Académique Documents
Professionnel Documents
Culture Documents
Table of Contents
Foreword 0
Part I Motor-CAD 13
1 What
...................................................................................................................................
is Motor-CAD? 13
2 Electromagnetic
...................................................................................................................................
Modelling of Electric Motors 14
3 Thermal
...................................................................................................................................
Modelling of Electric Motors 16
4 Motor-CAD
...................................................................................................................................
DEMO 19
5 BPM-EMag
................................................................................................................................... 20
6 BPM-Therm
................................................................................................................................... 21
7 IM-Therm
................................................................................................................................... 23
8 IM1PH-Therm
................................................................................................................................... 25
9 SRM-Therm
................................................................................................................................... 35
10 BPMOR-Therm
................................................................................................................................... 36
11 PMDC-Therm
................................................................................................................................... 54
12 SYNC-Therm
................................................................................................................................... 67
13 CLAW-Therm
................................................................................................................................... 83
14 Default
...................................................................................................................................
Motor Type 96
15 Contact
...................................................................................................................................
Information 97
16 F1...................................................................................................................................
Help 97
17 Licence
...................................................................................................................................
Information 98
14 Font
...................................................................................................................................
Size 124
15 About
...................................................................................................................................
Box 126
16 Lock
...................................................................................................................................
Results 127
17 Export
...................................................................................................................................
of Results in CSV Format 127
18 Geometry
...................................................................................................................................
Export 135
19 Geometry
...................................................................................................................................
Import 137
Rotor Duct
.........................................................................................................................................................
Geom etry 309
Rotor End.........................................................................................................................................................
Winding Geom etry 312
Shaft Geom
.........................................................................................................................................................
etry 313
Shaft Groove
.........................................................................................................................................................
Geom etry 314
Shaft Hole
.........................................................................................................................................................
Diam eter 314
Shaft Spoke
.........................................................................................................................................................
Geom etry 316
Slot Geom.........................................................................................................................................................
etry 317
SRM Geom .........................................................................................................................................................
etry 318
Stator Duct
.........................................................................................................................................................
Geom etry 319
SYNC Rotor
.........................................................................................................................................................
Geom etry 324
Tooth Tip.........................................................................................................................................................
Geom etry 326
Tooth Width
......................................................................................................................................................... 328
Wafter Geom
.........................................................................................................................................................
etry 328
WJ Spiral.........................................................................................................................................................
Channel Geom etry 330
Radial Duct
.........................................................................................................................................................
Geom etry 331
3 Cooling
...................................................................................................................................
Options 332
Cooling Options
.........................................................................................................................................................
[Input Data Editor] 332
Miscellaneous
.........................................................................................................................................................
Data [Cooling Options] 334
Fixed Tem.........................................................................................................................................................
peratures 337
Lam ination
.........................................................................................................................................................
Model 337
4 Thermal
...................................................................................................................................
model losses 339
Losses [Input
.........................................................................................................................................................
Data Editor] 340
Loss Variation
.........................................................................................................................................................
w ith Speed 341
Copper Loss
.........................................................................................................................................................
Variation w ith Tem perature 343
Loss Variation
.........................................................................................................................................................
w ith Tem perature & Load 344
Stall Analysis
......................................................................................................................................................... 349
Stall Copper
.........................................................................................................................................................
Loss Distribution 367
Stator Copper
.........................................................................................................................................................
Loss Distribution 368
Uneven Copper
.........................................................................................................................................................
Loss Distribution 369
Loss Distribution
......................................................................................................................................................... 375
Brushless.........................................................................................................................................................
AC m otor Loss Model 375
5 Materials
...................................................................................................................................
& Weight 380
Materials.........................................................................................................................................................
& Weights - electrom agnetic m odel 380
Materials.........................................................................................................................................................
& Weights - therm al m odel 381
Material Help
......................................................................................................................................................... 383
6 Interface
...................................................................................................................................
Gaps 386
Interface.........................................................................................................................................................
Gaps [Input Data Editor] 386
Therm al .........................................................................................................................................................
Contact Resistance 387
Interface.........................................................................................................................................................
Gaps 390
Bearing Model
......................................................................................................................................................... 395
7 Radiation
................................................................................................................................... 401
Radiation.........................................................................................................................................................
Data [Input Data Editor] 401
Internal Radiation
.........................................................................................................................................................
Heat Transfer 403
View Factor
......................................................................................................................................................... 410
Em issivity
......................................................................................................................................................... 411
8 Convection
...................................................................................................................................
Correlations 413
Convection
.........................................................................................................................................................
Heat Transfer 413
Mixed Convection
......................................................................................................................................................... 415
9 Natural
...................................................................................................................................
Convection 416
Natural Convection
.........................................................................................................................................................
Data [Input Data Editor] 416
Natural Convection
.........................................................................................................................................................
Correlations 417
Horizontal
.........................................................................................................................................................
Cylinder [Natural Convection Correlation] 418
Vertical Cylinder
.........................................................................................................................................................
[Natural Convection Correlation] 419
Vertical Flat
.........................................................................................................................................................
Plate [Natural Convection Correlation] 420
Horizontal.........................................................................................................................................................
Flat Plate [Natural Convection Correlation] 421
Horizontal.........................................................................................................................................................
Flat Plate - Upper & Low er [Natural Convection Correlation] 422
Horizontal.........................................................................................................................................................
Square Tube [Natural Convection Correlation] 423
Horizontal.........................................................................................................................................................
Servo Housing [Natural Convection Correlation] 423
Vertical Fin
.........................................................................................................................................................
Channel [Natural Convection Correlation] 424
Horizontal.........................................................................................................................................................
Fin Channel [Natural Convection Correlation] 425
Horiz Fins/Flat
.........................................................................................................................................................
Plate [Natural Convection Correlation] 426
Horiz Fins/Flat
.........................................................................................................................................................
Plate & Cylinder [Natural Convection Correlation] 427
Horiz Fins/Flat
.........................................................................................................................................................
Plate & Tube [Natural Convection Correlation] 428
Horiz & Vert
.........................................................................................................................................................
Fin Channel Average [Natural Convection Correlation] 429
Horiz/Vert.........................................................................................................................................................
Fins & Horiz Cylinder [Natural Convection Correlation] 429
10 Forced
...................................................................................................................................
Convection 430
Blow n Over
.........................................................................................................................................................
[Input Data Editor] 430
Blow n Over
.........................................................................................................................................................
Flow Options 431
Blow n Over
.........................................................................................................................................................
Fluid Flow 433
Blow n Over
.........................................................................................................................................................
Heat Transfer 435
Blockage.........................................................................................................................................................
and Leakage 437
Air Velocity
......................................................................................................................................................... 439
Air Flow ......................................................................................................................................................... 441
Convection.........................................................................................................................................................
Heat Transfer 443
Mixed Convection
......................................................................................................................................................... 445
Forced Convection
.........................................................................................................................................................
Correlations 445
Flat Plate.........................................................................................................................................................
Convection Correlation 446
Fin Channel
.........................................................................................................................................................
Convection Correlation 448
Enclosed.........................................................................................................................................................
Channel Convection Correlation 451
Across Fin.........................................................................................................................................................
Flow 455
11 Through
...................................................................................................................................
Ventilation 456
Through .........................................................................................................................................................
Ventilation Model 456
Rotor to Stator
.........................................................................................................................................................
Duct Flow 457
Through .........................................................................................................................................................
Ventilation [Input Data Editor] 457
Through .........................................................................................................................................................
Ventilation Flow Options [Input Data] 458
Through .........................................................................................................................................................
Ventilation Fan Characteristic [Input Data] 469
Through .........................................................................................................................................................
Ventilation Fluid Flow [Input Data] 471
Through .........................................................................................................................................................
Ventilation Heat Transfer [Input Data] 473
Inlet Definition
.........................................................................................................................................................
[TVent] 475
Rotation Entrance
.........................................................................................................................................................
Effects 477
Flow Model.........................................................................................................................................................
[Through Ventilation] 478
Flow Resistance
.........................................................................................................................................................
Minor Loss Factor 497
Through .........................................................................................................................................................
Ventilation & Wet Rotor Airgap Model 507
Woods Minor
.........................................................................................................................................................
Loss Factor Graphs 510
Endspace.........................................................................................................................................................
Rotational Effects 513
12 End
...................................................................................................................................
Space 516
End Space .........................................................................................................................................................
[Input Data Editor] 516
End Space .........................................................................................................................................................
Convection Correlation 519
Endcap Vents
......................................................................................................................................................... 525
End Space .........................................................................................................................................................
Altitude Effects 528
k1, k2, k3.........................................................................................................................................................
When Not Air 529
13 Wet
...................................................................................................................................
Rotor 531
Wet Rotor.........................................................................................................................................................
Model 531
Wet Rotor.........................................................................................................................................................
[Input Data Editor] 532
Wet Rotor.........................................................................................................................................................
Inlet/Outlet Position 537
Wet Rotor.........................................................................................................................................................
Convection Correlation 538
Wet Rotor.........................................................................................................................................................
Wedge/Divider Flow 541
Wet Rotor.........................................................................................................................................................
Active Fluid Flow only 543
14 Housing
...................................................................................................................................
Water Jacket 546
Housing Water
.........................................................................................................................................................
Jacket Model 546
Housing Water
.........................................................................................................................................................
Jacket Flow Options [Input Data Editor] 548
Housing Water
.........................................................................................................................................................
Jacket Fan Characteristic [Input Data] 549
Housing Water
.........................................................................................................................................................
Jacket Fluid Flow [Input Data Editor] 550
Housing Water
.........................................................................................................................................................
Jacket Heat Transfer [Input Data Editor] 552
Housing Water
.........................................................................................................................................................
Jacket Cooling Options 555
15 Shaft
...................................................................................................................................
Spiral Groove 557
Shaft Spiral
.........................................................................................................................................................
Groove Model 557
Shaft Spiral
.........................................................................................................................................................
Groove Fluid Flow [Input Data Editor] 558
Shaft Spiral
.........................................................................................................................................................
Groove Heat Transfer [Input Data Editor] 561
Shaft Spiral
.........................................................................................................................................................
Groove Cooling Options 564
16 Spray
...................................................................................................................................
Cooling 565
Spray Cooling
.........................................................................................................................................................
Model 565
Spray Cooling
.........................................................................................................................................................
Nozzle locations 566
Spray Cooling
.........................................................................................................................................................
Heat Transfer [Input Data] 569
Spray Cooling
.........................................................................................................................................................
Fluid Flow [Input Data] 571
Spray Cooling
.........................................................................................................................................................
Convection Correlation 575
Spray Cooling
.........................................................................................................................................................
Nozzle Positions 579
Spray Cooling
.........................................................................................................................................................
Model Setup 584
Spray Cooling
.........................................................................................................................................................
Calibration exam ple 602
Subm erged
.........................................................................................................................................................
double jet im pingem ent m ethod 615
17 Rotor
...................................................................................................................................
Water Jacket 628
Rotor Water
.........................................................................................................................................................
Jacket Model 628
Rotor Water
.........................................................................................................................................................
Jacket [Input Data] 632
18 Slot
...................................................................................................................................
Water Jacket 635
Slot Water
.........................................................................................................................................................
Jacket Model 635
Slot Water
.........................................................................................................................................................
Jacket [Input Data] 639
19 Submersible
................................................................................................................................... 644
20 Flooded
................................................................................................................................... 645
21 Transient
...................................................................................................................................
Data 646
Transient.........................................................................................................................................................
Model 646
Duty Cycle
.........................................................................................................................................................
Settings [Input Data Editor] 647
Duty Cycle
.........................................................................................................................................................
Definition [Input Data Editor] 651
Duty Cycle
.........................................................................................................................................................
Data File 654
Soak Back......................................................................................................................................................... 655
Transient.........................................................................................................................................................
Fault Analysis 657
Fault Tolerant
.........................................................................................................................................................
m ulti-phase open circuit 660
22 Settings
................................................................................................................................... 672
Settings .........................................................................................................................................................
[Input Data Editor] 672
Models [Settings]
......................................................................................................................................................... 673
Airgap Model.................................................................................................................................................. 674
Include Rt[Endcap]
..................................................................................................................................................
Axial & Radial 674
Include Fin ..................................................................................................................................................
Efficiency 676
Include Coil..................................................................................................................................................
Divider Node 677
Stator Tooth..................................................................................................................................................
Nodes 677
Axial Slice ..................................................................................................................................................
Model 680
Rotor Model .................................................................................................................................................. 687
1 Control
................................................................................................................................... 803
2 Back
...................................................................................................................................
EMF 806
3 Cogging
...................................................................................................................................
Torque 807
4 Loss
...................................................................................................................................
Calculation (Open Circuit) 808
5 Loss
...................................................................................................................................
Calculation (On Load) 809
6 Torque
................................................................................................................................... 810
7 Torque
...................................................................................................................................
Speed curve 811
8 Power
...................................................................................................................................
Speed curve 812
9 Electromagnetic
...................................................................................................................................
+ Thermal model 813
10 Custom
...................................................................................................................................
Drive Cycle 817
11 Stator
...................................................................................................................................
Skew 819
12 Iron
...................................................................................................................................
Loss calculation methods 820
13 Iron
...................................................................................................................................
Loss Coefficients 822
14 Magnet
...................................................................................................................................
and Sleeve Loss calculation methods 823
15 Torque
...................................................................................................................................
calculation methods 824
16 Harmonics
................................................................................................................................... 827
17 FEA
...................................................................................................................................
display 828
18 Total
...................................................................................................................................
Harmonic Distortion 829
19 Power
...................................................................................................................................
Factor 830
20 Custom
...................................................................................................................................
Geometry 830
11 Transient
...................................................................................................................................
Reduced Node Model 873
12 Reduced
...................................................................................................................................
Node Model - Best Practices 887
13 Transient
...................................................................................................................................
Results 888
14 Transient
...................................................................................................................................
Saved Temperature Data 890
15 Transient
...................................................................................................................................
Saved Temperature Data Options 891
16 Transient
...................................................................................................................................
Saved Temperature Graph 898
17 Transient
...................................................................................................................................
Speed Increase Data 899
18 Transient
...................................................................................................................................
Temperature Data 900
19 Transient
...................................................................................................................................
Temperature Graph 902
20 Transient
...................................................................................................................................
Temperature Graph Setup 903
21 Transient
...................................................................................................................................
Test Temperature Data 904
22 Transient
...................................................................................................................................
Test Temperature Data Selection 905
23 Transient
...................................................................................................................................
Test Temperature Graph 910
24 Zoom
...................................................................................................................................
[Graph] 910
11 Winding
...................................................................................................................................
[Electromagnetic Output Data] 978
12 Miscellaneous
...................................................................................................................................
[Output Data] 979
Miscellaneous
.........................................................................................................................................................
Output Param eters (Housing) 980
Miscellaneous
.........................................................................................................................................................
Output Param eters (Fluids) 981
Miscellaneous
.........................................................................................................................................................
Output Param eters (Airgap) 983
Miscellaneous
.........................................................................................................................................................
Output Param eters (Convergence) 985
Miscellaneous
.........................................................................................................................................................
Output Param eters (Other) 986
13 Housing
...................................................................................................................................
Water Jacket [Output Data] 988
Tem perature
.........................................................................................................................................................
[Housing WJ] 990
Flow [Housing
.........................................................................................................................................................
WJ] 992
Dissipation
.........................................................................................................................................................
[Housing WJ] 993
Area [Housing
.........................................................................................................................................................
WJ] 995
Therm al .........................................................................................................................................................
Resistance [Housing WJ] 996
14 Shaft
...................................................................................................................................
Spiral Groove [Output Data] 997
Tem perature
.........................................................................................................................................................
[Shaft Spiral Groove] 999
Flow [Shaft
.........................................................................................................................................................
Spiral Groove] 1001
Dissipation
.........................................................................................................................................................
[Shaft Spiral Groove] 1002
Area [Shaft
.........................................................................................................................................................
Spiral Groove] 1003
Therm al.........................................................................................................................................................
Resistance [Shaft Spiral Groove] 1004
15 Slot
...................................................................................................................................
Water Jacket [Output Data] 1005
Tem perature
.........................................................................................................................................................
[Slot Water Jacket] 1006
Flow [Slot
.........................................................................................................................................................
Water Jacket] 1007
Dissipation
.........................................................................................................................................................
[Slot Water Jacket] 1008
Area [Slot
.........................................................................................................................................................
Water Jacket] 1009
Therm al.........................................................................................................................................................
Resistance [Slot Water Jacket] 1010
16 Spray
...................................................................................................................................
Cooling [Output Data] 1011
17 Rotor
...................................................................................................................................
Water Jacket [Output Data] 1012
18 Wet
...................................................................................................................................
Rotor Output Parameters [Wet Rotor] 1013
Tem perature
.........................................................................................................................................................
[Wet Rotor] 1015
Flow [Wet
.........................................................................................................................................................
Rotor] 1017
Dissipation
.........................................................................................................................................................
[Wet Rotor] 1019
Area [Wet
.........................................................................................................................................................
Rotor] 1021
Therm al.........................................................................................................................................................
Resistance [Wet Rotor] 1023
19 Through
...................................................................................................................................
Ventilation 1024
Tem perature
.........................................................................................................................................................
[Through Ventilation] 1025
Flow [Through
.........................................................................................................................................................
Ventilation] 1026
Dissipation
.........................................................................................................................................................
[Through Ventilation] 1027
Area [Through
.........................................................................................................................................................
Ventilation] 1028
Therm al.........................................................................................................................................................
Resistance [Through Ventilation] 1029
20 Housing
...................................................................................................................................
Peripheries 1029
21 Blown
...................................................................................................................................
Over Flow Rates and Air Temperatures 1031
22 Stall
...................................................................................................................................
Parameters 1033
23 Material
...................................................................................................................................
[Electromagnetic Output Data] 1035
4 *.gsf
...................................................................................................................................
datafile 1042
5 Defaults
...................................................................................................................................
[Defaults.INI] 1043
6 History.INI
................................................................................................................................... 1044
Index 1116
1 Motor-CAD
1.1 What is Motor-CAD?
Motor-CAD is Computer Aided Design (CAD) package for the design of electric motors. The modules
currently supported by Motor-CAD are:
BPM-EMag
BPM-Therm
BPMOR-Therm
IM-Therm
SRM-Therm
PMDC-Therm
SYNC-Therm
CLAW-Therm
IM1PH-Therm
The particular module is selected using the Motor Type pull down menu. Each motor type uses the
same data-file structure and user interface, simplifying the transfer of data from one module to
another.
Motor-CAD has been developed to give the motor designer a fast method of analysing design
changes on the performance of the machine. In doing so not only can the optimum design solution
quickly be identified, but the user fully understands what design variables (geometry, materials, etc)
have the most influence on the machine performance and cooling. The latest user-interface
techniques are used so that the package is easy to use and the user is never more than a mouse
click away from the information required. Motor-CAD has been written so that the user need not be
an expert in electromagnetics or heat transfer - all the difficult parameters are calculated
automatically by Motor-CAD.
In this manual most of the details are valid for both the electromagnetic and thermal models in
Motor-CAD. However there are some cases where the interface or methods are only used for one of
the models and are highlighted in the heading sections as such.
Electromagnetic Model: See Electromagnetic Modelling of Electric Motors for more details on the
finite element and analytical methods used to calculate the
electromagnetic performance of the motor.
Thermal Model: See Thermal Modelling of Electric Motors for more details of the 3-
dimensional lumped circuit model (Schematic) used to calculate the
steady-state and transient thermal characteristics of the motor.
User Interface: Motor-CAD has an easy to navigate and intuitive main User Interface.
Online Help: On-line help is available at any point in Motor-CAD - just press the F1 key
.
Web Demonstration: A 5 minute Motor-CAD overview presentation can be downloaded from the
Motor-CAD website.
Contact Information: If you require further information please contact Motor Design Ltd.
Analytical Methods
Numerical Methods (Finite Element Analysis)(as used in Motor-CAD):
A brief description of the two methods together with a list of their main advantages/disadvantages is
given below:
Analytical Methods
The analytical modelling of the electric motors relies on the electromagnetic field and
generalized forces theorem. In a simplified form, the latter theorem states that the electromagnetic
torque (or force) is given by the system coenergy or energy variation with incremental rotor
displacement, if the currents and flux-linkages are constant during the elemental rotor movement. As
the system is an electrical motor, the energy or coenergy can be expressed as a function of
products between flux-linkages and currents. Furthermore, the flux-linkages are expressed as
products between reactances and currents. Consequently, all the developed analytical models that
compute the electromagnetic torques in AC motors are relying on the equivalent circuits parameters
(resistances and reactances) that can have fixed or variable (linear or non-linear) values. Thus, the
precision of any analytical model that estimates the electromagnetic torque in a rotating motor
depends on the accuracy level that characterizes the motor parameters.
Advantages:
o the analytical methods are based on measurable physical parameters and permit the
inclusion of non-linear effects. Therefore, many electrical machinery designers address
new prototypes development through the use of analytical tools.
Disadvantages:
o there are still physical phenomena that occur in rotating motors (e.g., stray load
losses) and cannot yet be mathematically modelled.
o several simplifying assumptions are necessary for any analytical motor model.
o the accuracy can be low unless significantly improved if the most important non-linear
effects (i.e., saturation, core loss, windage and friction loss, and harmonics) are
modeled through a sufficiently high number of elements in the mathematical model of
the motor.
Numerical modelling
The numerical modelling of electric motors has its basics in the electromagnetic fields
theory. There are several mathematical approaches such as finite-element method (FEM), finite-
difference method (FDM), boundary element method (BEM) to solving the system equations in
numerical methods. Regardless of the mathematics of these methods, the electromagnetic torque is
estimated using either the Maxwell stress theory, virtual work (energy variation) or Laplace method
(magnetizing currents).
Advantages:
o If the problem settings are correctly formulated, the numerical modelling of rotating
motors will usually lead to a higher accuracy level for the estimated results than the
analytical modelling
Disadvantages:
o 2D models still ignore the end-effects and the three-dimensional eddy currents effect.
o 3D models though potentially more accurate, require one or two orders of magnitude
more of computer resources;
o 3D models are still beyond the bounds of economic viability, especially in the electrical
machinery industry where tens of design versions for only one motor might be requested
in one day..
Thermal analysis techniques used in motor design usually fit into one of the following three
categories:
A brief description of the three methods together with a list of their main advantages/disadvantages
are given below:
Traditionally when designing a new motor, the thermal rating was estimated from prior knowledge of
one or more of the following:
thermal resistance of winding to ambient (C/W)
thermal resistances of winding to stator (C/W) and stator to ambient (C/W)
housing heat transfer coefficient (W/m2/C)
winding current density limit (A/mm2)
specific electric loading (A/m)
2-dimensional & 3-dimensional finite-difference and finite element analysis (FEA) can be used to
model the heat transfer due to conduction within a motor. However, 3-dimensional Computational
Fluid Dynamics (CFD) is required when dealing with convection heat-transfer.
be required.
reduced prototype requirement
gives the customer confidence in the companies design capabilities
There is a recent trend to develop lumped-circuit thermal models - with different levels of
sophistication. Motor-CAD uses a 3-dimensional lumped circuit model (example shown in
Schematic) that can be used to calculate the steady-state and transient thermal characteristics of
several motor types. The latest user-interface techniques are used so that the package is easy to
use and the user is never more than a mouse click away from the information required. Efficient &
robust mathematical routines are used in the network solver.
The disadvantages associated with lumped-circuit design methods that completely new features (e.
g. new form of cooling fin, liquid cooling down shaft, etc.) may be required to be built into the model
structure. However:
In many cases the new features may be modelled adequately using the existing version
of Motor-CAD with the correct assumptions. Expert backup is available from
support@motor-design.com for licenced Motor-CAD users (also from local
representatives).
Motor-CAD has been written with future developments in mind - both the graphical
interface (cross-section editor, schematic, etc.) and circuit model have been written in
such a way that new constructional features can be added with minimum effort - free to
licensed Motor-CAD users.
Motor-CAD has many adjustment factors available that can sometimes be used to
adequately model the new features.
Motor-CAD has a Circuit Editor to allow the user to make changes to the thermal
network, deleting existing components and adding new components.
convection and radiation are difficult to model accurately when using certain
constructional features. This is minimised when using Motor-CAD as it benefits from the
expert technical input provided by Motor Design Ltd and its university associates. Also,
Motor-CAD is under constant development and is used by companies worldwide so is
undergoing continuous validation.
Design Optimisation - Motor-CAD can be used as an integral part of the design process.
By optimising the thermal circuit in parallel with the electromagnetic and mechanical design
a true optimum design is achieved. Often the thermal design aspects are left until the end of
the design process, at which point it is too late to alter the design and a sub-standard motor
is produced.
certain key parameters that affects a motors thermal performance, e.g. interface gaps, rotor
losses. In many cases these parameters can be estimated by matching Motor-CAD output
with data that can be readily measured, e.g. component temperatures, stator losses. For
instance the interface gap between stator-lamination and housing can be varied until a
match is given with the measured values of T[stator] & T[housing].
Sensitivity Analysis and Robust Design - It is very easy to vary input parameters and
examine the effect on the temperature distribution. Sensitivity analysis can be used on
parameters such as dimensional tolerances, impregnation goodness, interface gaps,
material properties, etc and plot graph of the variation in critical temperature nodes. This can
be used to gain an insight into which are the critical design variables in the machine. It can
also be extended to form a thermal model in Robust Design techniques such a 6-Sigma.
The whole calculation process together with the variation in the relevant parameters can be
automated using ActiveX technology or by using the inbuilt Sensitivity Analysis tool.
When Motor-CAD is run it looks for a valid licence which is stored on a Security Key. If a valid
license is not found then the user can run Motor-CAD in demo mode.
In demo mode the user can input data for their motor design. But when it comes to calculate the
thermal performance (steady-state and transient) the user is presented with results for internal
default data rather than their design. The message shown below is displayed indicating that "Your
dongle is not licensed for this motor type":
The demo is meant to give the user an idea of how easy it is to use the software and what
calculations are made.
The demo mode is also available to view Motor & Application Modules that the user does not a valid
license for.
If a dongle is present and you expect to be able to run Motor-CAD then please check the licence
diagnostics to ensure that the required modules are available.
1.5 BPM-EMag
Module of Motor-CAD for the electromagnetic design of Brushless Permanent-Magnet Motors
(BPM).
The BPM electromagnetic model is selected by setting the Motor Type to BPM or BPMOR and the
Model Type to one of the E-Magnetic options.
1.6 BPM-Therm
Module of Motor-CAD for the thermal design of Brushless Permanent-Magnet Motors (BPM).
The BPM motor type is selected using the Motor Type pull down menu.
1.7 IM-Therm
The IM motor type is selected using the Motor Type pull down menu.
1.8 IM1PH-Therm
Module of Motor-CAD for the thermal design of Single Phase Induction Machines (IM1PH).
The IM1PH motor type is selected using the Motor Type pull down menu.
Introduction
The Single Phase Induction Machine module (IM1PH-Therm) was not one of the original motor types
in Motor-CAD v1. It was added in v4 together with this sectio of help to assist the user in using the
new module.
Motor Cross-Section
If Motor-CAD default data is to be used then this will show the radial cross-section shown below:
The default machine has 18 slots which is more suitable for a three phase machine. So we may
wish to make this 20 in this case - as shown below:
The second parameter in the left hand parameter editor column is the number of poles in the
machine [IM Pole Number]. This is important if the machine is to have some slots that have a
reduced slot depth. We can see that by default some of the slots are made shorter than others.
More details of how to set the slot depth reduction for individual slots is given in the section on the
Winding Editor.
Winding Editor:
The winding editor shown below is used to set the number and size of wire for both the main and
auxiliary windings in each slot:
The wire size for the main and auxiliary conductors can now be set individually:
The table at the bottom of the editor is used to set up the following:
Which slot will have temperatures calculated for and shown in the [Schematic], [Output Data] and
[Transient Graph]. The slot is selected using the check-box .
Which slot will be displayed in the [Winding Editor]. This will be the current row selected - which is
shown in yellow and has the symbol.
Number of main conductors in each slot
Number of auxiliary conductors in each slot
The depth reduction of each slot - compared to the standard [Slot Depth] which is set in the radial
The number of slots shown in the table is dependent upon the [Slots Number] and [IM Pole Number].
The number is set to make sure that there is symmetry in te machine and is equal to will be equal to
[Slots Number] / [IM Pole Number] / 2.
The final four columns in the table present the following information:
Slot Area [Area] - this is a function of the slot depth reduction for that slot
Slot Fill [Fill] - this is a function of the slot area and the number of main and auxiliary conductors in
that slot
Copper Loss [Pcu] - this is a function of the main and auxiliary copper loss and the number of main
and auxiliary conductors in that slot
Copper Loss / Slot Fill Ratio [Pcu/Fill] - this is to try to identify which is the hottest slot. The
concept is that a slot is likely to be hot if the copper loss in that slot is high and the slot fill is low.
Hence the slot with the highest Pcu/Fill ratio is likely to have the hottest temperature and maybe the
user should select this slot to be the one calculated.
If the user selects the combination of wire size and number of conductors in a slot that leads to an
invalid slot fill of greater than one then this condition is highlighted in red as shown below:
Both the losses in the Main and Auxiliary winding are input the editor shown below:
The main problem with modelling the single phase motor in Motor-CAD is that it assumes that all the
slots are the same, i.e. same size and same losses. This is clearly not true in the single phase
induction motor. To overcome this we apply the same type of analysis as is used in the Stall and
Fault analysis in Motor-CAD. The total loss in the slot is based on the slot chosen to be modelled
by the user, i.e. copper loss = copper loss in slot chosen multiplied by the number of slot. We then
apply the following rules to make sure that the dissipation in the machine compatible with the total
loss in the machine:
In the steady state calculation an additional loss is removed or added to the stator back iron node to
account for the loss in the chosen slot is greater or less than the addition of the main and auxiliary
copper losses.
In a transient calculation the power taken from the back iron is adjusted to account for the storage
term due to the thermal capacitance. In this case the amount of power taken from the node is equal
to the additional power multiplied by the instantaneous total power into the stator node plus the
power from the end windings divided by the total copper loss.
Details of the adjustments made to achieve a valid calculation are given in the [Output Data]:
1.9 SRM-Therm
Module of Motor-CAD for the thermal design of Switched Reluctance Motors (SRM).
The SRM motor type is selected using the Motor Type pull down menu.
1.10 BPMOR-Therm
Module of Motor-CAD for the thermal design of outer rotor brushless permanent magnet machines
(BPMOR).
The BPMOR motor type is selected using the Motor Type pull down menu.
Introduction:
The Outer Rotor Brushless Permanent Magnet module (BPMOR-Therm) was not one of the original
motor types in Motor-CAD v1. It was added in v3.1 together with this section of help to assist the
user in using the new module.
The BPMOR machine usually has a much more complex thermal model than the traditional
brushless permanent magnet inner rotor (BPM) motor. This is because:
The stator winding which is the main loss producing component is in the centre of the
machine.
The machine is often embedded in apparatus which it is driving, e.g. in the wheels of
a traction system. The apparatus needs to be taken into account when constructing
the thermal model as it can be a main cooling/heating component.
The blue housing is in fact the rotating magnet cover in this case. The user can also have a separate
rotating magnet return path component shown in light blue. The magnets are shown in green. The
static stator lamination is shown in red and is mounted on the dark green static shaft. The windings
are shown in yellow and the rotating endcaps in dark olive.
When the Motor-CAD user selects the BPMOR construction the Housing Type (the rotor housing in
this case) is set to the [Round] shape. It is unlikely that a finned construction that is typical in the
BPM motor housing will be used on this rotating surface.
The Mounting Type is assumed to be [Not Mounted] for the BPMOR. It is assumed that there is no
cowling.
The construction shown above has the Rotor Mounting set to [Open]. It is assumed that the outer
surface of the rotor housing has a mixture of natural and forced convection. The forced convection
fluid velocity is set equal to the peripheral speed of the rotor surface. Often the motor is built into
some other equipment such as a wheel. This can be modelled by setting the Rotor Mounting = [
Insulated]. In such cases an extra dark green housing cover is drawn as shown below:
The Cover Ins Thickness and Cover Ins Length are used to set its dimensions. The outer surface of
the cover is assumed to have a combination of natural and forced convection in a similar fashion to
that of the outer surface of the rotating housing in the open construction.
The layered Winding Model is used as with BPM. However the slot is now similar to that used in a
permanent magnet dc machine:
As mentioned previously the outer surface of the rotor housing or cover if used has a combination of
natural and forced convection. To model this Motor-CAD forces the Cooling Type to Blown Over:
Many of the usual selection options are not enabled as the Blown Over model is automatically set
up to force the surface fluid velocity to be equal to the circumferential velocity of the rotor housing
and/or housing cover. The Air Velocity Leakage parameter is also set up automatically to account
for the fact that the rotor housing overhang may not be covered and so will have a different radius and
thus velocity to the active section of the housing. The endcap axial sections will also have a different
velocity. The mixed convection is calculated to account for cases where the speed is low and natural
convection dominates.
Sometimes a water jacket is used on the static axle as shown below. See Shaft Spiral Groove
Model for details.
BPMOR Losses:
The losses in the machine are set using the editor below (Losses [Input Data Editor]):
BPMOR Materials:
The properties of the solid materials in the machine are set using the editor below (Materials &
Weights [Input Data Editor]). The [Housing Cover] material has been added to the list of
components:
The interface gaps between components are shown below (Interface Gaps [Input Data Editor]):
An important interface is that of the stator lamination to axle. As this is a laminated surface, then
the roughness can be greater than it would be with two solid components and a larger effective gap
can result. This can lead to a significant temperature rise across this boundary in a heavily loaded
machine.
The convection around the end-windings and internal surfaces in the endcaps is calculated using the
same method as in the BPM machine (End Space [Input Data Editor]). However some different
surfaces in the BPMOR are now rotating and some static compared to the BPM. This is taken
account of in the effective surface velocity calculation:
At present it is not possible to pot the end-windings in the BPMOR model. This feature will be added
at a later date.
BPMOR Schematic:
The static axle is shown at the top of the diagram and the rotating housing cover at the bottom of the
diagram. If parts of the rotor housing are covered then extra resistances are introduced into the
circuit as shown below. The axial outer surfaces of the endcaps are assumed to be rotating and can
be shown as separate resistances to the radial surfaces if they are covered:
If a water jacket is used then it is shown at the top of the diagram. The fluid temperature and
dissipation to the fluid is indicated on the diagram:
1.11 PMDC-Therm
Module of Motor-CAD for the thermal design of Permanent Magnet DC Motors (PMDC).
The PMDC motor type is selected using the Motor Type pull down menu.
Introduction:
The Permanent Magnet DC machine module (PMDC-Therm) was not one of the original motor types
in Motor-CAD v1. It was added in v2 together with this section of help to assist the user in using the
new module. In Motor-CAD v4.1 a new Brush Holder model was added.
PM DC Machine Selection:
PM DC Cross Section:
The dimensions are changed using the editors shown in Figure 1 and Figure 2. There are several
parameters specific to the PMDC motor geometry - their long names and the help on the status bar
should be sufficient to identify what dimension each parameter controls. For more details see:
Endcap Thickness [F]
Endcap Thickness [R]
Gap [Endcap-Comm]
Gap [Comm-Winding]
Comm Length
Comm Dia
Comm Cu Thickness
Brush Number
Brush Width
Brush Length
Brush Height
Figure 1: PMDC motor type selected and the Radial Cross Section Editor
The winding editor is shown in Figure 3. Parameters specific to the PMDC machine are Comm
Connections and Comm Connection Lmult. These are used to calculate the thermal conductive
length and cross sectional area of the leads connecting the end winding to the commutator - this
acts as a cooling path for the commutator. The other cooling paths for the commutator are via the
insulating part of the commutator to the shaft and convection from the commutator circuit. The
thermal resistances and loss associated with the commutator are shown in Figure 4. The full thermal
circuit is shown in Figure 5.
The Commutator brush friction and VI losses are set in Figure 7. Half the loss is put on the
commutator and half on the brushes.
The materials used to make the commutator (copper and insulation to shaft) and brushes are set in
Figure 8.
Interface gaps are shown in Figure 9. The gap between commutator and brushes is modelled as an
interface gap.
The end space editor accounts for the fact that some components are rotating and others are not -
see Figure 10 (i.e. the winding is rotating rather than being static). The only complex surface is that
of the end-winding bore. In most cases the end-space air will be shielded from this surface (the
winding bore will be close to the shaft). To account for this the [Air Velocity] multiplier is set to 0.1
as a default. Cooling from the commutator and brushes to the end space air is also considered in
the model.
Figure 11 shows a typical transient as calculated by the new PMDC-Therm module. Here we can
see that the brushes and commutator are the hottest component.
Figures 12 and 13 show the new Brush Holder = Bayonet model that was added to Motor-CAD v4.1.
The new dimensional parameters associated with the brush holder are:
Br Holder Comm Gap
Br Holder Outer Dia
Br Holder Ins Thick
Br Holder Height
1.12 SYNC-Therm
The SYNC motor type is selected using the Motor Type pull down menu.
Introduction:
The Synchronous Machine module (SYNC-Therm) was not one of the original motor types in Motor-
CAD v1. It was added in v2 together with this section of help to assist the user in using the new
module. Improved rotor winding modelling capabilities (layered winding model) and a new rotor type
more suited for aerospace generators than the existing industrial generator topology were added in
Motor-CAD v4.1
The radial cross-section is shown in Figure 1. The axial cross section is shown in Figure 2.
The dimensions used to draw the rotor are shown in Figure 3 and 4.
Stator Winding
The stator winding model used in Motor-CAD is the same as in the brushless permanent magnet
(BPM) and induction motors (IM). That is a layers model. This can have parallel teeth or parallel
slots. This model is also used for form wound windings, the user selecting an appropriate wire size
that matched the rectangular conductors used. In the future new parameters may be added to give a
better model for the form winding. The winding is defined in the Stator Winding Editor.
Rotor Winding
The rotor has a bobbin wound coil. The thickness of insulation between the winding and the rotor
pole base and sides are set independently using the cross section editors (Figures 1 and 2). The
liner material properties is set as in Figure 5. The losses in the rotor winding are set as shown in
Figure 6. The winding is defined in the Rotor Winding Editor.
The resistance/power source network that we are using to model the rotor winding and rotor poles is
shown in Figures 8 and 9 (without and with through ventilation). More details are given below.
These component winding resistances are output in the [Output Data] as shown in Figure 10.
In Motor-CAD v3.1(5) a layed winding model was implemented for the rotor winding. The new winding
model is selected using the [Input Data] - [Calculation Options] - Sync Rotor Winding radio button
selector as shown in Figure 8. The new winding method is default for new designs. The new winding
editor is shown in Figure 9.
The resistance/power source network that we are using to model the rotor winding and rotor poles is
shown for the simple winding model used prior to Motor-CAD v3.1(5) is shown in Figures 10 and 11
(without and with through ventilation). The rotor coil losses are modelled as a single power source
which is transferred to the pole iron and interpole air via the coil sides, base and top. The more
sophisticated layered winding model for the rotor displays the same circuit with the inner most
winding node temperature and the total heat transfer to the pole and surrounding air displayed to the
user. But in reality many more nodes are used to model the heat transfer through the rotor winding.
More details or local temperatures thoughout the rotor winding can be obtained using the Circuit
Editor (Figure 12) or the [Output Data] table of temperatures - Figure 13.
This simplified winding model should be calibrated to give good results while the more sophisticated
layed winding model should model the temperature distribution more accurately just using the
winding details.
Figure 12: Circuit Editor showing the layered winding model for the rotor
Results:
The calculated steady-state temperatures, resistances and power flow can be viewed in the
Schematic or Output Data [Numerical Values]. The transient performance can also be calculated
and viewed - Transient Graph.
The original rotor type geometry (Type 1) is described above. This rotor type is typically used for
industrial type machines. The new rotor Type 2 was implemented in Motor-CAD v3.1(5) to suit
geometries typically used in aerospace machines. The new geometry is shown in Figure 14:
The layered winding model associated with the Type 2 rotor is shown in Figure 15:
1.13 CLAW-Therm
Module of Motor-CAD for the thermal design of Claw Pole Machines (CLAW).
The CLAW motor type is selected using the Motor Type pull down menu.
Introduction:
The Claw Pole machine module (CLAW-Therm) was not one of the original motor types in Motor-
CAD v1. It was added in v2 together with this section of help to assist the user in using the new
module.
The Claw Pole machine is selected using the Motor Type selector.
The radial cross-section is shown in Figure 1. On the rotor we see the tips of the far poles drawn
between the poles of the near section. The axial cross section is shown in Figure 2. We see the
tapered poles and the yellow winding on the rotor. We can also see the green insulation material
between the coil and the rotor poles.
The dimensions used to draw the rotor are shown in Figure 3 to Figure 6. In Figure 5 and Figure 6 we
can see that the taper of the poles in the axial direction is set by the [Claw Pole Spacing] and [Claw
Tooth Width] parameters.
Figure 5: Claw Pole dimensions (rotor surface drawing - circumference made into a flat
section)
Figure 6: as Figure 5 but different value of Claw Pole Spacing and Claw Tooth Width
Stator Winding
The stator winding model used in Motor-CAD is the same as in the brushless permanent magnet
(BPM) and induction motors (IM). That is a layers model. This can have parallel teeth, parallel slots
or be slotless. This model is also used for form wound windings, the user setting the size of the
rectangular conductors used. In the future new parameters may be added to give a better model for
the form winding. The winding is defined in the Stator Winding Editor.
Rotor Winding
The rotor has a bobbin wound coil. The thickness of insulation between the winding and the claw
pole base and sides are set independently using the axial cross section editor (Figure 2). The liner
material properties is set as in Figure 7. The losses in the rotor winding are set as shown in Figure
8. At present the rotor winding model is not as sophisticated as that of the layered stator winding
model. The Rotor Winding model is defined in the Rotor Winding Editor.
The resistance/power source network that we are using to model the rotor winding and claw poles is
shown in Figure 10. In the present model we model the rotor coil losses as a single power source.
This then passes to the claw pole iron via the coil base and the coil sides (assumed that coil top is
insulated). The thermal path between the coil and rotor iron base and pole consists of three
resistances:
slot liner
liner-rotor interface gap (may be partially or completely filled in impregnation)
winding impregnation - resistance length taken as ½ coil depth multiplied by (1-
Slot Fill) for the coil base and ¼ coil axial length multiplied by (1-Slot Fill) for the
coil sides.
This simplified winding model should be calibrated to give good results. A more sophisticated layers
model for the rotor winding may be added in a later version (this will require less calibration as it
more closely matches reality).
These component winding resistances are output in the [Output Data] as shown in Figure 11.
Figure 10: Claw Pole rotor and winding thermal resistance and power source network
Results:
The calculated steady-state temperatures, resistances and power flow are shown in Figure 12 and
Figure 13. The transient performance can also be calculated.
Future Work:
Used to set the default Motor Type that is displayed when Motor-CAD is first executed. The
information is stored in Defaults.INI. Before a selection has been made the default motor type is set
to the first valid motor type licence that is found on the Security Key
When the Default Motor Type Pull Down Menu command is initiated it brings up the form shown
below:
1.16 F1 Help
On-line help is available at any point in Motor-CAD. Press the F1 key and the online help will appear
with the help page set to give information on the interface component active (being pointed to with
the mouse) at that time, i.e. edit-box, grid, graph, etc.
Motor-CAD requires a hardware licence key (dongle) fitted to either the parallel or USB ports in order
to work correctly. Motor-CAD is licensed to be used by the company named in the About Box. The
Program Module's licensed are also listed together with the type of license (Perpetual, Rental or
Evaluation) and the limit date.
Details of the valid licenses are given in the About Box or License Diagnostics
Support Dates:
For a Perpetual license the limit date is used to compare with the Program Release Date to
determine if the company has a support contract for that version of the software. If not then they
cannot run that version of the software. They can however run versions with a release date prior to
the support date. A user requires a new dongle update after paying support to set the support dates
on the dongle. The update can be sent by email and read into dongle memory using the Security
Key Update option in the Pull Down Menu
Motor-CAD makes a check to see if the Windows clock has been turned back in order to extend the
licence period. The dongle stores the date on which the last time the software was used. If a
computer date that pre-dates the last used date then the dongle will become non-operational and
must be sent back to the supplier to be re-programmed - the user must then apply for a new dongle
license.
Splash Screen:
If the dongle is found and the licence is valid then the following splash screen is displayed while
Motor-CAD is loaded into memory:
Once the splash screen disappears the radial-cross section editor is displayed.
A "Checking Motor-CAD Licence" windows dialogue box is displayed when stating Motor-CAD
indicating that it is checking the dongle. If the dongle is not found the following screen is displayed:
One option is to run Motor-CAD in Motor-CAD DEMO mode buy pressing the Motor-CAD Demo
button.
If there is a problem with the license data on the dongle the a screen similar to that shown below will
be shown. This will give information on the error encountered. In such cases you should contact
Motor Design Ltd or your local representative with information on the error reported.
Dongle Problems:
If the user is having problems with the dongle or License Information held on it then they can run the
License Diagnostics routine that should show up the problem. This can then be reported to support.
Support may then ask the user to use the Advanced Options for which a password will be supplied.
Dongle Update:
An update for the security key (dongle) is in the form of a .gen file. This can be loaded onto the
dongle using the Security Key Update option in the Pull Down Menu. Even in the user has a
Perpetual licence they will require a new dongle update after paying support to set the support dates
on the dongle.
Very occasionally a user will notice that the splash screen does not look like that shown above, but
like the one shown below. This is because Windows has been set up to use Large Fonts rather than
Small Fonts. Motor-CAD user interface is written to use Small Fonts. If Large Fonts are used, many
of the screens in Motor-CAD will be distorted and have slide-bars displayed. Motor-CAD is not
designed for use with large fonts, we recommend that Small Fonts are used. The following procedure
should be used to convert from Large Fonts to Small Fonts - Windows Start menu Settings
Control Panel Display Settings Advanced Small Fonts.
2 Tutorials
The following tutorials are available from the Motor-CAD website and show how the different features
of Motor-CAD are used. The files should also be available in your Motor-CAD installation folder.
DataLogging Tutorial
3 User Interface
3.1 User Interface
Geometry
Control
Winding Editor
Input Data Editor
E-Magnetics
Temperatures
Output Data
Transient Graph
Circuit Editor
Sensitivity
Scripting
File:
Export Results Export Steady-State, Transient or Parametric results to an Excel file (CSV
Format)
Export Matrices Export the model matrix values to files for use in system simulation or
circuit reduction.
Copy to Clipboard put the current Motor-CAD screenshot into the clipboard to allow to paste
into another application
Geometry Export
Geometry Import
Properties Gives details of the current loaded model file
Differences to previous Gives differences between the current loaded file and the file previously
file loaded. Useful for comparing the differences between 2 different models.
Notes View and change the Data File Notes
Exit Exit Motor-CAD
Model:
Allows the model type and model context to be selected.
Motor Type:
Allows the Motor type to be selected.
Options:
Messages Options for setting the warning and Error Message Display
Show Text Option for displaying text in the main tabsheet headings
Show Icons Option for displaying icons in the main tabsheet headings
Show Toolbar Option for displaying the toolbar
Show Motor Type Option for displaying the Motor Type
Show Title Option for displaying the model Title
Defaults:
Units... Set units used within the program (defaults or those to be used with current
datafile)
Display Size Set default size which Motor-CAD will be displayed when executed
Font Size Set the default font sizes to be used within Motor-CAD
Motor Type Used to set the Motor Type that will be displayed when Motor-CAD starts
File Locations Used to set the initial folder path for the [Open] and [Save As] commands
Register for ActiveX When enabled Motor-CAD automatically registers itself for ActiveX. Note: If
ActiveX registration fails then this can be removed by editing the defaults.ini
file.
Editors:
The following options are the same as pressing the Main Option Selector buttons in the user
interface to obtain the editors:
View:
Performs zoom functions with the displayed pictures.
Zoom Reset
Zoom Out
Zoom In
Results:
The following options are the same as pressing the Main Option Selector buttons in the user
interface to obtain the results:
Schematic
Node Temperatures
Output Data
Transient Graph
Lock Results
Cancel Steady state - aborts a steady state calculation
Copy results to comparison data
Tools:
Circuit Editor
Sensitivity
SPEED... Import from or Export to the SPEED software, compare thermal results
with SPEED.
Flux2D Export a thermal or electromagnetic model to Flux2D from Motor-CAD.
2D calibration model This modifies the Motor-CAD so that it is 2D for comparison with a 2D
FEA solution.
Reduced node model Provides control to allow the user to reduce the network to a simpler form
analysis for use in system simulations or control strategies
Licence:
Diagnostics Full licence diagnostics.
Update Update the licence information in the dongle.
Configuration Local or network licence settings.
Advanced Advanced licence options (requires password)
Print:
Print Print current screen
Help:
The Tool Bar is an alternative to the pull-down menu for selecting to import or export from SPEED.
The F1 key can also be pressed at any time - which will give help information regarding the current
section of the program being used.
BPM-EMag - electromagnetic model for inner and outer rotor brushless permanent magnet
motors
Note: due to the amount of model information present in the combined electromagnetic thermal
model the interface has been split so only the electromagnetic or thermal interface is shown at one
time.
The E-Magnetic(interface) + thermal and E-Magnetic + thermal(interface) models are identical the
difference is the interface that is displayed to the user. Swapping from one to the other will not have
any effect on the model but will allow different model data to be viewed.
Motor-CAD requires a hardware licence key (dongle) fitted to either the parallel or USB ports in order
to work correctly. Motor-CAD is licensed to be used by the company named in the About Box. The
Program Module's licensed are also listed together with the type of license (Perpetual, Rental or
Evaluation) and the limit date.
Details of the valid licenses are given in the About Box or License Diagnostics
Support Dates:
For a Perpetual license the limit date is used to compare with the Program Release Date to
determine if the company has a support contract for that version of the software. If not then they
cannot run that version of the software. They can however run versions with a release date prior to
the support date. A user requires a new dongle update after paying support to set the support dates
on the dongle. The update can be sent by email and read into dongle memory using the Security
Key Update option in the Pull Down Menu
Motor-CAD makes a check to see if the Windows clock has been turned back in order to extend the
licence period. The dongle stores the date on which the last time the software was used. If a
computer date that pre-dates the last used date then the dongle will become non-operational and
must be sent back to the supplier to be re-programmed - the user must then apply for a new dongle
license.
Splash Screen:
If the dongle is found and the licence is valid then the following splash screen is displayed while
Motor-CAD is loaded into memory:
Once the splash screen disappears the radial-cross section editor is displayed.
A "Checking Motor-CAD Licence" windows dialogue box is displayed when stating Motor-CAD
indicating that it is checking the dongle. If the dongle is not found the following screen is displayed:
One option is to run Motor-CAD in Motor-CAD DEMO mode buy pressing the Motor-CAD Demo
button.
If there is a problem with the license data on the dongle the a screen similar to that shown below will
be shown. This will give information on the error encountered. In such cases you should contact
Motor Design Ltd or your local representative with information on the error reported.
Dongle Problems:
If the user is having problems with the dongle or License Information held on it then they can run the
License Diagnostics routine that should show up the problem. This can then be reported to support.
Support may then ask the user to use the Advanced Options for which a password will be supplied.
Dongle Update:
An update for the security key (dongle) is in the form of a .gen file. This can be loaded onto the
dongle using the Security Key Update option in the Pull Down Menu. Even in the user has a
Perpetual licence they will require a new dongle update after paying support to set the support dates
on the dongle.
Very occasionally a user will notice that the splash screen does not look like that shown above, but
like the one shown below. This is because Windows has been set up to use Large Fonts rather than
Small Fonts. Motor-CAD user interface is written to use Small Fonts. If Large Fonts are used, many
of the screens in Motor-CAD will be distorted and have slide-bars displayed. Motor-CAD is not
designed for use with large fonts, we recommend that Small Fonts are used. The following procedure
should be used to convert from Large Fonts to Small Fonts - Windows Start menu Settings
Control Panel Display Settings Advanced Small Fonts.
Pull Down Menu for dealing with the dongle License Information. There are 4 sub-menues:
License Configuration
Update
Diagnostics
Advanced Options
If a network license is selected in License Configuration then only the licence configuration item will
be enabled:
Motor-CAD uses a hardware security key (dongle) to hold license information. The parallel port or
USB dongles are red in colour and should have a unique 6 digit number on them. If the user if having
problems with the dongle or License Information held on it then they can run the License Diagnostics
routine that should indicate what the problem is. This can then be reported to support.
The typical screen for a lease licence that will be displayed is shown below. The lease expiry date is
shown in the 3rd column. The licence is only valid if this date is later than the current date.
A typical screen for a perpetual licence is shown below. The support date is shown in the 3rd
column. The licence is only valid for this version of Motor-CAD being run if the support date is later
than the release date of this version of Motor-CAD. The 4th column shows the limit date for this
perpetual licence and is normally set at 2100.
Motor-CAD uses a hardware security key (dongle) to hold license information. The parallel port or
USB dongles are red in color and should have a unique 6 digit number on them.
There is an option in the Pull Down Menu to import a dongle update file (.gen file):
The user should then select the .gen used to update the dongle.
An error will be given if the dongle update does not work correctly.
Details of the current valid licensed modules is given in the About Box or License Diagnostics
Motor-CAD can use a local security key (dongle) connected to the users machine or a remote
dongle that is on a server on the network. This licence configuration option is used to set up the
selection of dongle location.
To use the network option the Motor-CAD License Server software needs to be installed on the
network. The network licence functionality is an additional option. Please contact support if you
would like a quotation for transfer to a network license.
If you have a local dongle on the users machine then the setting shown below should be selected:
If you are using a remote dongle via the network then the user must specify the licence server name
or IP address as shown below.
Two servers can be specified. The first server is normally used however if this server does not
respond then the second (backup) server will be used instead.
When the [Query Server] button is clicked Motor-CAD will attempt to connect to the licence server.
The normal response is as shown below with the 'Successfully connected to server" message:
The user must select the licenses that they wish to use by selecting the checkboxes as shown
below. Motor-CAD will then request the selected licenses from the licence server.
Pressing the [Apply] button will show in green the modules which have been requested and
available. The licenses that are not available or are not requested are shown in red. For example
below the user has selected the BPM-EMag, IM-EMag, BPM-Therm, IM-Therm, ActiveX, Sensitivity,
CircuitEditor and FE-Therm modules and all are available except for the IM-Mag module that is
indicated in red.
Clicking the [Swap Servers] button will change the licence server being used to the backup one. This
allows easy verification that both licence servers are running. In normal operation if the first licence
server is not available then Motor-CAD will automatically attempt to access the licenses from the
backup server.
Motor-CAD uses a hardware security key (dongle) to hold license information. The parallel port or
USB dongles are red in colour and should have a unique 6 digit number on them. If the user if having
problems with the dongle or License Information held on it then they can run the License Diagnostics
routine that should show up the problem. This can then be reported to support. Support may then
ask the user to use the Advanced Options for which a password will be supplied.
- Printer setup
- SPEED Import/Export
- Write all output parameters to a text file (Steady.dat) - note this can take some time
- Exit Motor-CAD
The toolbar can be hidden and shown as required using the menu option shown below:
The F1 key can also be pressed at any time - which will give help information regarding the current
section of the program being used.
3.12 Title
An edit box on the top of the main form so the user can add useful information to be stored with the
datafile. Larger notes can also be stored in the Data File Notes.
The title can be hidden and shown as required using the menu option shown below:
The status line at the bottom of the screen gives the following information:
a short description of the current parameter/component
the position of the cursor in the radial and axial cross-section editors.
length measurement values
the Units of Length used
todays date
Motor Design Ltd's web site address
Used to set the default font sizes for the main Motor-CAD user interface and also set specific fonts
in the schematic, node temperature, circuit editor and flow circuit displays.
The Font Sizes are set using the Pull Down Menu. When initiated it brings up the form shown below:
The main font size should normally be set to 8 for the best viewing. If the Windows operating system
is configured with large fonts then it may be necessary to reduce this setting to a value of 7 as
otherwise the text may be too large. Motor-CAD attempts to reduce the main font size if it detects
that large fonts are being used in Windows.
The font sizes can also be set for the following screens:
Schematic
Circuit Editor
Axial and Radial Node Temperatures
Flow Circuit
Program Name
Licence Information, i.e. Company Name, Support/Renew Date & Key Number
Contact Information
Program Version
Program Release Date
The Program Release Date is used to determine if the user is licensed to use the current version of
the software.The dongle contains information relating to the customers support contract dates.
When a user extends their support contract a dongle update will be sent by email to update their
dongle date information.
There are however times when the user may wish to be able to view the model without the
calculations being re-run. The Lock Results option, shown below, allows the user to keep the
temperature results and avoid any recalculation while they view the model.
When the results are locked the locking is highlighted in the main application window as shown
below.
Values from the different results tables generated by Motor-CAD can be exported in the csv file
format that is suitable to be read by other software packages (e.g Excel). This allows further user
analysis and graphing of the results.
The results are exported from the file menu as shown below.
The Steady State Solution, Transient Solution or Parametric Solution results can be exported. The
Transient Solution results are only available for export once a Transient Solution has been
generated. The Parametric Solution results are only available for export once a Sensitivity Analysis
has been run.
For the Steady State Solution the export tables can be selected as shown below:
The user must define the folder and file into which the data will be exported:
For the Transient Solution the export tables can be selected as shown below (the transient must be
calculated first):
As with the steady state export the user must define a folder and file to export the data to. The data
can then be opening applications such as Excel:
If the user has Loaded previous transient results for comparison then there are also the options to
export the saved temperature data and the temperature difference data as shown below:
If the user has Loaded test temperature data for comparison with the current model then there is
also the option to export the External Test Temperature Data as shown below:
For the Parametric Solution to be exported the Sensitivity analysis must have been run. The csv file
generated can be opened in Excel as shown below:
The user is then presented with the interface below from which they select the export options
required:
The dxf generated is automatically displayed on the view for which it has been generated to allow the
user to check it as shown below:
If a DXF format file has been generated then this can be viewed in Motor-CAD using the Geometry
Import functionality.
The Motor-CAD model geometry can also be exported in 3D format for use with other CAD, FE or
CFD design packages. The STL format describes the 3 dimensional surface geometries. A file is
produced for each component. The components exported are those displayed in the 3D view.
The Motor-CAD geometry import functionality allows a DXF file to be imported into Motor-CAD. This
is useful for verifying that the Motor-CAD geometry matches the machine CAD drawings.
The DXF import can also be used as the custom geometry for an electromagnetic model.
There are options to select the view on which to display the dxf and also scaling, rotation and
positioning options as shown below:
A DXF of a motor design can then be superimposed on top of the Motor-CAD drawing as a visual
check as can be seen below:
4 Geometry
The geometry tab holds the radial and axial cross section editors and the 3D view.
Visual editor used to edit the motors radial configuration and dimensions.
Configuration:
The user can select the following motor configuration parameters from the options provided in the
form of a drop-down list:
Housing (i.e. generic fin/frame type)
Mounting (i.e. type of mounting used, flange/foot)
Slot Type (i.e. parallel slots, parallel teeth or slotless)
Stator Ducts (i.e. rectangular, circular)
Rotor Ducts (i.e. type of rotor ducting used with through ventilation cooling if used)
Dimensions:
The radial cross-section dimensions can be edited and the motor drawing updated using the redraw
button.
Redraw:
Motor Drawing:
An image of the motor defined by the dimensional parameters is drawn to provide the user with visual
feedback - making input errors less likely.
Check boxes used to enable/disable the drawing of the machine cooling flow paths, Flange Mounted
Plate and Foot Mounted Base if used (see Mounting).
Status Line:
Units:
The units of length (mm or inches) are indicated on the status line.
Editor for modifying the following input parameters that have influence on the motors radial cross-
section:
See Radial Geometry, Airgap Geometry, Sleeve Thickness, for details of the other individual
parameters.
SRM Data:
Stator Poles Stator Pole Angle Stator Pole Depth Stator Pole Radius
Rotor Poles Rotor Pole Angle Rotor Pole Depth
BPMOR Data:
Cover Ins Thickness
PMDC Data:
Comm Dia Comm Cu Thickness
Brush Number Brush Width Brush Height
Br Holder Comm Gap Br Holder Outer Dia Br Holder Ins Thick Br Holder Height
Cowling Data:
Cowling Thickness Cowling Radial Gap
If have Stator Ducts between housing and stator lamination (Stator Ducts = Rect Lam-H Divider, Circ
Lam-H Divider) :
Number Stator Ducts Stator Duct Ratio Stator Duct Gap
If have Circular Stator ducts or Rectangular Stator ducts (Stator Ducts = Circular Ducts or Stator
Ducts = Rectangular Ducts) (See Stator Duct Geometry):
If have Rotor ducts (Rotor Ducts = Circular Ducts or Shaft Spoke) (see Rotor Duct Geometry):
Circular RDuct Layers
L1 Circ RDuct Rad Dia L1 Circ RDuct Channels L1 Circ RDuct Dia L1 Circ RDuct Angle
L2 Circ RDuct Rad Dia L2 Circ RDuct Channels L2 Circ RDuct Dia L2 Circ RDuct Angle
L3 Circ RDuct Rad Dia L3 Circ RDuct Channels L3 Circ RDuct Dia L3 Circ RDuct Angle
L4 Circ RDuct Rad Dia L4 Circ RDuct Channels L4 Circ RDuct Dia L4 Circ RDuct Angle
If have Radial stator and/or rotor ducts (Stator Ducts = Radial Ducts or Rotor Ducts = Radial Ducts)
(See Radial Duct Geometry).
The status line gives a brief description of the parameter currently being pointed by the cursor.
The parameters should be edited and the Redraw button pressed to update the motor cross-section
drawing.
Visual editor used to edit the motors axial configuration and dimensions.
Configuration:
The user can select the following motor configuration parameters from the options provided in the
form of a drop-down list:
Housing Type (i.e. generic fin/frame type)
Mounting Type (i.e. type of mounting used, flange/foot)
EWdg Cavity (i.e. is the end winding potted or not)
Feedback Type (i.e. is an encoder fitted)
Cowling (i.e. type of fan cowling used)
Shaft Type (whether solid or spider shaft)
Fan (The fan is currently not used in the thermal model directly but allows for visualisation and
export to CFD for modelling of characteristics. In the future it is intended to have a fan model in
Motor-CAD)
Dimensions:
The radial dimensions and axial dimensions relevant to the axial cross-section can be edited and the
motor drawing updated using the redraw button.
Redraw:
Motor Drawing:
An image of the motor defined by the dimensional parameters is drawn in order to provide the user
with visual feedback - making input errors less likely.
Check boxes used to enable/disable the drawing of the machine cooling flow paths, Flange Mounted
Plate and Foot Mounted Base if used (see Mounting).
Status Line:
Units:
The units of length (mm or inches) are indicated on the status line.
Editor for modifying the following input parameters that have influence on the motors axial cross-
section:
If have encoder:
Enc Shaft Dia Enc Case Dia
Enc Case Thick Enc Radial Gap
If have axial housing channels (Housing = Axial Fins (Cv) or Water Jacket (Axial) ):
Fin Base Thickness Fin Cover Thickness
If have shaft spiral groove channels (Liquid Cooling = Shaft Spiral Groove) :
Shaft Groove Wall Shaft Groove Height
BPM Data:
Magnet Thickness
IM Data:
EndRing Add [Outer F] EndRing Add [Outer R]
EndRing Add [Inner F] EndRing Add[Inner R]
SR Data:
Stator Pole Depth Rotor Pole Depth
PMDC Data:
Comm Dia Comm Cu Thickness
Brush Number Brush Height
BPMOR Data:
The status line gives a brief description of the parameter currently being pointed by the cursor. The
parameters should be edited and the Redraw button pressed to update the motor cross-section
drawing.
Editor for modifying the following input parameters that have influence on the motors axial cross-
section:
If have encoder:
If have shaft spiral groove channels (Liquid Cooling = Shaft Spiral Groove) :
Shaft Groove Width Shaft Groove Spacing
PMDC Data:
Gap [Endcap-Comm] Gap [Comm-Winding] Comm Length
Brush Width Brush Length
BPMOR Data:
Cover Ins Length
The status line gives a brief description of the parameter currently being pointed by the cursor. The
parameters should be edited and the Redraw button pressed to update the motor cross-section
drawing.
4.3 3D Viewing
The tree structure allows the user to add and remove from view and change the transparency of the
components.
The view can be rotated by holding down the left mouse button and moving the mouse.
The view can be panned by holding down the right mouse button and moving the mouse.
The view can be zoomed by using the mouse scroll button.
The animation option allows the machine to be rotated and built for viewing.
The 3D view can be exported for use in other 3D design packages, see Geometry Export for details.
Please note: Not all geometry features are currently supported. Please contact support@motor-
design.com if you have any geometry features that you require that are not drawn.
5 Control
This is the main page for defining the machine control and provides the following settings:
Current Definition - define the current in terms of peak current, RMS current or RMS current
density in the slot.
Phase advance (in electrical degrees) of the current to the back EMF waveform.
DC Bus voltage - the voltage of the main DC bus. Motor-CAD will check to ensure have sufficient
supply voltage for the specified speed and current.
Drive Mode - the machine can be driven with sinusoidal, square wave or custom current waveforms
Stator Skew - stator skew is taken into account when calculating the torque waveforms and mean
lenght per turn (MLT) of the windings.
The temperatures options below are not editable when the E-Magnetic + Thermal model is enabled.
In this case the values shown are calculated by Motor-CAD thermal model from the loss values
calculated in the electromagnetic model.
Winding Temperature - this temperature is used to calculate the winding electrical resistivity that is
used in the calculation of the copper losses
Magnet Temperature - this temperature is used to calculate the effective magnet remanence
Stator lamination Temperature - this temperature is used for calculating the electrical resistivity of
the stator laminations that is used in the stator eddy current loss calculation.
Rotor lamination Temperature - this temperature is used for calculating the electrical resistivity of the
rotor laminations that is used in the rotor eddy current loss calculation.
Stator sleeve Temperature - this temperature is used for calculating the electrical resistivity of the
stator sleeve that is used in the sleeve loss calculation.
Rotor banding Temperature - this temperature is used for calculating the electrical resistivity of the
rotor banding that is used in the banding loss calculation.
Performance Tests
The tests to be performed are selected using the checkboxes shown below.
Open Circuit:
Back EMF - the machine is rotated through 360 electrical degrees with no current to measure the
voltage induced in the windings. Note: this uses a multi-static FE solution
Cogging Torque - the machine is rotated through a slot pitch with no current and dense airgap mesh
to measure the cogging torque of the machine. Note: this uses a multi-static FE solution
Loss Calculation (open circuit) - the machine is rotated through 360 electrical degrees with no
current to measure the losses induced in the machine. Note: this uses a transient FE solver.
On Load:
Torque - the machine is rotated through 360 electrical degrees with a drive current to calculate the
torque waveforms of the machine. Note: this uses a multi-static FE solution so does not calculate
magnet or iron losses.
Loss Calculation (on load) - the machine is rotated through 360 electrical degrees with drive currents
to calculate the losses in magnet, stator and rotor iron (also banding and sleeve if present). Torque
waveforms of the machine are also calculated. Note: this uses a transient FE solution.
Torque Speed Curve - the machine performance for different angle of phase advance is calculated to
give torque speed and power speed envelopes.
The model is automatically solved when the data has been changed by viewing the results sheets.
The model solved can also be started by clicking on the solve button shown below.
When the electromagnetic + thermal model is has been selected then the Solve E-Magnetic and
Thermal Model button will also be enabled. When this is clicked then the combined
electromagnetic and thermal model is solved.
During the solving the status bar at the bottom of the Motor-CAD window will indicate the progress of
the calculation. The calculation can be stopped at any time by clicking on the Cancel Solving
button.
6 Winding Editor
6.1 Winding Pattern
The winding pattern page is shown for electromagnetic models only. For thermal models it is
sufficient to know the number and size of conductors in the slot.
Winding Type
Automatic – creates a correct winding if the above conditions are respected. Winding
feasiblity rules are applied for the automatic winding type to ensure that the winding is valid.
Custom – allows user to implement any winding configuration, without checking its validity or
balanced conditions. Coils can be added and removed using the Add Coil and Remove Coil
buttons. The coil details should then be entered in the table.
Pattern (single phase) - Displays the coils locations of the selected phase. It is used in the
finite-element solver for transferring ampere-conductors to the correct slots
Pattern (All Phases) - Displays the coils of all phases. The coil-list table is replaced by a table
showing, for each slot, the number of conductors from each phase, together with the total
number of conductors. This table is useful for checking that all slots contain the correct
number of conductors.
Phasors - Displays a vector diagram of individual coil MMFs for each phase. It is used to check
that a winding is balanced, i.e. for 3-phase windings there magnetic phase axes are shifted
with 120 electrical degrees.
Harmonics - Displays the winding MMF harmonics. Fourier analysis of the winding ampere-
turns distribution in the stator slots when the current is assumed to be 1A.
Motor-CAD reports the mechanical winding MMF harmonics, i.e. the fundamental electrical
will correspond to the (Poles/2)th mechanical harmonic.
Factors - a winding factor is the ratio between the magnetic flux linking the phase coils of a
winding and the magnetic flux corresponding to a fully pitched, single-layer, non-skewed
stator winding.
General rules:
Slots / Phases /t = Integer (t = Highest Common Factor between Poles/2 and Slots)
Number of slot phasors: N_phasors = Slots / t (t = Highest Common Factor between Poles/2
and Slots)
For the Non-Overlapping winding type there are several different coil divider placement options.
With the solid divider winding type there is currently assumed to be no conduction through coil
divider.
With the air divider winding type their is cooling from the inner surfaces of the slot. This is modeled
in the End Space model where a relative air velocity of 0.1 is used to calculate the heat transfer
coefficients. The air divider types can be used with the wet rotor cooling to pass a cooling fluid down
the coil divider.
Method of wire diameter input, i.e. from wire tables or input diameters directly.
Pull down list of wire sizes available in currently selected wire table. The wire tables available at
present are:
AWG Table
SWG Table
Metric Table
The wire sizes given in Motor-CAD AWG, SWG and Metric tables are with heavy build insulation. If
other wire sizes are required then they can be entered directly using the Diameter Input option.
When Wire Type = AWG Table then the user can select the size as shown below:
The AWG wire gauge can be selected from the drop-down list. This sets the values of Wire Diameter
and Copper Diameter.
AWG wire gauges from 1 to 56 are available (mainly in 0.5 gauge steps).
When Wire Type = Metric Table then the user can select the size as shown below:
The metric wire size can be selected from the drop-down list. This sets the values of Wire Diameter
and Copper Diameter.
Metric wire diameters from 0.159mm (copper diameter = 0.125mm) to 2.12mm (copper diameter =
2.0mm) are available.
When Wire Type = Diameter Input then the values of Wire Diameter and Copper Diameter are
input directly as shown below:
When Wire Type = Rectangular Conductors the user can select the size as shown below:
The values of Copper Height, Copper Width and Insulation thickness for a rectangular conductor are
input directly.
Wedge Model is a pull-down list selector to set whether the slot area associated with the slot wedge
is thermally non conductive or used for the winding:
It can be advantageous in motors with a small slot number to extend the winding into the wedge
area. Both the wire area and the slot periphery available for dissipation of the copper losses are
increased. The slot area available for the winding is also increased.
When Wedge Model = Non Conductive that the wedge is assumed to be perfectly non conducting
even if its thickness is very small. An alternative model where the slot wedge has a finite thermal
conductivity may be added later.
When Wedge Model = Air then fluid can be passed down the wedge in the Wet Rotor cooling
method. This can be useful for cooling the conductors.
Note that the Wedge Model = Wound Space has only been implemented for an outer slot with a
parallel tooth at this time. Also, the Copper Depth [%] parameter is not implemented for such a
model.
In a electromagnetic model The winding can be defined by user input of the Copper Slot Fill or the
Wire Size. The selection is made from the Wdg Definition pull-down list shown below:
If it is an existing design then the number of conductors per slot and the wire size will be known and
it is better to input them directly. If you are looking changing the number of turns in the winding
pattern to optimise the performance then the copper slot fill will automatically adjust the wire size to
give the copper slot fill value specified.
In a thermal model The winding can be defined by user input of the Wire Slot Fill or the Conductors/
Slot. The selection is made from the Wdg Definition pull-down list shown below:
If it is an existing design then the number of conductors per slot and the wire size will be known and
it is better to input them directly. If you are looking at the effect of slot fill on temperature rise then
the wire slot fill can be input directly.
Calculated slot fill - based on round covered conductors and slot area available for winding after liner
insertion.
Required slot fill - based on round covered conductors and slot area available for winding after liner
insertion.
If Slot Fill is an input - the value of Conductors/Slot is an output and the edit box disabled as shown
below:
The value of Slot Fill may differ from Slot Fill (Wdg Area) as it is calculated using the copper layers
rather than the round conductors.
If Conductors/Slot is an input - the value of Slot Fill is an output and the edit box disabled as shown
below:
The value of Slot Fill may differ from Slot Fill (Wdg Area) as it is calculated using the copper layers
rather than the round conductors.
Impreg Layer1 Multiplier The gap between copper layers is referred to as the Impreg Thickness. It
is varied in order to achieve the required Slot Fill or Conductors/Slot given
the selected Wire Gauge. The gap between the outer copper layer and
slot liner (Impreg Layer1 Multiplier) can be made to be larger or smaller
than Impreg Thickness - this is achieved using the multiplier Impreg
Layer1 Multiplier.
Copper Diameter: Copper uncovered diameter (no insulation). The value may be input
directly or taken from the wire table pull-down list - see Wire Type.
Wire Diameter: Wire covered diameter, i.e. copper + insulation. The value may be input
directly or taken from the wire table pull-down list - see Wire Type.
Wire Insulation The insulation of the copper wire. (thin regions shown in purple in diagram
below)
Mat [Liner-Lam] is a pull-down list selector to set whether the Liner - Lam Gap is filled with
impregnation material or air.
The Impregnation Goodness parameters are used to define the thermal conductivity and weight of
the winding impregnation regions:
Imp Goodness [Liner-Lam] Multiplier used to enhance/degrade the Liner - Lam Gap thermal
conductivity. It is used to account for air pockets within the
impregnation.
Gap Slot Liner - Slot Wall Distance between the slot liner and the slot wall.
Gap Base Spacer - Base Distance between the base spacer and the bottom of the slot.
Gap Top Spacer - Wedge Distance between the top spacer and the slot wedge.
Base Spacer Thickness Thickness of the spacer at the bottom of the slot.
Phase Spacer Thickness Thickness of the phase spacer.
Top Spacer Thickness Thickness of the spacer at the top of the slot.
Coil Height Height of each coil per phase including the coil insulation.
Coil Width Width of each coil per phase including the coil insulation.
Coil Insulation Thickness Thickness of the coil insulation.
Bar Insulation Thickness Thickness of the copper bar insulation.
Bar Height Height of a copper bar (without insulation)
Bar Width Width of a copper bar (without insulation)
Bar rows Number of rows of conductors per phase in each slot.
Bar columns Number of columns of conductor in each slot
Percentage of windable slot depth full of copper, i.e. the copper can be forced towards the slot
bottom by inserting a larger slot wedge than usual. This can reduce the thermal resistance by
reducing the gap between conductors and any gaps between liner and lamination.
Clearance between coils in same slot of Non-Overlapping windings (see Winding Type). The
parameter is disabled when Winding Type = Overlapping.
These parameter defines the number of connections between the winding and commutator. This can
influence the heat transfer in the vicinity of the commutator. [Comm Connections] and [Comm
Connection Lmult] are used to calculate the thermal conductive length and cross sectional area of
the leads connecting the end winding to the commutator - this acts as a cooling path for the
commutator. The other cooling paths for the commutator are via the insulating part of the
commutator to the shaft and convection from the commutator circuit. The thermal resistances and
loss associated with the commutator are shown below:
In the very first version of Motor-CAD the end-windings were assumed to be the same temperature
as the outer layer of the active section of the winding, i.e. the thermal resistance between the two
was assumed to be zero based on the fact that the thermal conductivity of copper is much larger
than that of the insulation in the slots. This is however an over-simplification and there can be a
temperature difference between the end-winding and active section of the winding - especially if the
end windings are long and/or there is a significant amount of heat transfer between the two (either
the end winding cooling is very good resulting in the end winding having a higher temperature than
the active section or vice-versa). To allow such a temperature difference between the active and end-
windings the thermal resistances shown below were added to the circuit:
The only problem is to set realistic values for the resistances. The following formulation is used:
The [Winding Rt[Active-EWdg] Length Multiplier] parameter can be used to calibrate the model. A
value of 1 means that the length used in the resistance calculation is the full length of the end
winding from where it exits the slots to the furthest point from the slots. Also, if [Winding_Rt[Active-
Ewdg]_Length_Multiplier] is set to zero then it reverts to the previous model with no resistance
between active and end windings.
The copper loss is divided between the active and end windings according to the winding volume in
the two regions (for each end-winding):
Pactive = Pcu x (2 x L_Stator)/(2 x L_Stator + EWdg_MLT)
Pend = Pcu x (EWdg_MLT)/(2 x L_Stator + EWdg_MLT)/2
The diagrams below show the user the distribution of losses in the active and end windings:
The axial slice model allows the user to select the number of radial slices in the active regions of the
machine. In the future it is also hoped to add slices to the end winding regions of the machine as
depicted below:
This allows much higher resolution of temperatures throughout the winding. The other good thing is
that it is much easier to estimate an accurate value for a thermal resistance effective area and length
if there are more nodes (less physical distance between nodes). The only downsides are that the
calculation speed will be reduced and it will be harder to visualise the effective heat transfer paths. In
effect the more the number of nodes used in the model the closer the solution becomes to finite
element analysis (with both its advantages and disadvantages).
The user can place an extra thickness of insulation at the slot base on top of that of the Liner
Thickness. This will result in added temperature rise of the winding. At present the added thickness
of insulation is not drawn in the Winding Editor. A separate thickness can be set for the Ins [Tooth
Side] Thickness. The material properties are set in the Materials & Weights [Input Data Editor]:
The user can place an extra thickness of insulation at the tooth side on top of that of the Liner
Thickness. This will result in added temperature rise of the winding. At present the added thickness
of insulation is not drawn in the Winding Editor. A separate thickness can be set for the Ins [Slot
Base] Thickness. The material properties are set in the Materials & Weights [Input Data Editor]:
If it is an existing design then the mean length per turn (MLT) will be known and it is easy to
calculate the EWdg MLT from MLT - 2 x Stator Lam Length. Given this information and the wire size
it is possible to calculate the total amount of copper in the end windings. This will be a proportion of
the space available for the end windings (doughnut volume as indicated in the diagram below). The
ratio of copper volume to available volume is the EWdg Fill - this can be input if the user does not
have details of the MLT.
Mean length per turn in the end-winding. EWdg MLT is an input parameter.
The value of EWdg Fill is calculated from the EWdg MLT, the wire size and the end-winding volume
using the following equation:
Copper fill factor for front & rear end-winding, i.e. total wire volume/end-winding volume. EWdg Fill is
an input parameter.
The end-winding volume is calculated from the area depicted in the axial cross-section editor
drawing.
For both of the above end-winding definitions the EWdg Fill is then used to calculate the volume/
weight of copper and impregnation used in the end-winding.
The spacing between layers in the end-winding is assumed to be the same as in the active section
of the winding. Thus the increase in heat transfer between layers due to the end winding being
present will be proportional to the increase in copper weight due to the end-windings. The total (end
winding + active) copper weight and active copper weight are used in the calculation of EWdg Layer
Cond Mult. This parameter is used to calculate the proportional increase in winding thermal
conductivity between layers due to the end winding.
The end windings can be surrounded by air or potted to improve the heat transfer to the housing and
end-caps. EWdg Cavity is used to select Potted or not Potted from a pull-down list. The axial cross-
section is drawn with potting or not accordingly:
Potting Goodness [EWdg] Multiplier used to enhance/degrade the end-winding potting thermal
conductivity, i.e. used in calculation of thermal resistance between end-winding and housing/endcap.
It is used to account for air pockets within the potting material.
This parameter has no effect when End Winding Cavity = Not Potted - in this case it is disabled.
Visual editor used to edit the motors winding configuration, wire-selection and winding input
parameters. The layer model or conductor view is selected using the Winding View button selection.
Configuration:
The user can select the following winding configuration parameters from the options provided in the
form of a drop-down list:
Winding Type (i.e. overlapping or non-overlapping (solid divider or air divider))
Wdg Definition (i.e. slot-fill or conductors/slot)
EWdg Definition (i.e. EWdg-Fill or EWdg-MLT)
EWdg Cavity (i.e. is the end winding potted or not)
Mat [Liner-Lam] (i.e. Liner-Lamination gap air or impregnation)
Wedge Model (i.e. is the slot wedge area wound or a non conducting wedge used)
Note - a non-overlapping winding is one where the end-windings of a phase do not overlap that of any
other phase. The coil is often bobbin bound around an individual tooth. This winding type is common
in modern precision wound segmented motors and in switched reluctance motors. The overlapping
winding is the more traditional type of winding where the end windings of all phases overlap each
other.
Wire Selection:
The following Wire Type parameters are selected (drop-down list) or input:
Wire Type Wire Gauge
Wire Diameter Copper Diameter
The following winding input parameters can be edited and the winding drawing updated using the
redraw button:
Slot Fill Conductors/Slot
Redraw:
Winding Drawing:
An image of the layers used in the Winding Model is drawn in order to provide the user with visual
feedback - making input errors less likely. The impregnation layer thickness between copper layers
is calculated iteratively (to satisfy the required Slot Fill or Conductors/Slot).
Some other less widely changed winding settings are available in the winding settings sheet.
A selection of Winding Output Parameters are displayed such that the user gets immediate
feedback as to the validity of the input parameters. Such data as slot fill and slot/winding area's are
presented to the user.
Units:
The units of length (mm or inches) are indicated on the status line.
Note: The conductor positions do not have any effect on the winding layer model and the winding
temperatures calculated using it. The size of the blocks of conductors is used in the cuboidal
winding model to determine the size of the cuboids.
The conductor positions are also important for use with the FEA module slot temperature calculation
.
Motor-CAD will populate the position table using the current winding as shown below:
When the automatic positioning is enabled then the conductors are positioned based on the position
of the conductor at the corner of the slot next to the slot wall and slot base. This position can be set
using the Conductor - Base Gap and Conductor - Tooth Gap parameters (see Automatic Separation
option in Winding Editor View.)
More accurate positioning of the conductors can be done by disabling the automatic positioning. The
conductor columns can then be accurately placed by defining the centre position of the top left or
top right conductor of the group of columns. The mesh is complex to construct for all the conductors
in the slot so there is an algorithm for the meshing that assumes that the conductors are in
continuous columns down the slot. Please be aware that moving the conductors around the slot may
cause problems in meshing if this mesh algorithm is broken so you should try to keep the column
positions of the conductors relative to each other.
The copy conductor positions buttons allow the duplication of both sides of the slot.
Motor-CAD does not check for the conductor positions being valid. This is left to the user as allows
more flexibility in the placement. The default conductor positions are not always valid as shown
below:
Many electrical machines have random wound mush winding in which it is impossible (or desirable)
to model the position of each individual conductor. Even with precision and form wound windings it is
often not necessary to model each individual conductor. (See form wound winding model). Also, the
end windings have in many cases a more random nature than the active section of the motor. For
these reasons a layered model is used to represent the winding in the thermal circuit as it gives a
rough indication of the temperatures throughout the winding and has some physical relationship to
the geometry and slot fill used - this is useful for understanding and visualisation purposes. The
typical layered model used is shown below.
In order to calculate a thermal resistance we need a component length, cross-sectional area and
thermal conductivity (R = l/(kA)). If we have network nodes on the stator lamination (tooth and back
iron nodes) we then need a set of thermal resistances to represent the temperature rise within the
winding and to define how the winding is coupled to the stator lamination nodes. There is a slot liner
at the boundary of the stator slot (with an unknown gap between it and the lamination wall). We
know the thickness of liner, cross-section area of the slot wall and liner material (see - Liner Layers)
so we can calculate the slot liner thermal resistance. On the inside of the liner there is a mix of
copper, impregnation and air. By placing layers of copper and insulation within the slot-liner (having
the same general shape as the liner) as shown in the diagram above we are able to define
component (copper and insulation) lengths and areas and so calculate a set of thermal resistances
like those shown below:
The good thing is that the layered model gives an indication of the actual temperature build up in the
actual winding with the hot spot at the slot centre.
In the original layered model developed in Motor-CAD the copper layer thickness and wire insulation
(wire enamel) layer thickness were made equal to Copper Diameter and ([Wire Diameter - Copper
Diameter]/2) respectively. The impregnation layer thickness (Impreg Thickness) between copper
layers was then calculated (iterative technique) in order to satisfy the required Slot Fill or
Conductors/Slot). This model made sure that the total amount of covered wire in the model was
equal to that of the actual motor. Also the amount of impregnation was equivalent to that in the
actual machine. The problem was that by setting the enamel layer thickness in the model to that of
the actual enamel for the wire it is impossible to achieve a copper slot fill equivalent to that of the
actual machine (the wire slot fill is correct but the copper slot fill is not). For this reason an
alternative Winding Layer Model = [Wire & Copper Area] where the wire insulation (wire enamel)
layer thickness is allowed to vary to give correct values for wire slot fill and copper slot fill was
developed. To select the new model use the radio button selector shown below (in Calculation
Options [Input Data Editor]):
The typical difference in winding layer model using the two model types is illustrated below:
It is seen that the [Wire Area & Copper Area] model has more wire insulation (coloured purple). This
is the more correct model as all the components (copper, enamel and impregnation) are included.
Losses:
The copper loss is distributed between the layers according to their volume. Also the active and end-
winding losses are distributed according to their respective volumes:
The diagrams below show the user the distribution of losses in the active and end windings:
The present winding model has just one axial slice to represent the heat transfer in the radial
direction for the active section of the winding. The Axial Slice model option allows more slices in the
axial direction and will give a more accurate representation of temperature differences along the
length of the machine at the cost of model size. There are just two resistances that link the active
section of the winding to the end winding (model the temperature difference between the active and
end windings due to axial winding heat transfer). The formulation used to calculate the resistances
between the active and end windings are detailed below.
The end winding length and cross section is used in the calculation as illustrated below. The user
can use the Winding Rt[Active-EWdg] Length Multiplier to calibrate the model.
The EWdg Active Layer Connect option in Input Data [Calculation Options] allows the user to make
the connection between the end winding node and the active winding at the mid layer node (Mid
Active Layer) rather than the default of the outer layer node (Outer Active Layer). This is to try and
help overcome any modelling deficiency due to the lack of nodes available in the present model to
give an accurate representation of the winding active and end-winding heat transfer. The new
connection can help give better results for a Non Overlapping winding where conductors furthest from
the tooth sides in the active section tend to be those showing at the extremity of the end-windings
(the EWdg Active Layer Connect = Outer Active Layer can give better results in this case). We
could have made the connection from the end-winding node to the inner most active layer - but it was
decided that this may over predict any cooling due to the end-windings. The end turn length will tend
to be longer for the end turns at the extremity of the end winding than the internal ones - Motor-CAD
uses an average end turn length. The original default tends to give good results for traditional
Overlapping windings where the conductors closest to the tooth sides in the active section tend to
be those showing at the extremity of the end-windings (winding outer layer). The increased axial
layer model that under development will make this option redundant and it will give finer resolution of
temperatures throughout the active and end-windings.
Impreg Thickness Liner Impregnation/airgap thickness between outer copper layer and slot liner.
Slot Depth (Windable) Radial depth of slot available for winding - from winding push-back to slot-
bottom.
The following winding output parameters are displayed so that the user gets immediate feedback as
to the validity of the input parameters:
Wire Slot Fill (Wdg Area) - the proportional area of wire (copper + insulation) in the winding area
available in the slot (after removal of wedge and liner)
Wire Slot Fill (Slot Area) - the proportional area of wire (copper + insulation) in the total slot area
Copper Slot Fill (Slot Area) - the proportional area of copper in the total slot area
Winding Depth
Wedge Depth
Tang Depth
Impreg Thickness
Impreg Thickness Liner
Wire Ins Thickness
Winding Periphery
EWdg Layer Cond Mult
Conductors Drawn
Slot Area
Winding Area (+Liner)
Winding Area
Covered Wire Area
Copper Area
Impreg Area
Wedge Area
Coil-Divider Area
Liner Area
Liner-Lam Imp Area
For further details of these parameters see Winding Geometry, Winding Areas and Winding
Peripheries
In a Claw Pole Machine and a Synchronous Machine there is a stator winding which is defined using
the Stator Winding Editor and a rotor winding which is defined using the more simple editor shown
above.
The rotor has a bobbin wound coil. At present the rotor winding model is not as sophisticated as that
of the layered stator winding model. We just set the following parameters:
Rotor Slot Fill
Rotor Gap Liner Lam
Rotor Impreg Goodness [Active]
Rotor Impreg Goodness [EWdg]
Rotor Impreg Goodness [Liner-Lam]
The resistance/power source network that we are using to model the rotor winding and claw poles is
shown below:
In the present model we model the rotor coil losses as a single power source. There is dissipation
via the coil sides, top and base. In the Claw Pole machine it is assumed that coil top is insulated.
The thermal path between the coil centre node and connecting node may pass through a slot liner
and liner-rotor interface gap (may be partially or completely filled in impregnation) if shown. In the
Synchronous machine we take the coil dimensions and slot fill to calculate an effective thermal
resistance for impregnation. In the Claw Pole machine the resistance length is taken as ½ coil depth
multiplied by (1-Slot Fill) for the coil base and ¼ coil axial length multiplied by (1-Slot Fill) for the coil
sides.
This simplified winding model should be calibrated to give good results. A more sophisticated layers
model for the rotor winding may be added in a later version (this will require less calibration as it
more closely matches reality).
A new winding model for the Synchronous machine rotor has been added in Motor-CAD v3.1.5
Details of this new winding model are given in the Synchronous machine section.
There are a series of editors for editing non-visual input parameters that were not set within the
cross-section and winding editors. A tab selector is used to separate the parameters into the
following topics:
Cooling Options
Losses
Materials & Weights
Interface Gaps
Radiation
Natural Convection
Blown Over - only visible if Blown Over (TEFC) cooling is enabled
Wet Rotor - only visible if the Wet Rotor cooling is enabled
Housing Water Jacket - only visible if the Water Jacket Housing cooling is enabled
Shaft Spiral Groove - only visible if the Shaft Spiral Groove cooling is enabled
Spray Cooling - only visible if the Spray Cooling is enabled
Rotor Water Jacket - only visible if the Rotor Water Jacket cooling is enabled
Slot Water Jacket - only visible if the Slot Water Jacket cooling is enabled
Duty Cycle
Settings
Material Properties
See also Radial Cross Section Editor, Axial Cross Section Editor, Winding Editor
The views below show different size plates mounted to the front and rear of the axle.
Pull down menu used to select the induction motor bottom cage type.
The Top Bar and Bottom Bar shapes should be selected to give the best possible match with that
used in the electromagnetic design. It is not as important to have an exact match of the geometry as
it is mainly the slot periphery and rotor copper loss that is important for the thermal calculation.
The range of rotor slot shapes than can be modelled are illustrated below:
Note that the Bottom Bar type is not available if single cage type chosen.
Pull down menu used to select the brushless permanent magnet rotor type.
Magnet Mounting:
Surface
Embedded
Inset
Spoke
Magnet Shapes:
Radial
Parallel
Breadloaf
Flat
V-Shape
The different types and their circuit changes are shown below:
The following new BPM rotor types have been added in v4.1
Surface Parallel
Inset parallel
Embedded Parallel
Surface breadloaf
Inset breadloaf
Embedded breadloaf
Interior V-Shape
Interior U-Shape
Spoke Magnet
Other rotor types will be added in future versions of Motor-CAD (please contact support if you have
any special requirements).
Now the user can introduce Bayonet type brushes into the model using the [Brush Holder]
parameter (note that at present this parameter can only be set on the radial cross section editor - it
will also be available on the axial cross section editor in a future version of Motor-CAD):
The inner section of the brush holder is usually metallic and the outer section (drawn in black) is
usually and insulating material. The materials can be selected in the editor shown below:
The interface gap between the brush holder and the brushes and the brush holder and the housing
can be set in the editor shown below:
Good default values are difficult to judge without some calibration using test. Air is assumed for the
interfaces. In some cases the brush holder will be glued into the housing structure - in such cases
the user can reduce the interface gap accordingly.
The dimensions used to define the brush holder are shown below:
(Brush Holder Height, Brush Holder Ins Thickness, Brush Holder Outer Dia, Brush Holder Comm
Gap)
The [Brush Holder Height] is taken to be the height of the metallic insert. The upper limit of the
insulating section is taken to be the outer diameter of the housing. The insulating outer section of
the brush holder is assumed to be round. The metallic inner section of the brush holder is assumed
to rectangular. The thermal resistance of the brush holder is taken to compromise of two
components:
The length of the outer section of the outer section of the brush holder. The thermal resistance length
is taken from central section of the housing to the end of the inner metallic insert and the area is
taken from the dimensions of the outer cylindrical and inner Toolsections of the brush holder - see
diagram below:
The radial thickness of the outer section of the brush holder. The thermal resistance length is taken
as the thickness of the outer brush holder material. and the area is taken from just the section of the
outer part of the brush holder that passes through the housing, i.e. only the area associated with the
housing thickness - see diagram below:
The weight of the brush holder is calculated and added to thermal capacitance of the brush node.
8.1.1.5 Cowling
Pull down menu used to select the fan cowling type - if used.
A drawing of the cowling is included in the Radial Cross-Section Editor and Axial Cross-Section
Editor
Pull down menu used to select the encoder (feedback) case design - if used.
The space between the rotor poles in a SR machine can be open to air flow or filled with some form
of insulation. If filled then this can result in a reduced windage loss but it will reduce the amount of
rotor dissipation area and the area that air can flow axially between poles.
At present the insulation is not plotted - this will be added to a later version of the software.
The Housed Outer Rotor BPM machine has a static outer housing and a rotating shaft like the more
traditional inner rotor brushless permanent magnet motor.
The outer rotating magnets are mounted on the inside of a rotor cup that is attached to the shaft.
The rotating shaft is coupled to the stationary axle via the bearings. The inner stator lamination is
mounted on the outside of the axle.
The machine has in effect 3 rotating airgaps, the innermost between the shaft and axle, the main
middle magnetic airgap between the stator lamination and rotating magnets, and the outer airgap
between the rotor cup and housing.
The Housed brushless permanent magnet outer rotor (BPMOR) geometry is shown below:
Housing Selection:
The [Housed] BPMOR machine is selected using the [Rotor Mounting] combo selector in the Radial
or Axial Cross Section Editors:
Dimensions:
The main new dimensions associated with the Housed BPMOR geometry are shown below:
Schematic:
The rotating shaft is shown along the top edge of the schematic (grey resistances). The shaft is
coupled to the dark green static axle resistances drawn below the shaft via the bearing at the front
and rear of the machine and the rotating airgap between shaft and axle (3 resistances - one for the
active part and one each for the fronr and rear overhanging sections of shaft and axle). The active
part of the axle is coupled to the stator lamination bore (stator lamination resistances shown in red).
The outer surface of the stator lamination is coupled to the inner surface of the rotating magnets
(geeen resistances) via the magnet airgap. The outer section of the magnets are connected to the
inner surface of the rotor cup (shown in light blue) that forms the magnetic return path. Base of the
cup is connected to the rotating shaft at the rear of the machine. The outer surface of the cup has an
airgap to the static housing that drawn in royal blue at the bottom edge of the schematic.
All three airgaps in the machine are modelled using the rotating airgap convection correlation based
on Taylor's theory on concentric rotating cylinders. Details of if the heat transfer mode across the
gap is via conduction (if laminar flow), or convection (if vortex or turbulent flow) is given in the Output
Data sheet, i.e. details of the Airgap Reynolds number together with the critical Reynolds numbers
for vortex and turbulent flow and the airgap heat transfer coefficient:
All gaps in the model also allow for both convection and radiation. The radiation component is only
included in the model when the [Include Internal Radiation in the Lumped Circuit Model] checkbox is
checked. If the radiation component is to be included then the user needs to provide the values for
the emissivity for the airgap surfaces:
Details of the new Cup component are included in the relevant editors where required, as shown
below:
Note that the Rotor Lam [Back Iron] is taken to be the active part of the rotor cup, the Rotor Cup OH
the overhanging cylindrical section of the rotor cup and the Rotor Cup Base the disc base section.
With the Housed BPMOR the user can select any housing type as the housing is static. In the other
types of BPMOR the outer housing is rotating and so the cylindrical surface is assume and fixed.
For instance below we have a radial finned housing on a Housed BPMOR machine:
The convection from the static housing is calculated as for the more traditional BPM machine, i.e.
natural convection or forced convection if Blown Over (TEFC). Note that in the case of the other type
of BPMOR machine in Motor-CAD the convection from the rotating outer surface of the housing is
calculated using the Forced convection correlations with the air velocity calculated from the
rotational speed and surface radius of the housing.
Pull down menu used to select one of the above housing types. Pictures of the different housing
types available are shown below. The user should select the housing type which closest matches
the one they intend to manufacture. The fin size and spacing and the corner cutout region for
mounting to the connected device can all be set in the Radial Cross-Section Editor and Axial Cross-
Section Editor editors.
BPM-EMag - electromagnetic model for inner and outer rotor brushless permanent magnet
motors
Thermal calculations can be carried out for an free-standing (not mounted) motor or for flange and
foot mounting, the mounting aiding heat dissipation.
The temperatures of both the Flange and Feet can be set to a fixed temperature (as can the front of
the shaft) using the Fixed Temperatures option.
The BMPOR rotor mounting type is set in Radial Cross-Section Editor or Axial Cross-Section Editor.
The parameter is used when Motor Type = BPMOR-Therm. It is used to select an open outer
rotating surface [Open] to embed the motor in an extra insulated layer [Insulated] or to give the
machine a static housing like a more conventional BPM machine.
(1) Open
(2) Insulated
The Insulated option is used to model the case where the motor is mounted within some form of
equipment - in many cases the equipment is not a good thermal cooling path.
In the insulated case the insulation thickness and axial length are set using the Cover Ins Thickness
and Cover Ins Length parameters.
The insulated cover material properties are set in the Materials & Weights [Input Data Editor]:
See BPMOR-Therm
The input parameters Spider Arm Length, Spider Radial Thickness and Spider Arm Width are used
to specify the spider geometry.
The construction of the model from the parameters can be quite involved as the parameters are quite
interlinked. The above example will help to understand the parameters.
The material properties for the new components have to be added in the material and weights editor:
The node temperature view shows the new nodes used for the single bearing model:
In the schematic view the circuit has been modified as shown below:
It is also possible to add an inner water jacket with cooling channels in the axle component and/or
stator plates as shown below. This is done by selecting the Shaft Spiral Groove Cooling.
Pull down menu used to select the slot/tooth type. Parallel tooth (default), parallel tooth SqBase,
Parallel slot, Slotless or Form Wound can be used.
The Tooth Width or Slot Width is then used to define the tooth or slot width.
Pull down menu used to select the induction motor top cage type.
The Top Bar and Bottom Bar shapes should be selected to give the best possible match with that
used in the electromagnetic design. It is not as important to have an exact match of the geometry as
it is mainly the slot periphery and rotor copper loss that is important for the thermal calculation.
The range of rotor slot shapes than can be modelled are illustrated below:
The form wound machine type is selected by selecting the Slot Type = Form Wound in the radial
cross-section editor.
The form wound machine winding is defined in the Winding Editor as shown below:
The slot into 2 coils (groups of conductors). Each coil is made up of several bars. The coils are
separated by a phase spacer.
Each bar has its own insulation.
See Form Wound Winding Definition for more details of the geometry definitions.
In the Materials & Weights editor the materials for the components inside the slot can be specified
and the weights of the components is calculated and displayed
In the Motor-CAD thermal model for the form wound machine there are 2 nodes one for each coil.
The winding layer model is not used as the position of the conductors is accurately known so
thermal resistance paths from the coils to the tooth wall and slot bottom can be calculated.
The Schematic with the form wound winding components is shown below:
The default form winding model created by Motor-CAD is usually quite accurate, however there are
thermal leakage paths though the slot wedge and the coil divider that mean that there may be a
small error in the calculated thermal resistance paths. If required the model can be calibrated from
test results or FEA Analysis see form wound calibration settings.
8.1.1.18 Fan
A shaft mounted fan can be defined using the options shown above. The fan can be placed at either
end of the machine and inside or outside of the endcaps.
At present the fan is not used in any Motor-CAD calculations however is used for visualization and
for export of the geometry to other 3D tools. It is planned to add a fan model into Motor-CAD in the
future.
Fan Axial Position - used to position the fan on the shaft. Can be to front or rear of machine or inside
the front or rear endcaps.
Fan Blade width - the axial length of the fan blades
Fan Blade offset - the relative axial position of the the fan blades relative to the hub
Fan Backplate width - the axial length of the fan backplate in which the blades are recessed.
Fan Hub Extension - the axial length of the fan hub.
8.2 Geometry
8.2.1 Airgap Geometry
Airgap: Airgap between rotor and stator. N.B.: not physical airgap but includes any magnet
retainment banding and/or stator bore sleeve:
The banding is only included in the thermal circuit if the thickness is greater than zero.
See Radial Cross-Section Editor, Axial Cross-Section Editor, and Airgap Model.
Stator Axial Offset: Amount that axial centre of stator stack is offset from housing axial centre
[+ve towards front of motor].
Magnet Axial Offset: Amount that axial centre of magnet is offset from stator stack axial centre
[+ve towards front of motor].
Rotor Axial Offset: Amount that axial centre of rotor stack is offset from stator stack axial
centre [+ve towards front of motor]:
Bearing Offset [F] Bearing Offset [Front of motor]. This is the distance the bearing is offset from
the endcap.
Bearing Offset [R] Bearing Offset [Rear of motor]. This is the distance the bearing is offset from
the endcap.
Sleeve Thickness:
Thickness of any stator bore sleeve. N.B. - The Airgap parameter includes any retaining sleeve or
banding:
The sleeve is only included in the thermal circuit if the thickness is greater than zero.
The stator sleeve is assumed to extend over the full length of the machine:
If a sleeve is used then the cooling around the end windings and inside of the housing and endcap is
adjusted automatically to account for the fact that the end space cooling of the these surfaces is
due to natural convection only, i.e. k2 in the End Space Convection Correlation set to zero:
If the Wet Rotor cooling method is used together with a sleeve the thermal resistance on the inside
of the end winding is automatically increased to account for the conduction through the sleeve and
any potting or airgap between the sleeve and the end winding. Conduction is assumed through the
airgap rather than convection due to the small thickness.
Gap [Endcap-Comm]: Gap between inner surface of endcap and the axial end of the commutator.
Gap [Comm-Winding]: Gap between axial end of the commutator and the end-winding.
The parameter only introduces a gap between fin sections when Housing Type = Axial Fins (Rd) if
Fin Extension > 0.
When axial fins are used Cutout Corner [%] represents the minimum housing corner cutout - the
actual cutout being determined by Fin Thickness and Fin Pitch/Thick
This parameter is not applicable to Housing Type = Round and Square.
Difference between Housing Dia and diameter of housing corner cutout (Housing Cutout Dia). The
area is used for inserting flange mounted bolts.
The parameter only applicable to the housings (Housing) that have cutouts.
A drawing of the cowling is included in the Radial Cross-Section Editor and Axial Cross-Section
Editor
Enc Radial Gap: Gap between encoder and encoder case in radial direction.
Enc Barrier Length: Axial length of thermal barrier between motor shaft and encoder shaft.
Enc Axial Gap: Gap between encoder and encoder case in axial direction.
EndRing Add [Outer F] Amount by which induction motor end ring extends outwards beyond slot
opening radius at front of machine.
EndRing Add [Inner F] Amount by which induction motor end ring extends inwards beyond slot
bottom radius at front of machine.
EndRing Add [Outer R] Amount by which induction motor end ring extends outwards beyond slot
opening radius at rear of machine.
EndRing Add [Inner R] Amount by which induction motor end ring extends inwards beyond slot
bottom radius at rear of machine.
EndRing Thickness [F] Thickness of the endring at the front of the machine.
EndRing Thickness [R] Thickness of the endring at the rear of the machine.
EndRing Extension [F] Extension of the rotor bar from the rotor lamination at the front of the
machine.
EndRing Extension [R] Extension of the rotor bar from the rotor lamination at the rear of the
machine.
EWdg-Housing [F] Gap between end winding and housing [front of motor].
EWdg-Housing [R] Gap between end winding and housing [rear of motor].
EWdg-Bore [F] Gap between end winding and stator bore [front of motor].
EWdg-Bore [R] Gap between end winding and stator bore [rear of motor].
Wdg Extension [F] Amount that winding extends beyond stator lamination before bending
into end winding [front of motor]
Wdg Extension [R] Amount that winding extends beyond stator lamination before bending
into end winding [rear of motor]
EWdg-Endcap [F] Gap between end winding and endcap [front of motor]
EWdg-Endcap [R] Gap between end winding and endcap [rear of motor]
Note: The channels that are created in Wdg Extension [F] and [R] can be used to pass air through
in the End Space Convection Correlation - by default this cooling path is neglected as its k1 factor is
set to zero.
More recently an addititional method of specifying the endwinding dimensions has been provided.
(see Geometry Settings)
Winding Add [Outer F]: Amount that winding extends radially outwards beyond the end winding
extension on the armature. Used for drive end (front) end-winding.
Winding Add [Outer R]: Amount that winding extends radially outwards beyond the end winding
extension on the armature. Used for non drive end (rear) end-winding.
Winding Add [Inner F]: Amount that winding extends radially inwards beyond the end winding
extension on the armature. Used for drive end (front) end-winding.
Winding Add [Inner R]: Amount that winding extends radially inwards beyond the end winding
extension on the armature. Used for non drive end (rear) end-winding.
The EWdg Insulation input parameters allow an extra layer of insulation to be modeled on the end
windings as shown below:
The thermal properties of the insulation are entered in the Materials & Weights editor as shown
below:
Fin Number [Quadrant] The number of fins on 1/4 of the housing periphery in a axial finned
machine. Used when Fin Input Options = Fin Number & Spacing
The Fin Spacing [Fin Base] is also used to define the dimensions of the fins on the housing. The
Corner Cutout [%] is automatically adjusted to accommodate the fins in this case.
In a radially finned housing the fin number is not specified and the Fin Spacing [Fin Base], Fin
Thickness and the axial length of the machine is used to calculate the number of fins that can be
fitted to the housing.
Fin Pitch/Thick Used when Fin Input Options = Fin Pitch/Thickness. Fin pitch/thickness
ratio, where:
Fin Pitch = Fin Pitch/Thick x Fin Thickness
Fin Pitch [Calc] This is not a true input parameter, but is calculated from the fin thickness &
fin pitch to thickness ratio:
Fin Pitch = Fin Pitch/Thick x Fin Thickness
Fin Spacing [Fin Base] The fin spacing at the base of fins placed on the housing. Used when Fin
Input Options = Fin Number & Spacing
The Fin Number [Quadrant] is also used to define the dimensions of the fins on an axially finned
housing. The Corner Cutout [%] is automatically adjusted to accommodate the fins in this case.
N.B.: The fin pitch is calculated at Housing Dia. in motors which have axial fins that extend out in a
radial direction
It is typical when performing rating tests on flange mounted servo motors to mount them to a plate
that acts as a heat sink. This is to represent the cooling that can be achieved in a practical
situation via conduction to the apparatus to which the motor is attached to. NEMA MG 7-1993
section 2.6.1.6, 'Mounting Configurations for Servo Motor Output Tests' gives recommended flange
mounted cooling plate dimensions. However, in practise manufacturers often use flange plate larger
than the recommendations in order to optimise their catalogue data. Motor users should be careful if
they choose a machine from a catalogue that is to be used in an application that does not allow
good cooling via the flange. In such cases the catalogue data should be de-rated. Motor-CAD is an
ideal tool for calculating the de-rating.
In some cases the motor may be mounted to a device that has a fixed temperature. The Fixed
Temperature options in Motor-CAD can be used to model such cases.
In some cases the motor may be mounted to a device that has a fixed temperature. The Fixed
Temperature options in Motor-CAD can be used to model such cases.
Feet Width [Motor]: Feet width at junction with motor [assuming 2 feet].
Feet Width [Base]: Feet width at junction with base [assuming 2 feet].
Housing Length: Housing axial length [including front and rear end-cap thickness]:
In order to stop confusion about the Housing Length (in that it includes the endcap axial thickness) it
was decided to rename Housing Length to Motor Length
Cover Ins Thickness: The parameter is used when Motor Type = BPMOR-Therm and Rotor
Mounting = Insulated.
It is used to set the radial thickness of the insulation on the outside of the machine.
Cover Ins Length: The parameter is used when Motor Type = BPMOR-Therm and Rotor Mounting =
Insulated.
It is used to set the axial length of the insulation on the outside of the machine.
Pole Arc: The pole arc for interior magnets in the rotor lamination in electrical degrees. The
angle has its centre at the centre of the machine.
Pole V Angle: The angle between the magnets in a pole in mechanical degrees.
Bridge Thickness: How far the outermost point of the magnet is embedded in the rotor lamination.
Magnet Separation: The distance between the magnets in a pole of the interior V-Shape rotor
lamination.
Magnet Width Reduction: The reduction in total magnet length for a pole from the maximum
possible for the interior magnet geometry in the rotor lamination.
Web Length: The length of the web between adjacent poles in the rotor lamination.
Web Thickness: The thickness of the web between adjacent poles in the rotor
lamination.
The distance that the radial magnets are embedded in the rotor lamination in a BPM Rotor =
Embedded Radial design
Magnet Inset : Minimum distance from rotor surface to the magnet outer surface.
Magnet Depth : Length of the magnet in the radial direction
Magnet Opening : Distance of Magnet opening at rotor surface
Magnet Thickness : Thickness of the magnet
Flange extension: Amount that the flange extends beyond Housing Dia. in a radial direction.
Flange Depth: Amount that flange extends axially beyond housing, i.e. flange mounting
interface.
Armature Diameter Rotor diameter in a PMDC motor and stator diameter in a BPMOR motor.
Back Iron Thickness Stator back iron thickness in a PMDC motor and rotor back iron thickness
in a BPMOR motor.
Housing Thickness Housing thickness in a PMDC motor and rotor housing thickness in a
BPMOR motor.
Bar Opening [T]: Induction motor cage top Bar Opening Depth [T]: Induction motor
bar opening. cage top bar opening depth.
Bar Width [T]: Induction motor cage top bar Bar Depth [T]: Induction motor cage top
width. bar depth.
Bar Opening [B]: Induction motor cage Bar Opening Depth [B]: Induction motor
bottom bar opening. cage bottom bar opening depth.
Bar Width [B]: Induction motor cage bottom Bar Depth [B]: Induction motor cage
bar width. bottom bar depth.
Rotor Tooth Width: Induction motor tooth width (only used when Top Bar = Parallel Tooth)
Bar Corner Radius: Corner radius at bottom of induction motor rotor slot (only used when Top Bar =
Parallel Tooth)
Bar Tip Angle [T]: Induction motor rotor tooth tip angle (top bar).
The type of rotor ducts used in a Through Ventilation model is set using the Rotor Duct Types
selector shown above. This is set in the Radial Cross-Section Editor. The choice is:
None
Circular Ducts with up to 4 layers of ducts
Shaft Spoke
Circular ducts
Arc ducts
The parameters to set the dimensions of the particular duct is also set in the Radial Cross-Section
Editor
When the Rotor Duct Type is set to Circular Ducts then the Circular RDuct Layers sets the number
of duct layers. Each duct layer can have a different location, size and number of ducts.
Rotor EWdg Ohang: Synchronous machine rotor coil axial overhang dimension.
See also Shaft Type for details of the spider shaft geometry.
See Shaft Spiral Groove [Input Data Editor], Radial Cross-Section Editor and Axial Cross-Section
Editor.
A hole can be put in the shaft. Its diameter is defined by the Shaft Hole Diameter parameter.
At present the shaft hole only has fluid passed down it for the Spray Cooling model. For other model
types there is no fluid passed down it, the hole is just used in calculating the shaft weight (transient
calculation) and thermal resistance of the shaft in the axial direction.
If Shaft cooling is requires then the Shaft Spiral Groove cooling type should be used to cool the shaft
with an external liquid.
When the Rotor Duct Type is set to Shaft Spoke then the Number Shaft Spokes sets the number of
ducts in the shaft. A spoke is a term for the spacer between ducts in this type of cooling system - it
resembles the spoke of a bicycle wheel. The Spoke Radial Depth and Spoke Thickness parameters
are used to set the spoke (duct) radial depth and the thickness of the spokes (spacing between
ducts).
When the Rotor Duct Type is set to Shaft Spoke then the Number Shaft Spokes sets the number of
ducts in the shaft. A spoke is a term for the spacer between ducts in this type of cooling system - it
resembles the spoke of a bicycle wheel. The Spoke Radial Depth and Spoke Thickness parameters
are used to set the spoke (duct) radial depth and the thickness of the spokes (spacing between
ducts).
Slot Width: width of slot for a parallel slot design (Slot Type):
Stator Pole Angle: Switched reluctance motor - stator pole angle at stator bore.
Rotor Pole Angle: Switched reluctance motor - rotor pole angle at rotor surface
Stator Pole Radius: Switched reluctance motor - corner radius of stator slot bottom
The Lamination - Housing Divider Ducts can only be used for Through Ventilation cooling.
If the Circular or Rectangular Ducts are in stator lamination then they can only be used for Through
Ventilation cooling.
If the Circular or Rectangular Ducts are in the housing (and the housing type is not square) then
they can only be used for the Housing Water Jacket cooling. (unless the Use Circular Ducts in
Housing for TVent Flow is enabled)
If the Circular or Rectangular Ducts are in the housing with a square housing then they can only be
used for Through Ventilation cooling.
See also Duct Calculations for details of the duct calculations used.
Winding Separator The axial length of each of the coil dividers in a Synchronous Machine.
Length
Wdg Separator Lower Parameter to define the lower limit of the coil divider in a Synchronous
[%] Machine. A percent of the lower limit of the rotor coil side. A value of 0
means that the separator will come to the lower limit of the coil side. A
value of 25% means that it will be 25% from the base of the coil side.
Wdg Separator Upper Parameter to define the upper limit of the coil divider in a Synchronous
[%] Machine. A percent of the upper limit of the rotor coil side. A value of 100
means that the separator will come to the upper limit of the coil side. A
value of 75% means that it will be 25% from the top of the coil side.
The winding separators effect the Through Ventilation Model for the airgap.
The following rotor geometry parameters are shown in the diagram below:
Tooth-tip depth: depth of slot opening from bore to tooth taper region.
If the Slot Width parameter is bigger than the Slot Opening, then 0 <= Tooth Tip Angle < 90.
If the Slot Width parameter is smaller than the Slot Opening, then -90 < Tooth Tip Angle <= 0.
If the slot width and slot opening parameters are the same then the Tooth Tip Angle can be anything
as it is not used for setting the geometry.
For induction machines the wafters are placed on the endrings. BPM machines can also have
wafters and the following parameters are used to specify the radial position of the wafters.
The dimensions of the housing spiral groove channel are shown below - only used when Housing =
Water Jacket (Spiral)
See Housing Water Jacket [Input Data Editor], Radial Cross-Section Editor and Axial Cross-Section
Editor.
The stator duct number and duct width and the rotor duct number and duct width are specified in the
axial cross section editor as shown below.
There are many different possible configurations of machine cooling available in Motor-CAD.
Cooling Type
Blown Over where an external fan is used to blow a fluid (normally air) across the outside of the
machine. The dissipation from the outside of the motor is assumed to be a combination of Mixed
Convection (forced convection due to fan combined with natural convection), radiation and
conduction. The fluid can be air or any other fluid as set in the External Fluid Data and Internal
Fluid Data. Editors are available to edit the Forced Convection Data [Input Data Editor] and
Radiation Data [Input Data Editor].
Motor Orientation
The motor orientation can have a strong influence on Natural Convection cooling. It also has a
weaker influence on the Mixed Convection cooling, especially when the Forced Convection (Fan)
cooling dominates.
Cooling Systems
Through Ventilation
Housing Water Jacket
Shaft Spiral Groove
Wet Rotor
Spray Cooling
Rotor Water Jacket
Slot Water Jacket
Submersible
Flooded
If a particular cooling option is not available (i.e. housing type not suitable for a water jacket) then
that option is not enables (greyed out) and cannot be selected.
Miscellaneous Data
Includes Ambient Temperature, Shaft Speed, Lamination Stacking Factor.
See Miscellaneous Data
Fixed Temperatures
See Fixed Temperatures
A Check Data button is included so the data can be checked for it validity while still within the
Cooling Options editor. A data check is automatically carried out when moving from the editor, i.e.
pressing any other tab-selector.
Ambient temperature for convection from the outside of the machine. This can be different to
Ambient Temperature [Radiation] as the motor may view a very hot surface for radiation but be
surrounded by cooler air for convection (or vice versa), i.e. in an automotive application the motor
may be under the bonnet and be placed in view of the hot exhaust system which will tend to heat the
motor via radiation.
Ambient temperature for radiation on the outside of the machine. This can be different to Ambient
Temperature [Convection] as the motor may view a very hot surface for radiation but be surrounded
by cooler air for convection (or vice versa), i.e. in an automotive application the motor may be under
the bonnet and be placed in view of the hot exhaust system which will tend to heat the motor via
radiation.
If Ambient Temperature [Convection] = Ambient Temperature [Radiation] then the circuit shown
below is solved:
If Ambient Temperature [Convection] =/= Ambient Temperature [Radiation] then the circuit shown
below is solved:
In this case we cannot connect the thermal resistance due to convection and radiation to the same
nodes so we use a power source/sink component to represent the radiation heat transfer. If a value
is negative then the heat transfer is from the radiation ambient into the motor (radiation ambient
hotter than the particular nodal temperature).
See Cooling Options [Input Data Editor], Losses [Input Data Editor] and Calculation Options [Input
Data Editor].
Fixed Plate, Fixed Base, Fixed Shaft and Fixed Endcap Temperatures can be set using this
interface.
Checkboxes are provides to set temperatures for the following nodes in the thermal network:
Flange Plate
Foot mounted Base Plate
Shaft extension at drive end (front)
Shaft extension at non drive end (rear)
Endcap at drive end (front)
Endcap at non drive end (rear)
This can be useful for modelling the effect of coupling the motor to a device of a known fixed
temperature, i.e. the motor may be flange mounted to a pump of a known temperature.
The Fixed Plate and Fixed Base temperature options are only available if the particular mounting
type is selected using Mounting Type.
See Fixed Temperature method for details on the methods used for setting the fixed temperatures in
other locations in the model.
The values are set in the Cooling Options [Input Data Editor]
Lamination stacking factors should be set for the stator and / or rotor laminations.
The stacking factors are entered in the Cooling Options input editor as shown below:
The lamination model assumes that the lamination stack consists of 2 materials, the laminations
themselves and the inter lamination material.
The thermal material properties for both materials should be entered in the Materials & Weights input
editor.
where:
Note: the axial resistance values are only included in the model if the model has more than 1 axial
slice.
The radial thermal resistance values only take into account the laminations and assumes that the
inter lamination material has poor contact so little influence:
The stator and rotor thermal capacitance values are also calculated taking into account the volumes
of lamination and inter lamination material.
Losses are very important to the thermal model, however it can be difficult to know exactly the size
and location of the losses in the machine.
Motor-CAD has many different loss models available to allow the user to put their knowledge of the
losses with the thermal model.
It is also possible to model uneven distribution of losses in the machine. This may be in as the
machine has multiple windings or in stall or fault cases.
The losses for the thermal model should be entered here. These losses may be obtained from
calculation, the Motor-CAD BPM-EMag module or another FE package or from test results.
The losses settings here are used for both the steady-state and transient calculations (when using
the torque-speed or per unit (pu) loss value method where these values with be scaled with the pu
value specified.
For transient calculations the losses specified in the transient duty cycle editor will also be scaled
when using the loss variation options listed above.
A group of edit-boxes for inputting the losses in the machine, i.e. Copper Loss, Stator Iron Loss,
Friction Loss, etc
The loss are made a function of the Speed [RPM] when the Speed Dependant Losses check-box is
selected ( ). In this case the loss at Speed [RPM] is calculated using the
formulation indicated in the screen capture below:
i.e. Loss [@Shaft Speed [RPM]] = P[Input] * ( Shaft Speed [RPM] / Speed[Ref] )** coef [A]
where:
P[Input] Component loss (component = copper, stator iron, windage, etc.) at the reference speed
specified. This loss value will then be scaled by Motor-CAD with the machine speed.
Speed The reference speed (RPM) at which the component loss (component = copper, stator
[Ref] iron, windage, etc.) is provided by the user.
coef[A] Component loss (component = copper, stator iron, windage, etc.) speed to power
coefficient.
Note that power can also be injected (or changed) from a range of nodes for which a loss is not
usually associated by adding the power source or power injection in the circuit editor.
If the Speed Dependant Losses check-box is not selected ( ) the losses are
now longer a function of speed and the parameters that are longer required are made inactive as
shown below:
Note the in the table the W/kg for each of the components is also given. This is given so that the
more experienced user can make a judgment if all seems ok and that no errors have been made in
the input of the losses or the Materials & Weights [Input Data Editor]
A useful feature of the program is to allow the copper losses to be input at a specified winding
temperatures given in the editor. The copper losses are then varied according to the winding
temperature calculated in the thermal model.
The copper losses at the average winding temperature are calculated as follows :
Copper loss temperature increase factor = Loss [Copper]Winding Av erage / Loss [Copper]Giv en
Loss [Copper]winding Av erage = Copper loss temperature increase factor x Loss [Copper]giv en
where:
TGiv en is the given winding temperature as input in the above user interface
RphWinding Av erage is the winding resistance at the winding temperature in the thermal model (TWinding
])
Av erage
Loss [Copper]Giv en is the copper losses for the given winding temperature
Loss [Copper]Winding Av erage is the copper loss calculated based on the average winding temperature.
These are the copper losses used in the thermal model.
When the Copper Loss Variation with Temperature check-box is not checked the copper losses are
assumed to be those given in Input Data [Losses].
The resistivity coefficient of 0.00393 for the above equation is set as copper at 20C by default but
can also be modified see Losses Settings.
A useful feature of the program is to allow the copper & iron losses to be input at specified values of
winding and magnet temperatures (Winding Temperature - Tw(i/p) & Magnet Temperature - Tm(i/p))
then use the simulated winding & magnet temperature to calculate the actual copper & iron losses.
This can be used with both the steady state and transient solvers. It can be useful to use the
transient solver to simulate a typical motor rating test, i.e. either the current is held constant
throughout the test or the torque is held constant by varying the input current (both options can be
modelled in Motor-CAD - see Constant Torque or Constant Current option below).
Note: To use the Loss Variation with Temperature & Load method with the transient solver the
torque-speed duty cycle data definition must be used.
The following input parameters are then used to calculate the losses at the actual component
temperatures:
Losses Vary With Temperature and Load checkbox
Constant Torque or Constant Current radio button option selector
Winding Temperature - Tw(i/p)
Magnet Temperature - Tm(i/p)
Shaft Torque (@Pcu Defined)
Motor Current [Arms] (@Pcu Defined)
Torque Constant [Nm/Arms] [Output Parameter]
Torque & Current Multiplier
Rph @Tw(i/p)
Magnet Temp Coeff Br [%/C]
Iron Loss To Power Ratio
When the Losses Vary with Temp & Load box is checked the losses are input at the specified
winding and magnet temperatures. The copper & iron losses are then calculated at the winding &
magnet temperature using the formulation given below. This can be useful when using the transient
solver to simulate a typical motor rating test, i.e. the torque is held constant throughout the test by
varying the input current.
Constant Torque or Constant Current Sets if the Loss Variation with Temperature &
Load Model is based on constant torque (current varied to maintain a constant torque) or constant
current.
Shaft Torque [??](@Pcu defined) Shaft Torque at which the copper loss is defined -
used in formulation given below for calculating the variation in copper & iron loss with temperature &
load.
When Constant Torque or Constant Current = Constant Current, the Motor Current is used to
calculate the copper loss using the formulation:
Motor Line Current [Arms] at which the copper loss is calculated - used in formulation given below
for calculating the variation in copper & iron loss with temperature & load.
The Torque & Current Multiplier can be used to vary the current level at which the steady-state
thermal performance is calculated.
Torque Constant
This is an output parameter.
Motor Torque Constant [Nm/Arms] calculated from Shaft Torque (@Pcu Defined) / Motor Current
[Arms] (@Pcu Defined)
The [Torque & Current Multiplier] allows the user to calculate the thermal performance at different
loads very quickly by just changing one parameter.
The Steady-State or simple thermal transient thermal calculation is carried out at this multiple of the
Shaft Torque (@Pcu Defined) or Motor Current [Arms] (@Pcu Defined). Note: This multiplier is
ignored when carrying out a duty cycle transient calculation.
Rph @ Tw(i/p)
Phase resistance at Winding Temperature - Tw(i/p) - used in formulation given below for calculating
the variation in copper & iron loss with temperature & load.
Formulations
The following formulations are used to calculate the losses when the Losses Vary With Temperature
and Load check-box is selected ( ):
Flux Ratio = 1 + (T [Magnet] - Magnet Temperature - Tm(i/p)) * Magnet Temp Coeff Br [%/C]/100
Loss [Stator Back Iron] = Loss [Stator Back Iron] @Tm(i/p) * Flux Ratio**Iron Loss To Power Ratio
Loss [Stator Tooth] = Loss [Stator Tooth] @Tm(i/p) * Flux Ratio**Iron Loss To Power Ratio
Loss [Rotor Back Iron] = Loss [Rotor Back Iron] @Tm(i/p) * Flux Ratio**Iron Loss To Power Ratio
Loss [Magnet] = Loss [Magnet] @Tm(i/p) * Flux Ratio**Iron Loss To Power Ratio
Loss [Retaining Can] = Loss [Retaining Can] @Tm(i/p) * Flux Ratio**Iron Loss To Power Ratio
Loss [Iron] @Tm(i/p) = Loss [Stator Back Iron] @Tm(i/p) + Loss [Stator Tooth] @Tm(i/p) + Loss
[Rotor Back Iron] @Tm(i/p) + Loss [Magnet] @Tm(i/p) + Loss [Retaining Can] @Tm(i/p)
Loss [Iron] = Loss [Stator Back Iron] + Loss [Stator Tooth] + Loss [Rotor Back Iron] + Loss
[Magnet] + Loss [Retaining Can]
3. Calculate Mechanical Losses, i.e. friction, windage, iron & encoder losses
Mechanical Loss @Tm(i/p) = Loss [Iron] @Tm(i/p) + Loss [Windage] + Loss [Friction - F Bearing] +
Loss [Friction - R Bearing] + Loss [Encoder]
Mechanical Loss = Loss [Iron] + Loss [Windage] + Loss [Friction - F Bearing] + Loss [Friction - R
Bearing] + Loss [Encoder]
Iph = Iph @Tw(i/p) & Tm(i/p) * 1/Flux Ratio * EM Torque/EM Torque @Tm(i/p)
When the Losses Vary With Temperature and Load check-box is not checked the input losses are
assumed to be at the component simulated temperatures.
For details of the parameters see also Losses [Output Data] and Temperature [Output Data]
At stall (zero speed) there is often unequal currents in the different windings of a 3-phase machine
(assumed all but the PMDC). The amount of unevenness is dependent upon how the windings are
connected and what type of drive is used (sinewave or squarewave). The winding connections can be
set in [Input Data] - [Losses] - Stall Copper Loss Distribution:
The default connection is Equal [Ia = Ib = Ic]. This is the connection sometimes used in the
laboratory to heat the machine up with a dc source with all windings connected in series so as to
have equal loss in all windings. The other two cases are the sinewave and squarewave drive. At zero
speed the squarewave drive will have two phases on and one phase off. The total copper loss will
therefore be concentrated in 2/3 of the winding. The time average loss in the conducting phases will
be 3/2 times than if rotating. At zero speed the sinewave drive could have a worst case condition of
peak current in one phase and half peak current in the other two phases. The current in the worst
phase will be Ö2 times the rms current so the loss will be 2 times (I2R) that of the rotational loss in
the same phase.
When rotating Motor-CAD models the heat transfer in the machine using a one-slot model which
represents the copper loss divided equally between all slots. This model is clearly not representative
of the stall condition. At stall Motor-CAD tries to model the worst slot. In fact we will assume that all
the slots have the worst loss:
To achieve the situation that all the slots have the worst loss the copper loss is
artificially by increased by 3/2 in a squarewave drive and by 2 in a sinewave drive. The
[Stall Pcu Multiplier] factor in Motor-CAD is set to the 3/2 or 2 depending on the drive
type:
[Stall Pcu Multiplier] = Worst Phase Loss / Phase Loss Assuming Even Loss Distribution
This additional copper loss is removed from the stator back iron as the total
dissipation from the full machine is equal to the losses defines in [Input Data] -
[Losses] rather than the artificially increased copper loss plus other losses (i.e. only a
proportion of the stator slots dissipate the worst case copper loss through the stator).
In the steady state calculation the total additional loss is removed from the stator
back iron node
In a transient calculation the power taken from the back iron is adjusted to account for
the storage term due to the thermal capacitance. In this case the amount of power
taken from the node is equal to the additional power multiplied by the instantaneous
total power into the stator node plus the power from the end windings divided by the
total copper loss.
Notes:
In versions before v3.1(1) the loss adjustment was made at both the stator
tooth and back iron modes (proportionally according to the tooth and back
iron periphery). Also no adjustment is made in the transient calculation for the
storage of the thermal capacitance.
In versions 2.1(20) to v3.1(0) an adjustment was made to the thermal
resistances associated with the winding and well as the adjustment of the
power values. The area used to calculate the conduction thermal resistances
associated with the windings was reduced by a factor of 2/3 in a squarewave
drive and 1/3 in a sinewave drive. This was to account for the fact that we are
only modelling the worst slots. It has been found that if we make this
adjustment together with that for the power that we are in effect accounting for
the stall condition twice and over predict the winding temperature rise.
In versions 2.1(20) to v3.1(0) an adjustment was made to the winding
capacitance. This was reduced according to the same rules as used for the
conduction thermal resistances associated with the windings. This has now
been removed.
The test case below illustrates the effects of Stall in a 12 slot machine.
Below we see the steady state calculation for a particular BPM motor at 1rpm:
Now at 0 rpm with a squarewave drive (Ia=Ib, Ic=0) we obtain the following results:
The winding hotspot is now 97C rather than 92C. The housing is as before. It is seen that the loss in
the winding has been artificially increased by a factor of 3/2. The powers displayed inside the stator
housing are as if all the slots have the worst loss. An adjustment is made at the stator back iron
node and the powers are now those for the actual machine. This assumes that the spreading of the
power in the back iron is quite efficient.
There is a steady increase in temperatures as the speed is constant at 1rpm. If we now reduce the
speed to 0 for the 2nd part of the transient, i.e.:
The transient below is predicted for the Equal drive, i.e. same as before:
It is seen that the heating is much increased over the 2nd part of the transient due to the larger loss
concentrated in the worst slots.
To try and confirm the assumptions made in putting together the model we can create alternative
models for the transients where in the zero speed period we set the [Stall Copper Loss Distribution]
to [Equal (Ia = Ib = Ic)] and just increase the power in all the slots to be equal to that of the worst
slot.
For the square wave drive the worst slot will have 3/2 times that of normal so the total copper loss is
increased from 120W to 180W. We also need to remove 60W from the back iron to ensure that the
total dissipation from the machine is equal to the total loss in the machine:
This is the same result as for the stall condition reported earlier.
For the transient simulation we set the losses as for the normal case:
Here we apply the full -60W from the stator tooth node during the stall period (not taking account of
the storage term as in the stall calculation). The resulting transient is very similar to the one reported
earlier:
Again the results compare well with those for the stall model for a sinewave drive.
Output Data:
Details of the parameters calculated in the analysis are given in Output Data [Miscellaneous - other]:
Fault Analysis
A very similar method to the stall analysis is used to analyse open circuit faults in BPM motors
under transient conditions (see Transient Fault Analysis)
When the motor is stalled (zero speed) there can be a situation that some of the slots have more
loss than others. The amount by which the loss varies in the slots depends on how the motor is
supplied. The above radio-button selector is used to choose the supply type:
Equal (Ia = Ib = Ic)
Sine (Ia/2 = Ib = Ic)
6-Step (Ia = Ib, Ic = 0)
The 1st connection type is not usually used in a three phase motor - unless all the phases are
connected in series in the lab and a dc supply is being used to heat up a stationery motor (can be
used for temperature rise testing). The sinewave and 6-step drive types are common in brushless
machines. In a 6-step driven machine two phases will have the same current and the other phase
zero current. The worst case condition in a sinewave driven motor is where the stall occurs when one
phase carried the peak current and the other two halve this value - in this case the phase with the
larger current will have 4 times the copper loss of the other two (but only twice the copper loss of a
phase where there is slow rotation with the rms same current as for the stall - all phases have the
same copper loss in this case)
The above interface sets up the Loss Distribution editor with the required loss values for each slot. It
is also possible to model the stall copper loss distribution for machines with other number of phases
using the Loss Distribution editor directly.
Loss balance is used to model the worst slot in the machine at stall. This is the one displayed in the
steady state schematic in this case. The theory used to calculate the worst slot is described in Stall
Analysis.
The data is set in Losses [Input Data Editor] and Stall Analysis
The Stator Copper Losses are divided between the active and end windings based on the active and
the total endwinding lengths as shown below:
split evenly between the front and rear endings (as shown above)
or
For details of the endwinding loss split settings see Losses [Settings].
The Loss Distribution editor allows the losses to by distributed unevenly between slots. This is
useful when there are some phases that are not energized for example in a multi winding fault
tolerant machine where one set of windings may not be energized.
Motor-CAD models the worst slot by calculating the losses in the slot from the following equation:
Copper Losses in worst case slot = (total copper losses) / (number of slots) * (Total number of
phases / Number of phases energized)
The extra copper losses calculated are then removed from the back iron to take account of the
adjacent slot not being energized and not having any copper losses.
Example
The example below will show how to model a machine with 6 phases but only 3 phases that are
energized.
We will first model the machine with all slots equally energised.
The distribution is set up as shown below with 1pu losses that are distributed between all 6 slots in
the phase belt.
The steady state temperatures for a machine with even distribution of losses is shown below.
The uneven distribution is set up as shown below with 1pu losses that are distributed between 3 of
the 6 slots in the phase belt.
So the energised slots now have x2 the losses. The uneven distribution option is then set as shown
below:
To compensate for the uneven distribution of losses Motor-CAD has put 2 x 120W of copper
losses in the slot and then to compensate for this additional copper loss 120W of extra losses will
be removed directly from the stator back iron (so instead of having 30W added it has now 90W
removed) as shown below:
The increase in the winding temperature of the worst case slot can also be seen above with the
average winding temperature rising from 109.1C to 116.4C.
This method of calculating the temperature of the worst case slot can be verified using the 2D FEA
calibration model as shown below. Can see the good agreement between the Motor-CAD model and
the FE results with the uneven copper loss distribution.
The loss distribution model is used when there is an even distribution losses between slots. This
may be the case if the machine has multiple windings, in with uneven copper loss distribution, fault
cases or in stall conditions.
The Use in FE checkbox specifies which copper loss distribution to use in the FE model of the slot.
See Uneven Copper Loss Distribution for its use.
The methodology is based upon loss parameters derived from 2 discrete time step finite element
analyses on the machine when the winding is:
1. Open Circuit
2. Has an ideal balanced short circuit imposed on its terminals
where
The Brushless AC motor loss model is enabled by check the Automatic Loss Calculation
checkbox shown below.
The Iron Loss Separation options allow the user to specify how the calculated iron losses are
separated between the back iron, tooth, tooth tip and rotor back iron.
The Loss model parameters are input using the Brushless AC motor Loss model editor shown
below:
The calculated copper, iron and magnet losses can be seen in the output data sheet as shown
below:
[1] P H Mellor, R Wrobel, D Holliday, 'A computationally efficient iron loss model for brushless AC
machines that caters for rated flux and field weakened operation', IEMDC 2009, Florida, May 2009
Table used to input material characteristic data for the thermal model and to calculate/display the
motor component weights.
N.B. - Parameters shown in Red are input, those in Black are output.
Various partially blanked out rows are used to display sub-total weights (e.g. copper, impregnation,
etc.). The total motor weight is also given on the last row.
Help on the properties for materials commonly used in electric machines is given in the online help.
A direct link to this page can be gained by pressing the [Material Help] button.
Material Database:
The user can input data for the materials using this editor or they can select a material from the
Material Properties Database. The material for a component is selected by using the drop down box
in the "Material from Database" column. When this is selected the values are set from the values in
the Solid Material database. However steel materials need to be used from the material database to
ensure that they have a valid BH curve.
If the values in the materials database are changed or the database is changed then values used for
the model are not automatically updated. The update is done by clicking on the "Update materials
from the Database" button:
Table used to input material characteristic data for the thermal model and to calculate/display the
motor component weights.
N.B. - Parameters shown in Red are input, those in Black are output.
Various partially blanked out rows are used to display sub-total weights (e.g. copper, impregnation,
etc.). The total motor weight is also given on the last row.
Help on the properties for materials commonly used in electric machines is given in the online help.
A direct link to this page can be gained by pressing the [Material Help] button.
Material Database:
The user can input data for the materials using this editor or they can select a material from the
Material Properties Database. The material for a component is selected by using the drop down box
in the "Material from Database" column. When this is selected the Thermal Conductivity, Specific
Heat and Density values are set from the values in the Solid Material database.
If the values of Thermal Conductivity, Specific Heat and Density are modified after the settings have
been taken from the database then the material name will be removed from the "Material from
Database" entry.
If the values in the materials database are changed or the database is changed then values used for
the model are not automatically updated. The update is done by clicking on the "Update materials
from the Database" button:
If the material is not available in the current database then the material will be added to the materials
database with the comment in the notes field that it has been autogenerated.
The following table contains data for material commonly used in electric machines. This data can be
input in the Materials & Weights [Input Data Editor]. Alternatively a material can be selected from
the Material Properties Database.
The properties given below are only intended as a rough guide to a materials thermal properties. It
will be found that many of the materials given below will have slightly different numerical values when
viewed in different sources of information.
Metals
Non-Metal Solids
Bakelite 0.19 - -
Epoxy 0.207 - -
Kapton 0.12 to 1.6 1090 1420
Mica 0.71 2900 -
Mylar - - 0.19 - -
Nomex - - 1400
Nylon 0.242 1700 1100
Paper 0.12-0.18 1200-1340 700-930
Plastic - ABS 0.25 1260 1014
Rubber (Soft) 0.13 2010 1100
Rubber (Hard) 0.16 - 1190
Teflon 0.22 1040 2200
Liquids/Gases
Skydrol LD-4 Hydraulic 0.132 (38C 1830 (38C) 990 (24C) 11.42e-6 (38C)
Fluid ) 0.125 (93C) 1976 (93C) 3.93e-6 (99C)
0.116 (149C) 2123 (149C) 1185e-6 (-54C)
Skydrol 500B-4 Hydraulic 0.132 (38C) 1753 (38C) 1045 (24C) 11.51e-6 (38C)
Fluid 0.125 (93C) 1896 (93C) 3.78e-6 (99C)
0.064 (149C) 2039 (149C) 2765e-6 (-54C)
Silicon Iron is a material commonly used in electric machines but which the designer may not have
thermal conductivity for - this is often not given in the data sheets for the lamination steel used. The
following graph is supplied to try to help the user set such data. It is noted that the low loss material
has a worse thermal conductivity than cheaper low silicon grades.
[1] Banks, P.J.: Thermal conductivity of sheet steel laminations, AEI Manchester Report No.TP/
R/1,188, 1961
[2] Roberts, T J : Determination of the thermal constants of heat flow equations of electrical
machines, Proc IME, 1969-70, Vol 184, Pt 3E, pp 84-92
[3] Schumichen, M: Longitudinal and transverse thermal conductivity of laminated transformer
sheets, Elektrie, 20(1966)12. In German.
[4] Gotter, G: Heating and cooling of electrical machines, 1954, Springer-Verlag, Berlin - In
German
The interface gaps can be defined either by using the drop down boxes in the details column to
specify the type of interface and use reference interface gap sizes or by entering an equivalent gap
directly in the gap column.
Notes:
A column of the table editor is included so that the user can add notes relating to the
individual interface gaps.
A column that gives an indication of the interface resistance in units of m2.C/W is output -
this can be useful for more experienced thermal analysis users.
Interface resistances are modelled in Motor-CAD as an effective airgap. The thermal resistance of
this airgap is calculated using the equation
R = Gap /kA where:
Gap = user input gap width
k = thermal conductivity of air at 20C
A = surface area
Due to the typical small lengths associated with interface gaps they are not drawn on the cross-
section drawing.
The accuracy of a motor thermal performance prediction is critically dependant upon the estimation
of the many thermal contact resistances within the machine (e.g. stator lamination to housing, slot-
liner to lamination, etc). The correct estimation of contact resistances can be crucial, especially in
heavily loaded machines, i.e. a large loss through even a small thermal resistance leads to a large
temperature difference (as a large current through a electrical resistance leads to a large voltage
drop).
The easiest way to deal with thermal contact resistances in a design algorithm is to base the
thermal resistance on an average interference airgap. This is the method used in Motor-CAD, the
user can set interface gaps for all contacting components. In order to assist the user to set realistic
values for these interface gaps and also to set realistic values for the default values some research
has been done into the subject.
The surface contact is a complex function of material hardness, interface pressure, smoothness of
the surfaces and air pressure.
Books on general heat transfer analysis such as Holman [1] and Mills [2] give typical values of
thermal resistance [m2C/W] and thermal conductance [W/m2/C] that can be expected between
various materials for various rms surface roughness. The definition of rms surface roughness is the
root mean square of the deviations of a surface from the reference plane [3] - typical values given by
Janna [3] are 0.0001mm for a mirror finish and 0.023mm for a rough finish. We can convert the data
given by Holman and Mills to equivalent airgap by using a thermal conductivity for air of 0.026 W/m/C
(m2C/W x W/m/C & 1/[W/m2/C] x W/m/C). The results are given in Tables I & II.
TABLE I
Contact Resistance and Interface Gap from Holman [1]
TABLE II
Inter facial Conductance and Interface Gap from Mills [2]
(Moderate Pressure & Usual Finish)
The effects of material hardness and surface roughness are clearly seen, the softer and smoother
materials clearly having the smallest effective gaps. For the typical material interfaces found in
electrical machines we find values of interface gap for aluminium-aluminium in the range 0.0005mm
to 0.0025mm, stainless-stainless of 0.007mm to 0.015mm, aluminium-stainless of 0.006mm to
0.009mm and aluminium-iron of 0.0006mm to 0.006mm . We can use these values as first
estimates of interface gaps in electrical machine analysis.
The problem in using interface gap data such as that given in Tables I & II is that it does not account
for all the complexities associated with electrical machines. Two gaps that we must look at in more
details are the lamination to housing interference gap and the gap between slot-liner and lamination.
The gap between lamination and housing is a function of how well the rough laminated outer surface
of the stator is prepared before the housing is fitted. A further complexity is that often there are
other features stamped into the outer surface of the lamination for the stacking operation which give
an increased effective gap. Also, if that housing is made from aluminium then due to its relative
softness compared to a cast iron frame this should lead to a reduced effective gap. The problem is
that the difference in thermal expansion rates between it and the stator lamination give rise to an
increasing effective gap at high temperatures - often eliminating the softness advantage.
Complexities in the slot-liner to lamination interface are that the liner material is quite pliable, the
slot surface is laminated, the gap may be filled or partially filled with impregnation and that a large
slot-fill will tend to push the liner towards the lamination.
A test program has been performed at Politecnico di Torino [4] to help identify typical gaps in
different sizes of machine and to relate the gaps to manufacturing and material differences between
machines. Table III shows typical values of lamination to housing interference gap found in a range
of machines. These have been measured by passing a know loss through the interface and
measuring the temperature on each side. The average of the values in the table is 0.037mm, but it
can vary significantly depending upon the manufacturing process and materials used. The gaps
found are typically around 10 times greater than those found in Tables I & II.
The Motor-CAD user can easily examine the importance of interface gaps in their machine be
varying them between expected upper and lower limits. The designer will get more accurate results
if they can perform calibration based on testing of motors that are constructed using materials and
manufacturing processes to be used in their new designs.
TABLE III
Examples Interface Gaps Found Between Housing and Lamination
References:
Interface gaps are not drawn in the cross-section editor as they are usually very small.
Stator Lam - Interface gap between stator lamination stack and housing.
Housing
Banding - Rotor Interface gap between magnet retainment banding and magnet.
Magnet - Rotor Interface gap between magnet and rotor lamination stack.
Lam
Rotor Lam - Interface gap between rotor lamination stack and shaft:
Shaft
Sleeve - Stator Interface gap between stator bore sleeve and stator lamination:
Endcap - Plate Interface gap between endcap and flange mounted cooling plate. See also Plate
Geometry.
Bearing - Shaft Interface gap between bearing and shaft [front of motor]:
[F] The Bearing-Shaft is not indicated on the above diagram - it is very similar to
Bearing - Endcap [F] but between the bearing inner race and shaft rather than outer
race and endcap.
Typically this interface gap is made larger than the physical gap in order to account
for the added thermal resistance presented by the bearings themselves. See
Bearing Model
Bearing - Shaft Interface gap between bearing and shaft [rear of motor]:
[R] The Bearing-Shaft is not indicated on the above diagram - it is very similar to
Bearing - Endcap [F] but between the bearing inner race and shaft rather than outer
race and endcap.
Typically this interface gap is made larger than the physical gap in order to account
for the added thermal resistance presented by the bearings themselves. See
Bearing Model
Bearing Effective This is the effective interface gap used to model the temperature drop across the
Gap [F] bearings at the front of the machine. See Bearing Model
Bearing Effective This is the effective interface gap used to model the temperature drop across the
Gap [R] bearings at the rear of the machine. See Bearing Model
Shaft - Enc Interface gap between motor shaft and encoder shaft:
Shaft
See Bearing Model for more information on the bearing model adopted in Motor-CAD.
Housing - Interface gap between active section of motor and housing overhang [front of
OHang [F] motor].
Used to model frameless motors that have Housing Dia = Stator Lam Dia and
negative value of Housing Step [F].
Housing - Interface gap between active section of motor and housing overhang [rear of motor].
OHang [R] Used to model frameless motors that have Housing Dia = Stator Lam Dia and
negative value of Housing Step [R].
In many cases this interface gap is zero - where the active and overhang sections of the winding are
made from the same piece of material.
Housing - Feet Interface gap between housing and feet. See Foot Mounted Base.
Feet - Base Interface gap between feet and foot mounted base. See Foot Mounted Base.
Cage - Rotor Interface gap between induction motor rotor bars/cage and rotor lamination.
Lam It can have a significant influence on the maximum rotor temperature in induction
machines. It is a function of whether a copper bars or a die-cast cage is used:
copper bars: points of contact with copper bars are relatively fewer than with a die
cast cage
die cast rotor: aluminium forced into the slots under pressure and fills the
corrugated type surface on the inside of the slots - improved heat transfer (5
times less thermal resistance [1] than if bars used) can cause short circuits of
laminations and increase in iron losses
[1] Rosenberry Jr, G.M.: The transient stalled temperature rise of cast aluminium
squirrel rotors for induction motors, Trans AIEE, 1955, Vol 74, Pt III, pp 819-824
Stator Lam - Interface gap between stator lamination and axle in an Outer Rotor BPM Machine
Axle
Comm - Brush Interface gap between commutator and brush in a PMDC Machine
Comm - Shaft Interface gap between commutator and shaft in a PMDC Machine
Housing - Cover Interface gap between housing and insulated cover in an Outer Rotor BPM Machine.
Only available if Rotor Mounting = Insulated
Work at Politecnico di Torino [1] has gathered more data on typical values for this equivalent gap.
The motors shown in Fig.1 have been tested. The procedure adopted is as follows. A Motor-CAD
thermal model of the motor under test was calibrated using the temperatures measured during a DC
supply test, i.e. a stationery motor was supplied with a dc current and the various unknown
parameters such as the impregnation goodness and stator lamination to housing interface gap were
varied until measured temperatures throughout the machine match measured ones. In this test the
rotor was at zero speed and only stator copper losses are present.
The second step was to perform a classical locked rotor test using a three phase sinusoidal supply.
In this condition the mechanical losses are again zero (rotor speed is zero) but both stator and rotor
losses now exist. The rotor losses can be computed as the difference between the input power
minus the stator copper losses as given in the following equation:
A special end shield as shown in Fig. 2 has been used to measure the temperature of the inner and
outer bearing races. The dc calibrated thermal model has shown very good agreement between the
measured and predicted temperatures for the windings, stator lamination and housing during the
locked rotor thermal simulation. Starting from this thermal model, the front and rear equivalent
bearing interface gaps have been changed until the temperature across Motor-CAD bearing
resistance is equal to the measured one. The required bearing interface gaps are reported in Table I.
The table shows that bearing equivalent interface gap equal to around 0.3mm gives good results. The
only real problem with such data is that the rotor and the bearing at stall. Politecnico di Torino are
now working on suitable tests for defining the bearing thermal behaviour with the rotor in running
condition.
If the predicted bearing temperature contact resistance difference is to be predicted with a good
accuracy, it is evident that an accurate model of the motor end shields is also required. It is possible
to predict both radial and axial thermal resistance values for the end shields if we know their effective
lengths and cross-sectional areas, Rt = k/(Al). However, this process is sometimes complicated by
the fact that the end shields can have complex shapes. We also have to take account of the
interference fit to the housing. The radial and axial endcap thermal resistances can be ignored or
included in the circuit (resistances based on simplified dimensions in the Motor-CAD radial and axial
cross section editors) by setting the Include Rt[Endcap] Axial and Include Rt[Endcap] Radial
checkboxes in Model Settings Editor].
TABLE I
In the testing of temperature rise of bearings on a range of induction motors it has been seen that
there is a tendency for a rapid but small increase in bearings outer race temperature when the
machine is turned off. One thought is that this could be due to the bearing thermal resistance at
standstill being smaller than when rotating so making the outer race closer to the hotter shaft. More
work is underway to investigate this effect in more detail. The effect can be modelled in Motor-CAD
by stopping the thermal transient when the motor is stopped and setting to use Previous Transient
Temperatures (see Duty Cycle Settings) . The bearing interface gap can then be reduced and the
transient continued. An ActiveX script can also be written to automate this calculation.
Bearing Model
A simple solution is to consider the bearing as an equivalent interface gap - this method is used in
Motor-CAD. The effective temperature drop across the bearings is simulated using the following
interface gaps:
Bearing - Endcap [F] - Interface gap between bearing and endcap [front of motor]. Typically
this interface gap is made larger than the physical gap to account for the added thermal
resistance presented by the bearings themselves.
Bearing - Endcap [R] - Interface gap between bearing and endcap [rear of motor]. Typically
this interface gap is made larger than the physical gap to account for the added thermal
resistance presented by the bearings themselves.
Bearing - Shaft [F] - Interface gap between bearing and shaft [front of motor]. Typically this
interface gap is made larger than the physical gap to account for the added thermal
resistance presented by the bearings themselves.
Bearing - Shaft [R] - Interface gap between bearing and shaft [rear of motor]. Typically this
interface gap is made larger than the physical gap to account for the added thermal
resistance presented by the bearings themselves.
Bearing Effective Gap [F] - This is the effective interface gap used to model the temperature
drop across the bearings at the front of the machine.
Bearing Effective Gap [R] - This is the effective interface gap used to model the temperature
drop across the bearings at the rear of the machine.
The default values for Bearing Effective Gap [F] and Bearing Effective Gap [R] in Interface
Gaps [Input Data Editor] are currently set at 0.4mm to account for the temperature drop
across the bearings. This default value used to be 1mm, however recent testing and
calibration of various motors it is thought that this value may be too large. A value in the
range 0.3mm to 0.5mm may be more appropriate.
In Motor-CAD there is a bearing node. Half of the equivalent bearing resistance calculated from the
bearing interface gaps is put on each side of this bearing node.
The default Bearing Loss location is Split between inner and outer race.
In Motor-CAD with the option to have the Bearing Losses Split between inner and outer race (See
losses Settings) then the bearing losses are split evenly between the inner and outer race, as show
below. This model predicts an average bearing temperature rather than just the temperatures on
the inner and outer race.
In Motor-CAD v5.3 an additional optional model for the bearings was added that has all the bearing
loss applied at the bearing central node as shown below. This is enabled in Motor-CAD using the
option to have the Bearing Losses on the Bearing Central Node (See losses Settings). Note: This
model tends to over-predict the bearing temperatures if the old calibrated effective interface gap
values are used.
It is recommended that the user keeps with the older model with the losses Split between inner and
outer race, especially if they are not going to carry out some independent bearing temperature
calibration as the default calibration figure for effective bearing gap of 0.4mm is only valid for the older
model. The new model tends to over-predict the bearing temperatures if the old calibrated effective
interface gap values are used.
[1] D. Staton, A. Boglietti, A. Cavagnino, "Solving the More Difficult Aspects of Electric Motor
Thermal Analysis", IEMDC 2003 Conference Proc. 1-4 June 2003, Madison Wisconsin, USA.
8.7 Radiation
8.7.1 Radiation Data [Input Data Editor]
Radiation can be a very important mode of heat transfer. In a Natural Convection (TENV) machine
the dissipation due to radiation can be equal or exceed that of convection. In Blown Over machines
that have shaft mounted fans and operate at low speeds the forced convection cooling approaches
that due to natural convection and so radiation is again important.
Internal radiation can also be important in space applications where there may be no internal
convection if the machine has a vacuum inside.
If the Include Radiation in Lump Circuit Model checkbox is checked then external radiation from
the housing is included in the model.
The external radiation heat transfer coefficient h[r] is calculated using the formula:
h[r] = 5.67e-8 * Emissivity * View Factor * [(Tnode + 273)**4 - (Ambient Temperature [Radiation] +
273)**4] / (Tnode - Ambient Temperature [Radiation])
Note that there are separate Ambient Temperature [Radiation] and Ambient Temperature
[Convection] values for radiation and convection respectively.
Check-box to determine whether to Include Internal Radiation in the thermal calculation. See Internal
Radiation Heat Transfer for full details.
The Delta Temperatures above are used in calculation of parameters in internal and external radiation
parameters. This is a tool to allow the user to see the radiation parameters and are not used for the
actual model calculations.
When calculating the actual steady-state or transient performance the actual calculated component
temperatures are used in the calculation of the radiation heat transfer data.
The Radiation Table Editor is used to input/calculate/display the motor radiation data. It has the
following columns:
Component - description of motor component surface that can dissipate heat by radiation
Emissivity - Emissivity of motor component surface
View Factor - View Factor to Ambient of motor component surface. Note: this is normally
not editable, to edit it see Editor Settings
hr @dT=xxxC [W/m2/C] - Radiation Heat Transfer Coefficient calculated at External
Radiation dT or Internal Radiation dT
Area [cm2] - area of motor component surface
Rt @dT=xxxC [C/W] - Radiation Thermal Resistance to Ambient calculated at External
Radiation dT or Internal Radiation dT
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See Units.
In the original version of Motor-CAD internal radiation was neglected. Only external radiation from the
housing and end-shields was accounted for. Internal radiation is not usually significant in comparison
with the convection except when there is a vacuum.
The option to include airgap radiation has since been added. Internal radiation is calculated between
the endwindings and the inside of the housing and end-shields and also between the axial ends of
the rotor and the inside of the end-shields.
The theory used to calculate the heat transfer due to radiation across the airgap is given in the
Appendix below:
Internal radiation is included in the model if the Include Internal Radiation check-box is checked:
When checked a rough indication of the effective radiation heat transfer coefficient for each side of
the airgap and the airgap radiation heat thermal resistance will be indicated. This is only a rough
indication as the heat transfer coefficient is a function of the temperature difference across the
airgap. This temperature difference is only calculated when a solution is carried out so in this table
the temperature difference is based on the value input for Internal Radiation dT:
For internal radiation the user can vary the emissivity for the inner and outer surfaces together with
the view factor in the table editor:
When internal radiation is included the separate convection and radiation plus the total thermal
resistance is shown in the schematic:
The internal radiation paths from the end windings to housing and end caps and the paths from the
rotor surface to the end caps are not shown in the schematic view as makes the schematic rather
complex.
They can be viewed if required in the circuit editor view:
The Radiation Exchange Factors (inner surface to outer surface) together with the radiation heat
transfer coefficient (inner surface to outer surface) and inner and outer surface areas are given:
The convection, radiation and total (convection and radiation in parallel) airgap thermal resistance is
also shown:
Appendix A
A1F12 = A2F21
We can calculate and effective radiation heat transfer coefficient for the two surfaces:
The values of h12 and h21 will be slightly different due to the difference in F12 and F21. However the
radiation thermal resistance between the two surfaces will be independent of if h12 or h21 as the
surface areas will also be different:
The View Factor of various motor component surfaces can be set within the Radiation Data [Input
Data Editor].
h[r] = 5.67e-8 * Emissivity * View Factor * [(Tnode + 273)**4 - (Ambient Temperature [Radiation] +
273)**4] / (Tnode - Ambient Temperature [Radiation])
The View Factor (also called configuration factor, shape factor or geometry factor) is a geometry
related value of a system in which two surfaces exchange energy by radiation. As radiation waves
travel in straight lines - if one surface cannot see another there is no direct radiation from the first to
the second. The View Factor gives an indication of the fraction of radiative energy leaving one surface
(i.e. motor component surface) that strikes the other surface directly (in this case the ambient
surroundings).
For very complex structures, such as a finned housing, the calculation of View Factor is very
difficult. It has been found that a simplification used in Motor-CAD does not lead to more than a
marginal error in virtually all cases. The simplification used is that all surfaces that have a clear view
of the ambient (i.e. the fin tips and fin base) have their View Factor set to 1. All surfaces that do not
have a clear view of the ambient (i.e. fins sides - these tend to radiate to the opposite fin surface)
have their View Factor set to 0. It is found that the small proportion of the fin side radiation that
would have been absorbed by the ambient is offset by the over estimation of the fin base radiation
(neglect of radiation from the base that is absorbed by the fin sides). The default values of View
Factor for the fins need not be altered except under special circumstances. The only exceptions
where much more radiation can be encountered are for radical designs such as that shown below -
for such designs the user needs to set a valid value for View Factor themselves.
8.7.4 Emissivity
The Emissivity of various motor component surfaces can be set within the Radiation Data [Input Data
Editor].
It is used in the calculation of external radiation heat transfer coefficient h[r] using the formula:
h[r] = 5.67e-8 * Emissivity * View Factor * [(Tnode + 273)**4 - (Ambient Temperature [Radiation] +
273)**4] / (Tnode - Ambient Temperature [Radiation])
It is used in the calculation of internal radiation heat transfer according to the formulation given in
Internal Radiation Heat Transfer
It is a myth that black paint has an emissivity approaching 1, while white paint has an emissivity
approaching 0. Some white paints can have very high emissivities - it all depends upon the materials
used in the paint and the surface finish.
Motor-CAD automatically calculates the convection from all external surfaces of the motor using
proven convection heat transfer correlation formulations (Natural Convection Correlation and Forced
Convection Correlation correlations). The formulation chosen for the particular surface is indicated in
the particular editor used (Natural Convection [Input Data Editor], Forced Convection [Input Data
Editor], etc). Full details of the formulations used are given in the online help. Each correlation if for a
separate geometric shape, e.g. horizontal or vertical cylinder, vertical plate, semi-open channel,
upper or lower sides of a horizontal plate, etc. Basically the authors of the various correlations have
made convection dissipation tests on various sizes of the particular geometric shape and fitted a
curve fit to their data. The curve fit is based on the fluid properties and the dimensions of the
geometric shape (the characteristic length). A process of Dimensional Analysis is used to develop
such formulations.
Dimensional Analysis:
In the technical literature the process of dimensional analysis usually followed to obtaining a
functional relationship between the physical properties and the fluid parameters. The use of
dimensional analysis makes the resulting correlation applicable to all such shapes and with any fluid
material - with the constraint that the flow conditions are similar to those of the tests, i.e. laminar or
turbulent.
A set of dimensionless numbers are used in the process:
Re = vL /
Gr = g ^2 L^3 / ^2
Pr = cp /k
Nu = h L / k
where
Natural Convection:
Nu = a (Gr Pr)^b
Forced Convection:
Nu = a (Re)^b (Pr)^c
There is generally a transition from laminar to turbulent flow for geometries having internal flow when
Re » 2300, with the flow becoming fully turbulent when Re > 5 x 104.
There is generally a transition from laminar to turbulent flow for geometries having external flow when
Re » 5 x 105.
Motor-CAD uses the best available correlation for the given surface. For some surfaces a composite
correlation is used, i.e. horizontal square housing = average of 2 x vertical plates + 1 x horizontal
upper surface + 1 x horizontal lower surface. The user can artificially increase or decrease the
amount of convection heat transfer from an individual surface using the h[adjust] parameter for the
particular surface - for instance this may be carried out if the user thinks that there is more
convection cooling than predicted due to drafts, etc. However, in the majority of cases the user will
just use the default calculation.
Full details of the correlation used in Motor-CAD are shown in Natural Convection Correlations and
Forced Convection Correlations.
The user can also completely ignore the default calculation for convection for a particular surface and
supply their own value of heat transfer coefficient, In order to do this the Input h check-box should be
ticked and a heat transfer value input in the h[input] column.
Altitude:
Altitude has a significant effect on convection cooling - thus the edit box is provided so the user
can select the altitude used for the calculation. More details regarding the altitude calculation used
in Motor-CAD is given in the US Standard Atmosphere.
Once heat transfer coefficient, h [W/m2/C] is calculated we can calculate the convection heat
transfer [W] for a given surface area, A [m2], and temperature [C]. The thermal resistance, Rt [C/W],
between the particular cooling surface and ambient due to natural convection is calculated using the
formula:
Rt = 1/(h.A)
This is in parallel with a radiation thermal resistance which is calculated from the Radiation Heat
Transfer - that is if the internal Radiation in include in the thermal calculation.
Units
See Natural Convection Data [Input Data Editor] and Forced Convection Data [Input Data Editor].
[1] Incropera, F.P & DeWitt, D.P.: Introduction to Heat Transfer, Wiley, 1990.
Motor-CAD automatically calculates the convection from all external surfaces of the motor using
proven convection heat transfer correlation formulations seeConvection Heat Transfer data for more
details.
The Delta Temperature is used in calculation of natural convection parameters. This is a tool to allow
the user to see the natural convection parameters and is not used for the actual model calculations.
When calculating the actual steady-state or transient performance the actual calculated component
temperatures are used in the calculation of the natural convection heat transfer data.
Altitude
This is the altitude at which the machine is operating at. The air properties are taken for this altitude
(see US Standard Atmosphere).
The Natural Convection Table Editor is used to input/calculate/display the motor natural convection
data. It has the following columns:
Component - description of motor component surface that can dissipate heat by convection
Input h? - Check-box for choosing to input or calculate hnc for the particular motor
component surface
Convection Correlation - An indication of the Natural Convection Correlation formulation
used to calculate hnc of the particular motor component surface
h[input] or h[adjust] [W/m2/C or dimensionless] - Depending upon the check-box in
column Input h?, the natural convection heat transfer coefficient (hnc) can be input directly
or an adjustment multiplier input - this being used to modify the calculated value of hnc
hnc @dT=xxxC [W/m2/C] - Natural Convection Heat Transfer Coefficient (hnc) calculated
at Convection Table dT.
Area [cm2] - area of motor component surface
Rt @dT=xxxC [C/W] - Natural Convection Thermal Resistance to Ambient calculated at
Convection Table dT
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See Units.
The following correlation formulae have been used in the calculation of the natural convection heat
transfer coefficient (hnc) from the various dissipating surfaces of the motor:
Horizontal Cylinder
Vertical Cylinder
Vertical Flat Plate
Horiz Flat Plate [Upper Face]
Horiz Flat Plate [Lower Face]
Horiz Flat Plate [Upper & Lower]
Horiz Square Tube
Horiz Servo Housing
Vertical Fin Channel
Horizontal Fin Channel
Horiz Fins/Flat Plate
Horiz Fins/Flat Plate & Cylinder
Horiz Fins/Flat Plate & Tube
Horiz & Vert Fin Channel Average
Horiz/Vert Fins & Horiz Cylinder
An indication of the formulation used is given in the Natural Convection Data [Input Data Editor] Table
Editor:
Simonson [1] gives a formulation for average Nusselt number of a cylinder of diameter d:
Nud = 0.525(Grd.Pr)**0.25
Nud = 0.129(Grd.Pr)**0.33
The fluid properties are evaluated at the average of the surface and bulk fluid temperatures, i.e. mean
film temperature.
Air Property Variation with Temperature can be modelled within the program see Air Properties.
hnc = Nud * k / d
k = Air Conductivity
d = Characteristic Length
[1] Simonson, J.R.: Engineering Heat Transfer, 2nd Edition, Macmillan, 1998
Simonson [1] gives a formulation for average Nusselt number of a vertical cylinder of height L:
NuL = 0.59(GrL.Pr)**0.25
NuL = 0.129(GrL.Pr)**0.33
The fluid properties are evaluated at the average of the surface and bulk fluid temperatures, i.e. mean
film temperature.
Air Property Variation with Temperature can be modelled within the program see Air Properties.
hnc = NuL * k / L
k = Air Conductivity
L = Characteristic Length
[1] Simonson, J.R.: Engineering Heat Transfer, 2nd Edition, Macmillan, 1998
Simonson [1] gives a formulation for average Nusselt number of a vertical flat plate of height L:
NuL = 0.59(GrL.Pr)**0.25
NuL = 0.129(GrL.Pr)**0.33
The fluid properties are evaluated at the average of the surface and bulk fluid temperatures, i.e. mean
film temperature.
Air Property Variation with Temperature can be modelled within the program see Air Properties.
hnc = NuL * k / L
k = Air Conductivity
L = Characteristic Length
[1] Simonson, J.R.: Engineering Heat Transfer, 2nd Edition, Macmillan, 1998
Simonson [1] gives formulations for average Nusselt number of a horizontal flat plate of mean length
of side L:
NuL = 0.54(GrL.Pr)**0.25
NuL = 0.14(GrL.Pr)**0.33
NuL = 0.25(GrL.Pr)**0.25
The fluid properties are evaluated at the average of the surface and bulk fluid temperatures, i.e. mean
film temperature.
Air Property Variation with Temperature can be modelled within the program see Air Properties.
hnc = NuL * k / L
k = Air Conductivity
L = Characteristic Length
[1] Simonson, J.R.: Engineering Heat Transfer, 2nd Edition, Macmillan, 1998
8.9.7 Horizontal Flat Plate - Upper & Lower [Natural Convection Correlation]
Area based average of Horizontal Flat Plate [Hot Surface Facing Up] & Horizontal Flat Plate [Hot
Surface Facing Down] - see Horizontal Flat Plate [Natural Convection Correlation].
Ref [1] gives a formulation for Nusselt number of u-shaped vertical channels (laminar flow):
Z = 24 * [1 - 0.483 . EXP(-0.17/a)]
/ [1+a/2]**3 . [1+ (1-EXP(-0.83a)) x (9.14 . SQRT(a) . EXP(-465 . fin_spacing) - 0.61)]**3
The fluid properties are evaluated at the wall temperature (except volumetric coefficient of expansion
which is evaluated at the mean fluid temperature).
Air Property Variation with Temperature can be modelled within the program see Air Properties.
hnc = Nur * k / r
k = Air Conductivity
r = Characteristic Length
[1] Van De Pol, D.W. & Tierney, J.K. : Free Convection Nusselt Number for Vertical U-Shaped
Channels, Trans. ASME, Nov. 1973.
Ref [1] gives a formulation for Nusselt number of u-shaped horizontal channels (laminar flow):
Air Property Variation with Temperature can be modelled within the program see Air Properties.
hnc = Nus * k / s
k = Air Conductivity
s = Characteristic Length
[1] Jones, C.D., Smith, L.F. : Optimum Arrangement of ToolFins on Horizontal Surfaces for Free-
Convection Heat Transfer, Trans. ASME, Feb 1970.
An average of the following is used to estimate hnc of fin section of horizontal axial finned motors:
(1) Horizontal Fins for the fins on the top and under-side of the motor
(2) Horizontal Flat Plate [Upper & Lower] x Min(Fin_Spacing,Fin_Depth)/Fin_Depth for fins on the
side of the motor
The second term in (2) is to limit dissipation area to depth down fin channel equal to fin spacing
(there will be little air circulation at the bottom of a deep channels on the sides of the motor).
An area based average of the following is used to estimate hnc of fin sides and base in horizontal
mounted axial finned motors:
Horiz Fins/Flat Plate
Horizontal Cylinder
The special form of average is used such that if the fins are deep compared to spacing then the
Horiz Fins/Flat Plate should predominate, but if the fins are not deep than the Horizontal Cylinder
should predominate. Also this form of average ensures that when the fins are virtually non-existent
then the correlation reverts back to that of a cylinder.
The graph below shows that in a Natural Convection (TENV) motor that a good prediction of the
natural convection can be achieved using such complex correlations. Here we see both calculated
and measured thermal resistance values between housing and ambient for the motors shown below,
the fan being at rest in this case. The calculated data is for Motor-CAD with default setting of all
parameters - all the user has done is to input the geometry for the motor and its foot mounting (the
cooling from the flange or foot mounting is important and is included in the analysis), the winding
details, the materials and the losses.
An average of the following is used to estimate hnc of fin sides & base in horizontal mounted axial
finned motors:
Horiz Fins/Flat Plate
Horizontal Square Tube
Very similar formulation to Horiz Fins/Flat Plate & Cylinder [Natural Convection Correlation] but for
the square tube rather than cylinder.
hnc = (1/k1 x h[[Horizontal Square Tube] + k1 x h[Horiz Fins/Flat Plate] / (k1 + 1/k1)
The special form of average is used such that if the fins are deep compared to spacing then the
Horiz Fins/Flat Plate should predominate, but if the fins are not deep than the Horizontal Cylinder
should predominate. Also this form of averag
e ensures that when the fins are virtually non-existent then the correlation reverts back to that of a
tube.
8.9.15 Horiz & Vert Fin Channel Average [Natural Convection Correlation]
An average of the following is used to estimate hnc of fin sides in horizontally mounted radial finned
motors:
Vertical Fins on sides of motor
Horizontal Fins on top and bottom of motor
An average of the following is used to estimate hnc of fin sides & base in horizontal mounted axial
finned motors:
Horiz & Vert Fin Channel Average
Horizontal Cylinder
hnc = (1/k1 x h[Horizontal Cylinder] + k1 x h[Horiz & Vert Fin Channel Average] / (k1 + 1/k1)
The special form of average is used such that if the fins are deep compared to spacing then the
Horiz Fins/Flat Plate should predominate, but if the fins are not deep than the Horizontal Cylinder
should predominate. Also this form of average ensures that when the fins are virtually non-existent
then the correlation reverts back to that of a cylinder.
A Blown Over machine has typically an external fan is used to blow a fluid (normally air) across the
outside of the machine. The dissipation from the outside of the motor is assumed to be a
combination of Mixed Convection (forced convection due to fan combined with natural convection),
radiation and conduction. The fluid can be air or any other fluid as set in the External Fluid Data.
This editor provides the setting of the Blown Over data for the forced convection.
Motor-CAD automatically calculates the forced convection from all external surfaces of the motor
using proven convection heat transfer correlation formulations see Convection Heat Transfer data for
more details.
The Blown Over editor is visible when the Blown Over cooling option is selected. (see Cooling
Options)
Flow options for specifng the flow direction and velocity (or flow rate if have enclosed channels) of
flow leaving the blower.
Fluid Flow for specifying how the flow reduces along the length of the machine due to leakage or
blockages.
Heat transfer that shows the calculated heat transfer coefficients.
Flow Direction - whether the flow comes from the rear to the front of the machine or from the front
to the rear of the machine. For radial finned machines there is also the option to have tangential
flow along the fin channels.
Reference Shaft Speed - this is shaft speed for which the Reference Flow Velocity (or
Reference Flow Rate) is specified by the user. The flow velocity (or flow rate) will be scaled linearly
with the machine speed.
Cowling grill definition - at present this is used not used in any Motor-CAD calculations. It is
used for visualization and also for exporting the 3D geometry to other 3D packages.
See also Blown Over [Input Data Editor],, Blown Over Fluid Flow and Blown Over Heat Transfer
This interface is used to define how the blown over flow changes as the fluid passes along the
outside of the machine.
The Graph Selection option allows the graph to show Flow, Velocity or Flow Rate values along the
axial length of the housing.
The View Typical Curves option allows the user to see the curves from different results so that
their flow values can be compared and adjusted if necessary to match these results. The example
below shows how the flow in the model is matched to the flow of a 15kW machine. The red dashed
line will be used in the Motor-CAD model.
See also Blown Over [Input Data Editor], Blown Over Flow Options and Blown Over Heat Transfer
Editor for setting the Convection Heat Transfer data when forced convection is used.
From the Air Velocity Motor-CAD calculates the forced convection heat transfer coefficients (hfc)
using the appropriate Forced Convection Correlation
Shaft Speed [rpm] -This is the speed at which the machine is operating at. See Shaft Speed
[RPM].
Altitude - This is the altitude at which the machine is operating at. The air properties are taken for
this altitude (see US Standard Atmosphere).
Turbulence Factor (Channel) - This is a Direct multiplier for Convection Heat Transfer coefficient
(hfc) calculated using the Fin Channel [Forced Convection] correlation to account for turbulence [1] -
tests indicate typical turbulence factors in the range 1.7 - 1.9 which seem independent of the flow
velocity [1]. This is only used for forced convection components within fin channels.
[1] Heiles, F. : Design and Arrangement of Cooling Fins, Elecktrotecknik und Maschinenbay, Vol.
69, No. 14, July 1952.
dT used in table below - The Delta Temperature is used in calculation of forced convection
parameters. This is a tool to allow the user to see the forced convection parameters and is not used
for the actual model calculations. Note: This temperature is not used for the actual model
calculations. When calculating the actual steady-state or transient performance the actual
calculated component temperatures are used in the calculation of the forced convection heat transfer
data.
Table
The Forced Convection Table Editor is used to input/calculate/display the motors forced convection
data. It has the following columns:
Component - description of motor component surface that can dissipate heat by convection
Input h? - Check-box for choosing to input or calculate hfc for the particular motor
component surface
Forced Convection Correlation - An indication of the Forced Convection Correlation
formulation used to calculate hfc of the particular motor component surface
h[input] or h[adjust] [W/m2/C or dimensionless] - Depending upon the check-box in
column Input h?, the forced convection heat transfer coefficient (hfc) can be input directly or
an adjustment multiplier input - this being used to modify the calculated value of hfc. See
also Blockage.
Air Velocity [pu] - Input estimate of local air velocity - in relation to reference velocity (
Default Velocity @Default Speed). This value must be set for each surface by the user to
take into account the flow over each outer surface of the machine. See Blockage and
Leakage and Air Velocity for details on how this is set.
Air Velocity [m/s] - Local air velocity. See also Leakage.
hnc @dT=xxxC [W/m2/C] - Natural Convection Heat Transfer Coefficient (hnc) calculated
at dT used in table below. See Natural Convection Data [Input Data Editor].
hfc @dT=xxxC [W/m2/C] - Forced Convection Heat Transfer Coefficient (hnc) calculated at
dT used in table below.
hmixed @dT=xxxC [W/m2/C] - Mixed Convection Heat Transfer Coefficient (hmixed)
calculated at dT used in table below. See Mixed Convection.
Rt @dT=xxxC [C/W] - Natural Convection Thermal Resistance to Ambient calculated at dT
used in table below
Surface Area - the surface area of the component
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See also Blown Over [Input Data Editor], Blown Over Flow Options and Blown Over Fluid Flow
See Units.
Typically in a Blown Over (TEFC) machine some of the fin channels on the outside of the machine
are blocked by bolt lugs and terminal boxes. Another deficiency of TEFC machines is that the air
leaks out of the open channels causing the local Air Velocity to be lower at the drive end than at the
non-drive end. The typical form of the reduction in velocity is shown below:
The prediction of the actual reduction in velocity is a complex function of many factors including the
fan, fin and cowling design and rotational speed. A more accurate model is formed if some
calibration is performed using testing and/or CFD. Below we see the typical accuracy that can be
expected with an un-calibrated model:
Here we have taken the default parameters in Motor-CAD and calculated the effective thermal
resistance between housing and ambient for the motors shown below. The open channel air leakage
data of DiGerlando [1] in the graph above is used as defaults in this case. It is seen that an accurate
estimate can be made if the user has a basic knowledge of the inlet air velocity or volume flow rate
to the fin channels. The fact that the larger machines have a higher air speed is confirmed by
comparing calculated and measured data in the graph above. Fin blockage is simply accounted for
by the user counting the total number of fin channels (N_Total) and the blocked channels
(N_Blocked). The factor used is then:
Kovalev [2,4] has performed testing on open and closed fin channel arrangements. He shows that
the reduction in heat transfer is only of the order of 10% in the open-channel arrangement, the inlet
velocity to both being the same. The small reduction compared with the larger reduction in velocity
along the channels is attributed to added turbulence in the middle and far end of the machine. The
closed fin channel requires a larger driving force in terms of a larger fan. Benerke [3,4] shows similar
results.
References:
[1] DiGerlando, A., Vistoili, I.: Thermal Networks of Induction Motors for Steady State and
Transient Operation Analysis. ICEM 1994, Paris.
[2] Kovalev, E.B.: Burkovski, A.N., Tokarenko, A.T.: Heat transfer in channels between frame-
ribbing of enclosed asynchronous motors, Elektrotechnika, 1965, No 11.
[3] Benecke, W: Temperature field and heat flow in case of small surface cooled three-phase
motors with squirrel cage rotor, ETZ(A), 1966, Vol 87, No 13.
[4] Pal, S.K.: Heat Transfer in Electrical Machines - A Critical Review, ERA Report No 71-76,
July 1971
See also Blown Over [Input Data Editor], Blown Over Fluid Flow and Blown Over Heat Transfer
The local air velocity is defined by the user using the Air Velocity [LEAKAGE] [pu] parameter in the
Forced Convection Data [Input Data Editor]) Table Editor, i.e.:
It is termed a LEAKAGE factor as it can be used to account for Leakage leakage from open fin
channels which are typical in Blown Over (TEFC) machines.
Depending upon the Constant Speed Fan parameter - the Local Air Velocity [m/s] is then calculated
using one of the following:
Local Air Velocity [m/s] = Air Velocity [pu] x Forced Convection Default Velocity
Local Air Velocity [m/s] = Air Velocity [pu] x Forced Convection Default Velocity x Shaft Speed
[RPM] / Forced Convection Default RPM
The Local Air Velocity [m/s] is given as output in the Table Editor for reference. This is then used to
calculate hfc using the relevant Forced Convection Correlation.
The graph below shows the variation in typical air velocities at the fan outlet for different machine
sizes with speed.
The graph below shows the typical form of velocity reduction with distance from the fan in axially
finned induction motors:
The velocity of air at points along the axial length of an open fin channel in a TEFC motor will be a
function of the Blockage and Leakage which can be defined in the Blown Over [Input Data Editor]
editor.
[1] DiGerlando, A., Vistoili, I.: Thermal Networks of Induction Motors for Steady State and Transient
Operation Analysis. ICEM 1994, Paris.
See also Blown Over [Input Data Editor], Blown Over Fluid Flow and Blown Over Heat Transfer
Fan Laws are a set of equations which relate the performance of geometrically similar fans[1]. Fans
are considered to be geometrically similar if all the dimensions of one fan are in a fixed ratio to the
corresponding dimensions of the other fan. It assumes that the efficiency of the fans are similar,
however in reality the fan efficiency increases with increasing size.
If the flow rate is known for a fan then fan scaling laws can be used to calculate the flow rates
achieved when resizing the fan diameter and varying the speed.
The graph below shows and example of the variation of flow over the machine with speed for a range
of fans.
See also Blown Over [Input Data Editor], Blown Over Fluid Flow and Blown Over Heat Transfer
[1]Woods Practical Guide to Fan Engineering, Woods of Colchester Ltd, 2nd Ed, June 1960.
Motor-CAD automatically calculates the convection from all external surfaces of the motor using
proven convection heat transfer correlation formulations (Natural Convection Correlation and Forced
Convection Correlation correlations). The formulation chosen for the particular surface is indicated in
the particular editor used (Natural Convection [Input Data Editor], Forced Convection [Input Data
Editor], etc). Full details of the formulations used are given in the online help. Each correlation if for a
separate geometric shape, e.g. horizontal or vertical cylinder, vertical plate, semi-open channel,
upper or lower sides of a horizontal plate, etc. Basically the authors of the various correlations have
made convection dissipation tests on various sizes of the particular geometric shape and fitted a
curve fit to their data. The curve fit is based on the fluid properties and the dimensions of the
geometric shape (the characteristic length). A process of Dimensional Analysis is used to develop
such formulations.
Dimensional Analysis:
In the technical literature the process of dimensional analysis usually followed to obtaining a
functional relationship between the physical properties and the fluid parameters. The use of
dimensional analysis makes the resulting correlation applicable to all such shapes and with any fluid
material - with the constraint that the flow conditions are similar to those of the tests, i.e. laminar or
turbulent.
A set of dimensionless numbers are used in the process:
Re = vL /
Gr = g ^2 L^3 / ^2
Pr = cp /k
Nu = h L / k
where
Natural Convection:
Nu = a (Gr Pr)^b
Forced Convection:
Nu = a (Re)^b (Pr)^c
There is generally a transition from laminar to turbulent flow for geometries having internal flow when
Re » 2300, with the flow becoming fully turbulent when Re > 5 x 104.
There is generally a transition from laminar to turbulent flow for geometries having external flow when
Re » 5 x 105.
Motor-CAD uses the best available correlation for the given surface. For some surfaces a composite
correlation is used, i.e. horizontal square housing = average of 2 x vertical plates + 1 x horizontal
upper surface + 1 x horizontal lower surface. The user can artificially increase or decrease the
amount of convection heat transfer from an individual surface using the h[adjust] parameter for the
particular surface - for instance this may be carried out if the user thinks that there is more
convection cooling than predicted due to drafts, etc. However, in the majority of cases the user will
just use the default calculation.
Full details of the correlation used in Motor-CAD are shown in Natural Convection Correlations and
Forced Convection Correlations.
The user can also completely ignore the default calculation for convection for a particular surface and
supply their own value of heat transfer coefficient, In order to do this the Input h check-box should be
ticked and a heat transfer value input in the h[input] column.
Altitude:
Altitude has a significant effect on convection cooling - thus the edit box is provided so the user
can select the altitude used for the calculation. More details regarding the altitude calculation used
in Motor-CAD is given in the US Standard Atmosphere.
Once heat transfer coefficient, h [W/m2/C] is calculated we can calculate the convection heat
transfer [W] for a given surface area, A [m2], and temperature [C]. The thermal resistance, Rt [C/W],
between the particular cooling surface and ambient due to natural convection is calculated using the
formula:
Rt = 1/(h.A)
This is in parallel with a radiation thermal resistance which is calculated from the Radiation Heat
Transfer - that is if the internal Radiation in include in the thermal calculation.
Units
See Natural Convection Data [Input Data Editor] and Forced Convection Data [Input Data Editor].
[1] Incropera, F.P & DeWitt, D.P.: Introduction to Heat Transfer, Wiley, 1990.
The following correlation formulae have been used in the calculation of the forced convection heat
transfer coefficient (hfc) from the various dissipating surfaces of the motor:
An indication of the formulation used for a given surface is given in the particular editor used in
Motor-CAD, i.e. Blown Over [Input Data Editor], Wet Rotor [Input Data Editor] and Housing Water
Jacket [Input Data Editor] Editors.
In some forced convection systems such as liquid cooled machines the fluid velocity is well defined
(from the flow rate and the known ducting cross-section). However, in Blown Over (TEFC) machines
with open fin channels the prediction of the local fluid velocity can be more difficult. Motor-CAD
includes channel Blockage and Leakage factors to try and help the user determine accurate
velocities.
The convection heat transfer coefficient, hfc, for a forced cooled flat plate is calculated using the
correlation [1]:
Laminar Flow ((Re < 5 x 10^5) and (0.6 < Pr < 50):
This assumes external flow, isothermal wall, laminar air flow over entire surface and air properties
calculated at film temperature = (Tfree-stream + Twall)/2.
Air Property Variation with Temperature can be modeled within the program see Air Properties.
The following graph plots the variation in h and Re with fluid velocity. The point at which the flow
becomes turbulent is clearly seen, i.e. rapid increase in h.
[1] Incropera, F.P & DeWitt, D.P.: Introduction to Heat Transfer, Wiley, 1990.
In some forced convection systems such as liquid cooled machines the fluid velocity is well defined
(from the flow rate and the known ducting cross-section). However, in Blown Over (TEFC) machines
with open fin channels the prediction of the local fluid velocity can be more difficult. Motor-CAD
includes channel Blockage and Leakage factors to try and help the user determine accurate
velocities.
The correlation used in Motor-CAD for open fin channel constructions is that of Heiles [1]. This is
based on testing on actual electric motors. In the correlation it is assumed that the flow is always
turbulent due to the fact that the radial fans and cowlings used in such machines create turbulence.
The convection heat transfer coefficient hfc is calculated using the formulation:
This assumes isothermal wall, laminar air flow with air properties calculated at the film temperature =
(Tfree-stream + Twall)/2.
Air Property Variation with Temperature can be modeled within the program see Air Properties.
Heiles also recommends the use of a Turbulence Factor to directly multiply hfc by - tests indicate
typical turbulence factors in the range 1.7 - 1.9 which seem independent of the flow velocity.
The inlet velocity to the fin channels must be estimated. We can use empirical data such as that
shown below. This shows that average velocity of the air in the fin channels as it leaves the fan. The
variation in velocity with shaft speed is as expected a linear relationship.
The actual variation in velocity from channel to channel can vary significantly and is a function of the
fan direction - see below. Alternatively we may know the volume flow rate. As we know the channel
dimensions and the inside diameter of the cowling we can calculate the velocity from the cross-
sectional area available for flow.
As slow or zero shaft speeds the velocity of air in the open fin channels becomes so low that the
natural convection can predominate. This is accounted for automatically in Motor-CAD as the Mixed
Convection is calculated. It is also reassuring that the Natural Convection Correlations used in
Motor-CAD have been developed such that they do not over-estimate the natural convection over
axial fin types that are not optimised for natural convection. An example of the natural convection
thermal resistance from housing to ambient both calculated using the default parameters and
measured on the 5 motors shown below is given the graph below [2]:
[1] Heiles, F., "Design and Arrangement of Cooling Fins", Elecktrotecknik und Maschinenbay, Vol.
69, No. 14, July 1952.
[2] Staton D. , Boglietti A., Cavagnino A., "Solving the More Difficult Aspects of Electric Motor
Thermal Analysis", IEMDC 2003 Conference Proc. 1-4 June 2003, Madison Wisconsin, USA.
It is noted that h increases dramatically as the flow regime changes from being laminar to turbulent
flow.
The following formulation is used to calculate the convection heat transfer coefficient (h) from
enclosed channels.
where
and
The flow is assumed to be fully laminar when Re < 2300 in Round/rectangular Channels and when
Re < 2800 in Concentric Cylinders
The flow is assumed to be fully turbulent when Re > 3000 (in practice the flow may not be fully
turbulent until Re > 10000)
A transition between laminar and turbulent flow is assumed for Re values between those given above.
Laminar Flow
For Concentric Cylinders (adaptation of formulation for parallel plates from Mills [1] which
includes entrance length effects):
The variable part of the above equation is the entrance length correction [2] which accounts for
entrance lengths where the velocity and temperature profiles are not fully developed.
For Round Channels (from Mills [1] which includes entrance length effects):
Turbulent Flow
From Gnielinski's [3] formula (fully developed turbulent flow, i.e. 3000 < Re < 1e6):
Typical results showing transition from laminar to turbulent flow for the enclosed channel correlation
is shown below. It is seen that the two formulations do not in fact join each other and a small
transition zone (taken from critical Re number dependent upon channel shape to 3000) is used to
make the two function join and so give numerical stability. A weighted average (based on Re) is then
used to calculate Nu in the transition zone.
It is seen that far larger heat transfer rates can be gained if turbulent flow can be achieved. Motor-
CAD shows how close to the critical Re number the design is working as shown below:
[2] Edwards, D.K., Denny, V.E., Mills, A.F. : Transfer Processes, 2nd ed., Hemisphere,
Washington, D.C. (1979)
[3] Gnielinski, V.: 'New equations for heat and mass transfer in turbulent pipe and channel flow', Int.
Chemical Engineering, 16, pp.359-368 (1976)
With radial fins it is possible to have axial flow with the flow passing across the tops of the fins as
shown above. Note: With radial fins the normal direction of blown over flow used is tangential flow
with the flow passing down the fin channels.
For this flow type the flow will pass over a proportion of the different fin surfaces based on the fin
geometry.
Forced convection correlations Fin Base [Ax] and Fin Side [Ax] are used for the cooling of the fin
base and side.
The aspect ratio of the fins is used to modify the heat transfer coefficient calculated using the Flat
Plate Convection Correlation.
Through Ventilation cooling is when there is a flow of fluid (usually air) through 3 parallel paths (stator
axial ducts, rotor axial ducts and airgap) of the machine. (See Flow Model [Through Ventilation])
In addition there is often some extra dissipation from the outside of the motor housing by Natural
Convection and Radiation
The fluid passing through the machine can be air or any other fluid as set to be the Internal Fluid
Data (See Cooling Options).
Through Ventilation cooling of the machine is enabled by selecting the Through Ventilation option
from the Cooling Options. The model is then set up using the Through Ventilation [Input Data Editor]
.
The through ventilation calculation is broken down into the following 2 steps:
1. The air flow through the machine is calculated. This depends on the Fan Definition (See
Through Ventilation Flow Options). This may be a fixed flow rate or calculated from the fan
characteristic or proportional to the machine speed or based on a non liner relationship with
machine speed.
2. Once the flow in the machine is known, the local velocities and heat transfer coefficients
(using the standard enclosed channel convection correlations) are calculated.
The Through Ventilation cooling type is selected using the Cooling Type selector in Cooling Options
[Input Data Editor]. Details of the model are given in Through Ventilation.
Specific editors are available for setting the Through Ventilation model data. The data is split into 4
pages:
This option is used to specify the fluid being blown through the machine. This may be air, any other
fluid from the fluid database or fixed fluid properties can be specified.
Flow Direction
The Flow Direction parameter is used to set the direction of flow in a Through Ventilation machine.
Front Inlet
Flow from drive end to non drive end. Flow through airgap and stator and rotor ducts if present.
Rear Inlet
Flow from non drive end to drive end. Flow through airgap and stator and rotor ducts if present
This option is enabled when there are radial ducts in the rotor and stator (see Radial Duct Geometry)
Average stator duct flow rate = stator duct inlet flow rate + rotor duct inlet flow rate / 2
Radial duct flow rates for rotor = rotor duct inlet flow rate
Radial duct flow rates for stator = rotor duct inlet flow rate
Heat transfer is calculated using the flow rates for the stator, rotor and radial ducts in the rotor and
stator.
No pressure drop calculation is currently made for entrance, exit effects and the flow across the
airgap or for the radial ducts. Radial flow is assumed to be separate from the airgap fluid.
See also Radial Duct Geometry
Flow Definition
The flow rate can be defined using one of the following options:
Constant Flow Rate There is an external blower that produces a constant specified
flow rate through the machine.
Constant Flow Rate from Fan There is an external blower that produces a constant flow rate
Characteristic through the machine. The flow rate is calculated from the fan
characteristic and the system resistance.
Flow Rate Proportional to SpeedThe fan is mounted on the shaft. A Reference Shaft Speed and
Reference Flow Rate are provided by the user. The flow rate
varies linearly with the shaft speed.
User Specified Flow Rate The fan is mounted on the shaft. A set of points of shaft speed vs
Variation with Speed flow rate are provided by the user. The flow rate varies non-
linearly with the shaft speed based on the set of points provided.
Inlet Definition
Use Circular Ducts in Housing for TVent Flow - when this option is enabled then TVent flow is
passed down any circular ducts in the housing.
Stator cooling only (No airgap or rotor flow) - when this option is enabled then the TVent flow passes
down the stator ducts only and assumed to be isolated from the rotor and so does not pass down
the airgap or any rotor ducts.
Editor for setting the Through Ventilation model Fan Characteristic Data and viewing the system flow
resistance curve and thus prediction of flow and pressure drop for the system.
The fan characteristic data is input using the editor shown above (Add Point and Remove Point
button functions). The minimum number of points is two. The Update Graph buttons sorts the data
into a logical order and plots the updated fan characteristic (in Red). Linear interpolation is used
between points. The system resistance characteristic is also calculated and plotted in blue. The
numerical values of intersection of the two characteristics is also indicted as shown above.
Note that if the fan characteristic is not known or if the user wishes to set the flow rate then this can
be done be entering two points with a flow rate equal to that required for the simulation as shown
below:
Editor for viewing the Through Ventilation model fluid flow and fine tuning the model.
The fluid is usually air - unless an alternative fluid is specified in Internal Fluid Data (see cooling
options).
It was intended to make it an option to include the airgap in the Through Ventilation flow path. But at
present the airgap is always included and so the Include Airgap Flow Path checkbox is always
checked and disabled.
k [Adjust]
The Through Ventilation Flow Resistance Minor Loss Factor (k factor) is calculated for all changes in
flow in the flow path through the machine. The k factors are based on proven formulations but the
user can adjust the formulations if required using the k [adjust] factors - this will affect the pressure
drops calculated for that particular flow restriction.
Table
The table section of the screen indicates full details of the flow path (path dependent upon the
options chosen in Flow Options). It indicated for each component of the flow path the following
information:
which minor loss factor (k) calculation algorithm is used
the value of k - this can be modified using the k [Adjust] parameter
Calculated Through Ventilation k-Factor (Flow Resistance Minor Loss Factor) data is displayed in
the table:
the area associated with the k factor calculation
the flow resistance - R
the volume flow rate - Q
the fluid velocity
Editor for viewing the Through Ventilation heat transfer and fine tuning the model.
For each surface within the flow path the following information is presented:
Input h - if this check-box is checked the user can input their own value of heat transfer coefficient
for the given surface
Convection Correlation - The Channel Correlation type chosen for each flow path component.
h[input] or h[adjust] - user adjustment factor for heat transfer coefficient - or the users own heat
transfer coefficient if the Input h check-box is checked
Local Velocity Multiplier - user adjustment factor for local velocity
Local Fluid Velocity - an indication of the local fluid velocity for the given surface.
Re Number - Reynolds number for the given flow path. This value is used to determine if the flow
is laminar or turbulent and therefore which correlation to use (Enclosed Channel Convection
Correlation)
Area - Surface area of the given surface. Together with the h value this is used to calculate the
TVent thermal resistance values put in the Schematic
h - Heat transfer coefficient for the given surface. Together with the Area value this is used to
calculate the TVent thermal resistance values put in the Schematic
Notes - area where the user can add notes
Input h
h[input] or h[adjust]
For all surfaces in the Through Ventilation flow path the heat transfer coefficient is calculated using a
proven formulation (i.e. Enclosed Channel Convection Correlation).
The user can however ignore this calculation and provide their own heat transfer coefficient data for
any surface.
The user can set their own heat transfer coefficient - just check the relevant Input h checkbox and
input a value in the h[input] or h[adjust] column.
The user can also adjust the internal calculated heat transfer coefficient by putting the multiplying
factor in the h[input] or h[adjust] column (default value = 1).
The Enclosed Channel Convection Correlation is velocity dependent. The local velocity is calculated
from the volume flow rate and cross-section area of that particular part of the flow path (data given in
Fluid Flow [Tvent]). The user can adjust the calculated local velocity using the Local Velocity
Multiplier (default = 1).
The inlet/outlet vents at the front (drive end) and rear (non drive end) in a Through Ventilation machine
can be placed in various positions. The Front Inlet Type and Rear Inlet Type pull down selectors
shown above are used to set where the vents are placed. The options are:
In this case the vents are in the endcap axial surface. The number of vents and the diameter of vents
(assumed to be circular vent holes) are set by the Number Holes [Front & Rear] and Hole Diameter
[Front & Rear] parameters.
In this case there is a vent in the radial circumference of the motor housing or endcaps that extends
for the full circumference of the machine. Only the vent axial length (Grill Length [Front & Rear]) is
required to define the vent area.
In this case there are a series of rectangular vents in the radial circumference of the motor housing
or endcaps.
As with the Radial Full Circle, the vents are in the radial surface of the endcap or housing. But rather
than covering the full circumference of the housing they only cover part of it - typically there may be
two vents each covering just 1/4 of the periphery. The dimensions of the vents are set using the Grill
Number [Front & Rear], Grill Length [Front & Rear] and Grill Width [Front & Rear] parameters. The
Grill Length [Front & Rear] is the axial length of each of the vents and the Grill Width [Front & Rear]
the circumferential width.
Front Grill Free Area Ratio Sets the ratio of free area for flow to total vent area in a Through
Ventilation inlet/outlet vent, i.e. a value of 0.8 means that 80% of a vent is free for air flow while the
other 20% is covered by the vent cover (to stop objects entering the machine). (1 = no grill, 0 =
blocked grill)
Front Vent Total Area Vent area calculated from the Vent dimensions (ignoring the
protection grill).
Front Vent Free Area Free flow area of Vent after taking off the protection grill area.
Rear Grill Free Area Ratio Sets the ratio of free area for flow to total vent area in a Through
Ventilation inlet/outlet vent, i.e. a value of 0.8 means that 80% of a vent is free for air flow while the
other 20% is covered by the vent cover (to stop objects entering the machine). (1 = no grill, 0 =
blocked grill)
Rear Vent Total Area Vent area calculated from the Vent dimensions (ignoring the
protection grill).
Rear Vent Free Area Free flow area of Vent after taking off the protection grill area.
The formulation is reported in [1] and [2] and is to account for the increase in pressure drop with
rotational speed in rotating ducts.
It is more debatable if such an adjustment should also be applied to the airgap and so the user can
select to apply the adjustment to the gap or not by using the appropriate check box -
. By default the adjustment is not applied to the airgap.
The data is set in Fluid Flow [Tvent] of Through Ventilation [Input Data Editor]
References:
[1] J.L. Taylor - Calculating air flow through electrical machines, Electrical Times, 21 July 1960
[2] Lightband & Bicknell
The flow path is assumed to have three parallel paths (stator axial ducts, rotor axial ducts and
airgap) as shown in Figure 1 below:
Turbulent flow has been assumed. This results in Q^2 rather than Q in the following relationship:
P = R Q^2 [1]
Where:
P = Pressure Drop [N/m2 or Pa]
Q = Volume Flow Rate [m3/s]
R = Flow Resistance [kg/m7]
Changes in flow condition - such as expansions and contractions in the flow circuit and
The flow resistance is calculated for all changes in the flow path using the formula:
R = k r / (2 A^2) [2]
Where :
k = Minor loss factor - dimensionless parameter that depends upon the type of expansion,
contraction, etc
r = air density (kg/m3)
A = area of flow section (relates to the k factor formulation)
Formulations have been added to Motor-CAD to calculate the k factors for all changes in flow section
within the motor. The resistances are then calculated from the associated flow areas. The system
resistance characteristic can then be calculated and its intersection with the fan characteristic
found. The flow path around the end-windings is quite complex making it difficult to choose whether
to use a bend or an expansion/contraction flow resistance. The solution is to calculate both and just
Radial Full Circle use the maximum value in each case (both values are however output in the
program).
There are two flow resistances associated with the inlet as shown in Figure 4 and Figure 5:
Inlet Entrance - to account for the protective cover over the vents.
Inlet Expansion - to account for the pressure drop due to a sudden expansion as the
air enters the motor through the restricted are of the vents.
Details of the Minor Loss (k) factors calculations used for the prediction of resistance to air flowing
through the vent grill are give in the following is given in the link - Flow Resistance Minor Loss Factor
. These are the values given in the [Fluid Flow] table in Motor-CAD. Area1 shown in Figure 5 is used
as the reference area in the flow resistance calculation in this case, i.e. used in equation 2.
Details of the Minor Loss (k) factors calculations used for the prediction of resistance to expanding
as it enters the motor is given in given in the link - Flow Resistance Minor Loss Factor. The area and
velocity associated with the Inlet Expansion is that before the expansion (Area1 in Figure 5). These
are the values given in the [Fluid Flow] table. The expanded areas used in the k-factor calculation is
shown in Figure 6.
Here we calculate the flow resistances for a 90 degree sharp bend and a contraction as the inlet air
is directed towards to stator ducting system. So as not to over-calculate the flow resistance at this
point we calculate both sets of resistances (using the formulations given in the link - Flow
Resistance Minor Loss Factor) but only use the largest resistance in the flow path calculation. The
position of the two resistances depends upon if the inlet is at the front or rear of the motor and if the
vents are on the axial surface of the endcaps or on the circumference of the endcaps/housing.
Typical area's used to calculate the two resistances are shown in Figure 8 and Figure 9. For the
bend we use the average of the areas at the two ends of the bend - this is the value given in the table
shown in Figure 7. The area and velocity associated with the Contraction is at the outlet end (smaller
area) - Figure 9. If the user feels that vents are positioned such that the Inlet Bend is not as dramatic
as that shown in Figure 8 and Figure 9 then they can use the k[adjust] factor in Figure 7 to weaken
the flow resistance.
These resistances are only calculated if the inlet is from an axial vent. It is to account for the
pressure drop as the air passes around the corner of the winding as shown in Figure 10 and Figure
11. The flow resistance is calculated as a 90 degree sharp bend and as a contraction/expansion
(contraction if the downstream area is smaller than the upstream area). Only the maximum of the
two resistances is included in the stator flow circuit.
The average of the two areas shown in Figure 10 are used for the k-factor and resistance calculation
for the bend. The areas shown in Figure 11 are used for the contraction/expansion - the choice of
expansion or contraction depends on the relative area values. The reference area is upstream for an
expansion and downstream for a contraction - these values are output in the table shown in Figure 7.
Figure 10: Stator Inlet EWdg Bend Areas used in k factor calculation
Figure 11: Stator Inlet EWdg Cont/Exp Areas used in k factor calculation
This is to account for the expansion in air as it passes over the winding extension before entering the
stator ducting - as shown in Figure 12. The reference area used in the calculation of the flow
resistance is the smaller upstream area. If the winding extension is set to zero then the resistance
does not exist.
Figure 12: Stator Inlet Wdg Ext Exp Areas used in k factor calculation
This is to account for the contraction in air as it passes into the stator ducting - as shown in Figure
13. The reference area used in the calculation of the flow resistance is the smaller downstream area.
This is to account for the expansion in air as it exits from the stator ducting - as shown in Figure 14.
The reference area used in the calculation of the flow resistance is the smaller upstream area.
This is to account for the contraction in air as it passes from the winding extension to the main body
of the end-winding - as shown in Figure 15. The reference area used in the calculation of the flow
resistance is the smaller downstream area. If the winding extension is set to zero then the
resistance does not exist and there is just the stator duct expansion.
Figure 15: Stator Outlet EWdg Contraction Areas used in k factor calculation
These resistances are only calculated if the outlet is from an axial vent. It is to account for the
pressure drop as the air passes around the corner of the winding as shown in Figure 16 and Figure
17. The flow resistance is calculated as a 90 degree sharp bend and as a contraction/expansion
(contraction if the downstream area is smaller than the upstream area). Only the maximum of the
two resistances is included in the stator flow circuit.
The average of the two areas shown in Figure 16 are used for the k-factor and resistance calculation
for the bend. The areas shown in Figure 17 are used for the contraction/expansion - the choice of
expansion or contraction depends on the relative area values. The reference area is upstream for an
expansion and downstream for a contraction - these values are output in the table shown in Figure 7.
Figure 16: Stator Outlet EWdg Bend Areas used in k factor calculation
Figure 17: Stator Outlet EWdg Cont/Exp Areas used in k factor calculation
Here we calculate the flow resistances for a 90 degree sharp bend and a contraction/expansion
(depending on the relative areas) as the stator air is directed towards the outlet vents. So as not to
over-calculate the flow resistance at this point we calculate both sets of resistances (using the
formulations given in the link - Flow Resistance Minor Loss Factor) but only use the largest
resistance in the flow path calculation. The position of the two resistances depends upon if the inlet
is at the front or rear of the motor and if the vents are on the axial surface of the endcaps or on the
circumference of the endcaps/housing. Typical area's used to calculate the two resistances are
shown in Figure 18 and Figure 19. For the bend we use the average of the areas at the two ends of
the bend - this is the value given in the table shown in Figure 18. The area and velocity associated
with the Contraction/Expansion is shown in Figure 19. If the user feels that vents are positioned such
that the Outlet Bend is not as dramatic as that shown in Figure 18 and Figure 19 they can use the k
[adjust] factor in Figure 7 to weaken the flow resistance.
Here we calculate the flow resistances for a 90 degree sharp bend and a contraction as the inlet air
is directed towards to rotor ducting system. So as not to over-calculate the flow resistance at this
point we calculate both sets of resistances (using the formulations given in the link - Flow
Resistance Minor Loss Factor) but only use the largest resistance in the flow path calculation. The
position of the two resistances depends upon if the inlet is at the front or rear of the motor and if the
vents are on the axial surface of the endcaps or on the circumference of the endcaps/housing.
Typical area's used to calculate the two resistances are shown in Figure 21 and Figure 22. For the
bend we use the average of the areas at the two ends of the bend - this is the value given in the table
shown in Figure 21. The area and velocity associated with the Contraction is at the outlet end
(smaller area) - Figure 22. If the user feels that vents are positioned such that the Inlet Bend is not
as dramatic as that shown in Figure 21 and Figure 22 then they can use the k[adjust] factor in
Figure 20 to weaken the flow resistance.
These resistances are only calculated if the inlet is from a radial vent. It is to account for the
pressure drop as the air passes around the corner of the winding as shown in Figure 23 and Figure
24. The flow resistance is calculated as a 90 degree sharp bend and as a contraction/expansion
(contraction if the downstream area is smaller than the upstream area). Only the maximum of the
two resistances is included in the stator flow circuit.
The average of the two areas shown in Figure 23 are used for the k-factor and resistance calculation
for the bend. The areas shown in Figure 24 are used for the contraction/expansion - the choice of
expansion or contraction depends on the relative area values. The reference area is upstream for an
expansion and downstream for a contraction - these values are output in the table shown in Figure
20.
2.3.3 MAX(Rotor Inlet EWdg Bore Bend, Rotor Inlet EWdg Bore Exp):
These resistances are to account for the pressure drop as the air passes around the inner (bore)
corner of the winding as shown in Figure 25 and Figure 26. The flow resistance is calculated as a 90
degree sharp bend and as an expansion. Only the maximum of the two resistances is included in
the stator flow circuit.
The average of the two areas shown in Figure 25 are used for the k-factor and resistance calculation
for the bend. The areas shown in Figure 26 are used for the contraction/expansion. The reference are
used in the flow resistance calculation is that upstream - this is the area shown in Figure 20.
This is to account for the contraction in air as it passes into the stator ducting - as shown in Figure
27. The reference area used in the calculation of the flow resistance is the smaller downstream area.
The rotor duct contraction and expansion are in parallel with the airgap flow path.
The k factor for the rotor duct entry contraction can be adjusted for rotation effects (increase in
pressure drop with rotational speed in rotating ducts). See Rotation Entrance Effects for more
details.
This is to account for the expansion in air as it exits from the stator ducting - as shown in Figure 29.
The reference area used in the calculation of the flow resistance is the smaller upstream area.
The rotor duct contraction and expansion are in parallel with the airgap flow path.
2.3.6 MAX(Rotor Outlet EWdg Bore Bend, Rotor Outlet EWdg Bore Cont):
These resistances are to account for the pressure drop as the air passes around the inner (bore)
corner of the winding as shown in Figure 30 and Figure 31. The flow resistance is calculated as a 90
degree sharp bend and as an expansion. Only the maximum of the two resistances is included in
the stator flow circuit.
The average of the two areas shown in Figure 30 are used for the k-factor and resistance calculation
for the bend. The areas shown in Figure 31 are used for the contraction/expansion. The reference are
used in the flow resistance calculation is that upstream - this is the area shown in Figure 20.
These resistances are only calculated if the outlet is from a radial vent. It is to account for the
pressure drop as the air passes around the corner of the winding as shown in Figure 32 and Figure
33. The flow resistance is calculated as a 90 degree sharp bend and as a contraction/expansion
(contraction if the downstream area is smaller than the upstream area). Only the maximum of the
two resistances is included in the stator flow circuit.
The average of the two areas shown in Figure 32 are used for the k-factor and resistance calculation
for the bend. The areas shown in Figure 33 are used for the contraction/expansion - the choice of
expansion or contraction depends on the relative area values. The reference area is upstream for an
expansion and downstream for a contraction - these values are output in the table shown in Figure
20.
Here we calculate the flow resistances for a 90 degree sharp bend and a contraction/expansion
(depending on the relative areas) as the rotor air is directed towards the outlet vents. So as not to
over-calculate the flow resistance at this point we calculate both sets of resistances (using the
formulations given in the link - Flow Resistance Minor Loss Factor) but only use the largest
resistance in the flow path calculation. The position of the two resistances depends upon if the inlet
is at the front or rear of the motor and if the vents are on the axial surface of the endcaps or on the
circumference of the endcaps/housing. Typical area's used to calculate the two resistances are
shown in Figure 34 and Figure 35. For the bend we use the average of the areas at the two ends of
the bend - this is the value given in the table shown in Figure 34. The area and velocity associated
with the Contraction/Expansion is shown in Figure 35. If the user feels that vents are positioned such
that the Outlet Bend is not as dramatic as that shown in Figure 34 and Figure 35 they can use the k
[adjust] factor in Figure 20 to weaken the flow resistance.
In the Synchronous Machine the through ventilation model for the airgap is more sophisticated and
takes account of the air passing down the interpolar gap and cooling from the outer and under
surfaces of the rotor coil and the outer surface of the rotor back iron. The contractions and
expansions of air passing down the interpolar gap due to the coil separators is also accounted for in
the pressure drop calculation.
References:
[1] J.L. Taylor - Calculating air flow through electrical machines, Electrical Times, 21 July
1960
[2] Woods Practical Guide to Fan Engineering, Woods of Colchester Ltd, 2nd Ed, June
1960.
[3] Lightband & Bicknell
[4] Mills, A.F., Heat Transfer, Prentice Hall, 1999
[5] Edwards, D.K., Denny, V.E., Mills, A.F. : Transfer Processes, 2nd ed., Hemisphere,
Washington, D.C. (1979)
[6] Gnielinski, V.: New equations for heat and mass transfer in turbulent pipe and channel
flow, Int. Chemical Engineering, 16, pp.359-368 (1976)
At present there are 6 types of flow resistance modelled in Through Ventilation model:
1. Inlet Grill/Guard
2. Outlet Grill/Guard
3. Sharp Bend
4. Sudden Expansion
5. Sudden Contraction
6. Sudden Cont/Exp
The link - Woods Minor Loss Factor Graphs - gives more details of some of the original curves/
formulations used to calculate the k-Factors
Used to calculate pressure drop at entry to system due to a Grill/Filter over the vents. The curve
used is shown below. This uses a combination of data from Woods [2] and Lightband & Bicknell [3]
(they give a value of k = 8 for a typical traction motor filter).
Note that the arrow in the diagrams relate to which area is used in the flow resistance calculation - it
also indicates which velocity is given in the Motor-CAD [Fluid Flow] table (outlet velocity in this
case).
An options for Louvres may be added later - Lightband & Bicknell (p351) give the following curve:
Used to calculate pressure drop at exit from system due to a Grill/Filter over the vents. The curve
used is shown below. This uses a combination of data from Woods [2] and Lightband & Bicknell [3]
(they give a value of k = 8 for a typical traction motor filter).
Note that the arrow in the diagrams relate to which area is used in the flow resistance calculation - it
also indicates which velocity is given in the Motor-CAD [Fluid Flow] table (outlet velocity in this
case).
For a right angle bend we use the worst case value of k = 1 and use the average area at each end of
the bend. This values seems to be confirmed by both Lightband & Bicknell and Woods - for a small
inner diameter bend.
k = (1 - Area1/Area2)2
This is plotted below. This is the same curve as that plotted by Lightband & Bicknell (p355) - but
they use Area2/Area1 rather than Area1/Area2. Woods has a similar curve but in terms of diameters
rather than areas. Both Lightband & Bicknell and Woods show data for gradual expansions rather
than sudden expansions (less pressure drop). In Motor-CAD we always assume the worst case of a
sudden expansion.
Note that the arrow in the diagrams relate to which area is used in the flow resistance calculation - it
also indicates which velocity is given in the Motor-CAD [Fluid Flow] table (inlet velocity in this case).
Both Woods and Lightband & Bicknell give data for sudden expansion minor loss factors. The
Lightband & Bicknell data is in terms of area and the Woods data in terms of duct diameters. Both
curves are effectively the same data. In Motor-CAD we use a curve fit of the Lightband & Bicknell
data.
Note that the arrow in the diagrams relate to which area is used in the flow resistance calculation - it
also indicates which velocity is given in the Motor-CAD [Fluid Flow] table (outlet velocity in this
case).
In this case depending on the areas at each side of the obstruction to flow a sudden expansion or
sudden contraction formulation is used.
References:
[1] J.L. Taylor - Calculating air flow through electrical machines, Electrical Times, 21 July
1960
[2] Woods Practical Guide to Fan Engineering, Woods of Colchester Ltd, 2nd Ed, June
1960.
[3] Lightband & Bicknell
[4] Mills, A.F., Heat Transfer, Prentice Hall, 1999
[5] Edwards, D.K., Denny, V.E., Mills, A.F. : Transfer Processes, 2nd ed., Hemisphere,
Washington, D.C. (1979)
[6] Gnielinski, V.: New equations for heat and mass transfer in turbulent pipe and channel
flow, Int. Chemical Engineering, 16, pp.359-368 (1976)
In Motor-CAD there are two models where fluid (air or liquid) is assumed to pass through the airgap:
The [Wet Rotor] model was implemented first. The airgap thermal resistances and the power taken
out of the airgap node are shown below:
Pw is the windage loss applied to each side of the airgap. Pairgap is the power taken out to the fluid
and is calculated using the formulation Power = dT x Volume Flow Rate x Cp. The thermal
resistance values between the bore the airgap node [Bore-Fluid] and
the rotor surface and airgap node [Rotor-Fluid] are calculated using the formulation:
Rt = 1/ (h Area)
The heat transfer coefficient (h) for each side of the airgap is calculated using a parallel plate channel
correlation with:
Hydraulic_Diameter = Dh = 2 x Airgap
Laminar Flow:
For concentric cylinders (adaptation of formulation for parallel plates from Mills [4] which
includes entrance length effects):
- The 2nd term in the above equation is the entrance length correction [5] which
accounts for entrance lengths where the velocity and temperature profiles are not
fully developed.
Turbulent Flow:
Calculated using Gnielinski's [6] formula for fully developed turbulent flow, i.e. 3000 < Re < 1e6:
Nu is calculated for both laminar and turbulent flow using the above formulations. A weighted
average (based on Re) is then used to calculate Nu.
It is noted that h increases dramatically as the flow regime changes from being laminar to
turbulent flow as shown in the typical plot below:
Fluid Velocity
The Channel_Fluid_Velocity used in the above is taken as the vector sum of the axial fluid velocity
(from flow rate and cross-sectional area) and the rotor peripheral velocity/2 (i.e. rotor surface is at the
rotor peripheral velocity and the stator surface is at zero velocity so I have taken the average of the
two). This same formulation is used to calculate h for both sides of the airgap.
The [Through Ventilation] model uses much the same circuit for the airgap - shown below:
Again it uses the vector sum of the average peripheral velocity (1/2 way through airgap) and the axial
fluid velocity in the calculation of Re and h.
The Through Ventilation model also calculates the pressure drop of the airgap due to the contraction,
expansion and surface friction. .
It is not clear rotational entrance effects should be applied to the airgap pressure drop as well as
those of traditional rotor ducts. By default the rotational entrance effects are not enabled for the
airgap pressure drops but can be enabled if required.
The Pressure drop calculation will be implemented for the Wet rotor in the near future.
References:
[1] J.L. Taylor - Calculating air flow through electrical machines, Electrical Times, 21 July
1960
[2] Woods Practical Guide to Fan Engineering, Woods of Colchester Ltd, 2nd Ed, June
1960.
[3] Lightband & Bicknell
[4] Mills, A.F., Heat Transfer, Prentice Hall, 1999
[5] Edwards, D.K., Denny, V.E., Mills, A.F. : Transfer Processes, 2nd ed., Hemisphere,
Washington, D.C. (1979)
[6] Gnielinski, V.: New equations for heat and mass transfer in turbulent pipe and channel
flow, Int. Chemical Engineering, 16, pp.359-368 (1976)
Data taken from "Woods Practical Guide to Fan Engineering, Woods of Colchester Ltd, 2nd Ed,
June 1960"
This model attempts to take account of the rotational movement of air in the machine endspaces.
When this option is selected then the endspace rotational velocity is taken into account when
calculating the cooling from the surfaces in the endspaces in the through ventilation model.
The rotational air velocities calculated in the endspace model shown above are used to calculate the
heat transfer coefficients in the through ventilation model as shown below:
The heat transfer coefficient is calculated using the mixed convection formulation
This area of machine cooling is one of the most difficult to predict accurately as fluid flow (air in most
cases) in the end space region of an electric motor is usually much more complex than that for flow
over its outer surfaces. This is because the fluid flow depends on many factors including the shape &
length of the end winding, added fanning effects due to simple fans and end-ring wafters & the
surface finish of the end sections of the rotor and turbulence. Several authors that have studied the
cooling of internal surfaces in the vicinity of the end-winding. In the majority of cases they propose
the use of a formulation such as that given below, i.e. a curve fit of the convection heat transfer
coefficient with local fluid velocity for surfaces in contact with the end space fluid:
h = k1 x [1 + k2 vel **k3]
The k1 x 1 term accounts for natural convection when the reference velocity is zero.
The k1 x k2 x vel **k3 term accounts for the added forced convection due to rotation.
A specific editor is available for setting the convection data for the surfaces surrounding the front and
rear end spaces within the motor.
The theory relating to the curve fits used to calculate the end space heat transfer coefficients is given
in End Space Convection Correlation.
End Space Velocity Multiplier [Front], End Space Velocity Multiplier [Rear]
The End Space Velocity Multiplier is used in the calculation of the End Space Reference Velocity.
The End Space Reference Velocity is calculated from the rotor peripheral velocity using the
formulation:
End Space Reference Velocity = End Space Velocity Multiplier x Shaft Speed [RPM] / 60 x pi x (
Stator Bore - 2 x Airgap)
The drop down boxes give some example values for different rotor smoothness. The smoother the
rotor then the lower the End Space Velocity Multiplier should be. By selecting the user defined
option the user can put in their own value for the multipliers.
Rotating Surfaces
The End Space Reference Velocity is not used for rotating surfaces. For rotating surfaces the local
fluid velocity is calculated from the rotation speed and average radius of the surface in such cases.
Endcap Vents
The default calculation is that the air within the endcaps is trapped. There is an alternative that there
are open Endcap Vents that allow an amount of flow into and out of the end space regions. Two
radio button selectors are available to set Closed or Open Endcap Vents for the front and rear of the
machine.
If the vents are open then an estimate of the volume flow rate (Volume Flow Rate ESpace) must be
provided. This flow rate can be either a fixed value or set to be proportional to the speed of the
machine. The endspace temperature is then calculated taken into account the flow through the vents
The inlet temperature is normally ambient but another temperature may be used for example if the
inlet comes from an enclosed cavity at a temperature different to ambient.
Shaft Speed
There is also an input option for Speed [RPM].
Internal Surface - description of motor component surface that can dissipate heat by
convection
k1, k2, k3 [dimensionless] - curve fir parameters used to calculate the heat transfer
coefficient (End Space [k1, k2, k3 factors])
Air Velocity [pu] - Local fluid velocity multiplier (End Space [local velocity multiplier])
Air Velocity [m/s] - Local fluid velocity. The Local fluid Velocity is used in the h = k1 x [1 +
k2 vel **k3] formulation.
h [W/m2/C] - Convection Heat transfer coefficient [h] of a given end space surface. h is
calculated using the data given in End Space Convection Correlation. It is used with the
surface area to calculate the thermal resistance of the given surface to the end space node.
Area [mm2] - Surface area of a given end space surface. The area is used with the heat
transfer coefficient to calculate the thermal resistance of the given surface to the end space
node.
Rt [C/W] - Convection Thermal Resistance to the end space node. Thermal resistance
between the given end space surface and the end space node. It is calculated from the
surface area and heat transfer coefficient of the given surface.
N.B. - Parameters shown in Red are input, those in Black are output.
See Units.
Data relating to cooling within the end space of the machine is set in End Space [Input Data Editor].
This area of machine cooling is one of the most difficult to predict accurately as fluid flow (air in most
cases) in the end space region of an electric motor is usually much more complex than that for flow
over its outer surfaces. This is because the fluid flow depends on many factors including the shape &
length of the end winding, added fanning effects due to simple fans and end-ring wafters & the
surface finish of the end sections of the rotor and turbulence. Several authors that have studied the
cooling of internal surfaces in the vicinity of the end-winding. In the majority of cases they propose
the use of a formulation such as that given below, i.e. a curve fit of the convection heat transfer
coefficient with local fluid velocity for surfaces in contact with the end space fluid:
h = k1 x [1 + k2 vel **k3]
The k1 x 1 term accounts for natural convection when the reference velocity is zero.
The k1 x k2 x vel **k3 term accounts for the added forced convection due to rotation.
Rotating Surfaces:
The reference velocity [vel] is set as the average velocity of the rotating surface, i.e. calculated from
the rotation speed, End Space [local velocity multiplier] and average radius of the particular surface.
Non-Rotating Surfaces:
The reference velocity [vel] is calculated from the rotor peripheral velocity and the values of End
Space Velocity Multiplier [Front], End Space Velocity Multiplier [Rear] and End Space [local velocity
multiplier].
The following table and graph show data from existing correlations of the end space cooling. It is
seen that the graphs give much the same trends - so that if one of the correlations is used the
resulting heat transfer prediction should not be too far from that achieved in practice
Comparison of Results:
The graph below compares the above formulations [with the exception of the Di Gerlando for h(end
winding/internal surfaces), which is gives h values far larger than the others at high velocities - this is
due to the vel**2 term)
It is seen that all the curve span quite a narrow band. It is possible to select any one of the
published characteristics directly in Motor-CAD using the function shown below in end space
settings:
The main goal to obtain an accurate estimate of the end space cooling is to select an appropriate
correlation form the ones given above and then to estimate the local velocity seen by all surfaces in
the end-space region. The velocity of the rotating surfaces is easy to calculate as it is just the
components peripheral velocity. The estimation of the velocities of the non-rotating components is
more difficult.
Motor-CAD calculates a reference velocity (ESpace - Ref Velocity [Front] and ESpace - Ref Velocity
[Rear]) for the front and rear end spaces. This is simply the rotor surface peripheral velocity
multiplied by a user setable factor (End Space Velocity Multiplier [Front] and End Space Velocity
Multiplier [Rear]). A value approaching 1 will be used for cases where there are large wafters of an
internal fan that lead to the velocity similar to the rotor peripheral velocity flowing over the non-
rotating surfaces. A lower value is used where no wafters or fan is used and the end space air is
relatively still. The default value of is 0.5.
The table editor contains an Air Velocity multiplier column so that the local air velocity can be
adjusted to be larger or smaller than the default. This is typically used to reduce the velocity of the
components surfaces shielded from the air flow, i.e. at the back of the end-windings.
Calculated Data:
For each internal convection surface, the following output parameters are calculated and displayed :
local air velocity
heat transfer coefficient (hr - W/m2/C)
component surface area
thermal resistance to ambient
These parameters can be useful for judging what contribution to cooling the internal convection from
each surface plays.
The two surfaces defined by Wdg Extension [F] and Wdg Extension [R] are provided in the end-
space model to allow air flow between winding extensions as shown below (channel formed between
the axial end of the stator laminations and the end winding turn). The surface area of each is equal to
2 x Slots x Slot_Depth x Winding_Extension. The dissipation is turned off by default (k1 = 0).
Altitude Effects
Altitude effects can have a significant effect on the convection. An option has been included to use a
new correlation that takes into account the effect of high altitudes. (See End Space Altitude Effects)
If a Sleeve is used in the airgap then the cooling around the end windings and inside of the housing
and endcap is adjusted automatically to account for the fact that the end space cooling of the these
surfaces is due to natural convection only, i.e. k2 set to zero:
The default calculation is that the air within the endcaps is trapped. There is an alternative that there
are open vents in the endcaps that allow an amount of flow through the end space regions. Two radio
button selectors are available to set Closed or Open Endcap Vents for the front and rear of the
machine. If the vents are open then an estimate of the volume flow rate (Volume Flow Rate ESpace)
must be provided.
Inlet Temperature ESpace: If there are inlet vents then there is the option to have the inlet
temperature at ambient or to specify some other temperature for the inlet.
The Open Endcap Vent cooling method is very similar to the [Wet Rotor] or [Through Ventilation]
models in that there is a flow of fluid through the machine. There is a fluid inlet and outlet that gives
rise to heat removal by the fluid according to the formulation:
P= T Q Cp [1]
where:
Formulation [2] is used to calculate the heat transfer from the surfaces in the end-space region and
formulation [1] is used to calculate the temperature of the fluid and the dissipation by the fluid.
If there is a flow of air past the end-winding then it may also be useful to pass air through the
channel formed between the winding extension (Wdg Extension [F] and Wdg Extension [R]) and
main body of end-winding and the axial edge of the stator lamination. The default values for End
Space [k1, k2, k3 factors] for the extension channels are set to zero so neglecting heat transfer for
these paths:
When the endcap vents are open an indication of the power taken out of the machine due to vents is
shown by the resistance paths in the Circuit Editor:
See End Space [Input Data Editor] and End Space Convection Correlation
When this option is disabled then Motor-CAD uses a correlation of the form:
(i)
where are curve fit coefficients that have been obtained by experimental investigations,
and ‘vel’ represents the local fluid velocity (m/s).
It has been established that none of the correlations of form (i) take into account the effects of
operating at high altitude where the air pressure is significantly lower. The effect of pressure must
clearly be taken into account since, at higher altitudes, there is a reduction in pressure which in turn
will lead to lower heat transfer coefficients and thus to a hotter machine.
A recent in depth analysis on this issue has been carried out by the University of Nottingham [1],
and has produced the following equation:
(ii)
where, the pressure ratio is taken directly from the International Standard Atmosphere (ISA)
table.
225.0000
200.0000
175.0000
150.0000
h [EWdg]
125.0000
Equation (i)
100.0000
Equation (ii)
75.0000
50.0000
25.0000
0.0000
0 2500 5000 7500 10000 12500 15000 17500 20000
Altitude (m)
From the figure above, it can be established that, using equation (i), the heat transfer coefficient is
being greatly over-predicted as the altitude increases.
Using the recommended equation (ii) a more realistic evaluation of end space transfer will be
obtained. This is done in Motor-CAD by enabling the Include Altitude Effects option.
Reference:
[1] G. Gilson, S. J. Pickering, C. Gerada and D. Hann, Analysis of the End Winding Heat Transfer
variation with Altitude in Electric Motors, IECON Conference, November 2009
One weakness of the end space convection correlation is that it is based on testing with air as the
cooling fluid. If we have a motor that is flooded with a fluid other than air then we need to either use
a [Wet Rotor] or [Through Ventilation] model or choose more appropriate values of k1, k2 and k3.
The [Wet Rotor] and [Through Ventilation] models are good if we have a flow of fluid through the
machine. Otherwise we need to alter k1, k2 and k3 - the fluid may be moving in this case but there
is no inlet and outlet so the heat transfer to the fluid is not directly removed from the machine
according to the formulation:
P= T Q Cp [1]
where:
There is a special type of end space cooling available in Motor-CAD where the endcaps are partially
open allowing a flow of fluid through the end-space. This is called the Open Endcap Vents. In this
case the user must specify the volume flow rate of fluid passing through the end space - see Figure
2. In this case formulation [2] is used to calculate the heat transfer from the surfaces in the end-
space region. Formulation [1] is used to calculate the temperature of the fluid and the dissipation by
the fluid.
End Space Cooling (Enclosed Fluids) - Modeling Fluids Other Than Air:
The heat transfer from surfaces in the end space region will be greater than that predicted using the
default k1, k2 and k3 data in formulation [2] if a fluid such as water is used rather than air. The
problem with formulation [2] is that it does not automatically account for a change in fluid material.
Most convection correlations in Motor-CAD are based on dimensional analysis of heat transfer and
so do automatically take account of a change in fluid (Re, Gr, Pr and Nu are functions of the material
data).
A formulation based on dimensionless analysis is implemented for the cooling surfaces in the end-
space that automatically scales the heat transfer for different internal fluids.
This is enabled by selecting the Include Internal Fluid Heat Transfer Coefficient Scaling option. This
should normally be enabled.
This scaling function scales the end space heat transfer coefficients assuming mainly turbulent flow
forced convection using the equation shown below:
h = Nu x k / l
h k x Re0.8 x Pr0.33
so
The calculated value of ESpace Fluid HTC Scaling Factor is given in the endspace output data sheet
.
Wet rotor cooling is where a fluid is passed down the airgap of the machine. The fluid also passed
over the shaft and end-winding surfaces in the endcap regions of the machine (these cooling paths
can be turned off if required). This form of cooling is often used for integrated pumps where leakage
from the pump is used to cool the motor.
In addition there is often some extra dissipation from the outside of the motor housing by Natural
Convection and Radiation
The fluid passing through the machine can be air or any other fluid as set to be the Internal Fluid
Data (See Cooling Options).
Wet Rotor cooling of the machine is enabled by selecting the Wet Rotor option from the Cooling
Options. The model is then set up using the Wet Rotor [Input Data Editor].
The wet rotor calculation is broken down into the following 2 steps:
1. The fluid flow through the machine is calculated. At the moment this is a fixed flow rate but
it is intended to include further flow options in the future.
2. Once the flow in the machine is known, the local velocities and heat transfer coefficients
(using the standard enclosed channel convection correlations) are calculated. (see Wet
Rotor Convection Correlation for details).
See Temperature [Wet Rotor], Flow [Wet Rotor], Dissipation [Wet Rotor], Area [Wet Rotor] and
Thermal Resistance [Wet Rotor] for the calculation results.
Wet Rotor cooling is where a fluid is passed down the airgap of the machine. The fluid also passed
over the shaft and end-winding surfaces in the endcap regions of the machine (these cooling paths
can be turned off if required). This form of cooling is often used for integrated pumps where leakage
from the pump is used to cool the motor.
Wet Rotor cooling of the machine is enabled by selecting the Wet Rotor option from the Cooling
Options.
The Wet Rotor sheet has the main user input sheets for defining the wet rotor flow and heat transfer.
Inlet Temperature:
Edit box for setting the wet rotor fluid inlet temperature.
Edit box for setting the wet rotor volume flow rate.
When enabled the duct wall friction of the different flow paths is calculated. The calculated flow
resistances and pressure drops can be viewed in the Wet Rotor Flow [Output Data].
At present the enabling of this function has no effect on the results of the model as the pressure
drops are not equalized between the parallel flow paths. In the future it is intended to calculate the
flow down the parallel paths of airgap, coil divider and slot wedge based on the pressures.
Fluid Properties:
Edit boxes for setting the wet rotor fluid thermal conductivity, density, specific heat capacity and
kinematic viscosity.
Outputs showing calculated values of the wet rotor fluid dynamic viscosity and Prandtl Number.
The wet rotor fluid properties can be either fixed values shown in the table or can have properties that
are dependent on temperature when the fluid is taken from the fluids database.
Radio-button selector for selecting if airgap heat transfer coefficient (calculated using the appropriate
Wet Rotor Convection Correlation) is based upon the:
Axial Fluid Velocity
Effective Fluid Velocity (combined axial & peripheral velocity)
Depending upon h[Wet Rotor] Velocity Selection, the mean velocity can be based upon the Axial
Fluid Velocity or the Effective Airgap Velocity (combination of axial and peripheral fluid velocity). The
default is to base the mean velocity on the Effective Airgap Velocity - this can give enhanced heat
transfer as the peripheral velocity can be much larger than axial fluid velocity.
Edit box for setting the wet rotor fluid axial airgap velocity multiplier. This is used in the calculation of
local fluid velocity in the Wet Rotor Table Editor - Wet Rotor Data [Input Data Editor].
Selector to select where the inlet and outlet for the wet rotor fluid is located.
The available options are:
Front Inlet - The inlet is at the front of the machine with the outlet at the rear end of the
machine.
Rear Inlet - The inlet is at the rear of the machine with the outlet at the front end of the
machine.
Central Inlet - The inlet is at the centre of the machine and it then passes to outlets at the
front and rear of the machine in two parallel paths.
Central Outlet - The inlets are at the front and rear of the machine with the fluid passing to
the outlet at the centre of the machine with two parallel paths.
Different inlet and outlet positions can be used. See Wet Rotor Inlet/Outlet Position for details.
Fluid Flow
There are options to pass the wet rotor fluid down the slot wedge and/or the coil divider. See Wet
rotor Wedge/Divider Flow for details.
Component - description of motor component surface that can dissipate heat by convection
Input h? - Check-box for choosing to input or calculate h[wet-rotor] for the particular motor
component surface
Convection Correlation - Indication of the Wet Rotor Convection Correlation formulation
used to calculate h[wet-rotor] of the particular motor component surface
h[input] or h[adjust] [W/m2/C or dimensionless] - Depending upon the check-box in
column Input h?, the forced convection heat transfer coefficient (h[wet-rotor]) can be input
directly or an adjustment multiplier input - this being used to modify the calculated value of h
[wet-rotor]
Local Velocity Multiplier [pu] - Multiplier used in determining the local fluid velocity.
Local Fluid Velocity [m/s] - Local fluid velocity.
Re Number [dimensionless] - Reynolds number = (equivalent diameter) * (mean velocity) /
(kinematic viscosity) - used to determine which convection heat transfer mode is applicable
(laminar or turbulent). See Wet Rotor Convection Correlation.
%Re(crit) [percentage] - the Reynolds number (Re) as a percentage of Critical Renolds
Number. This can be used to see how close to being turbulent the flow is. See Wet Rotor
Convection Correlation.
h [W/m2/C] - Wet Rotor Convection Heat Transfer Coefficient, h[wet-rotor], input in table or
calculated using the Wet Rotor Convection Correlation (Depending upon the contents of the
Input h? check-box)
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See Units.
Two extra surfaces are included in the Wet Rotor cooling method if the inlet and outlet are in the
radial position rather than the axial position. The new surfaces are on the end sections of the end
winding and the inner surface of the endcap.
Details of the new surfaces are given in the Wet Rotor [Input Data Editor]:
The fluid velocity is assumed to be totally due to the volume flow rate through the machine and is not
influenced by rotational effects - the effective velocity can be increased using the [Local Velocity
Multiplier].
The fluid in the airgap and the fluid in the regions between the shaft overhang and inner surface of the
end-windings is calculated using an Enclosed Channel Convection Correlation based on concentric
cylinders.
Depending upon h[Wet Rotor] Velocity Selection, the mean velocity can be based upon the Axial
Fluid Velocity or the Effective Airgap Velocity (combination of axial and peripheral fluid velocity). The
default is to base the mean velocity on Effective Airgap Velocity - this can give enhanced heat
transfer as the peripheral velocity can be much larger than axial fluid velocity.
An indication of the formulation used is given in the Wet Rotor Data [Input Data Editor] Table Editor:
The schematic indicates the heat extraction due to the wet rotor fluid and the average fluid
temperature at the given node as shown below:
If we do not wish to include a particular surface in the cooling path then we can simply turn it off by
setting its h[adjust] factor to zero as shown below:
The EW-Fluid thermal resistance will be automatically increased if the end windings are Potted or if
a Sleeve is used. The added resistance is due to the thickness of the potting and sleeve and the
thickness of any airgap between sleeve and airgap - this later gap is assumed to be a conduction
resistance rather than convection resistance (due to the small gap not allowing significant
convection).
See Wet Rotor Data [Input Data Editor] and Wet Rotor Output Parameters
Wet Rotor cooling passes a fluid down the airgap by default. There are some machines where it is
useful to provide some more direct conductor cooling. This is done by passing the fluid down the
slot.
Motor-CAD has options to pass the fluid down the slot opening and / or the coil divider.
To enable the slot opening or coil divider flow the slot must be defined with an air Divider and/or
Wedge as shown below.
Fluid Flow
The Checkboxes shown above are used to specify the flow paths.
At the moment the flow is split in proportion to the cross sectional areas of the paths. The
adjustment factors can be used to adjust the flow in the different paths.
In the future the flow rates in the different paths will take into account the Wet Rotor Duct Wall
Friction (see Wet Rotor [Input Data Editor] for details).
See Temperature [Wet Rotor], Flow [Wet Rotor], Dissipation [Wet Rotor], Area [Wet Rotor] and
Thermal Resistance [Wet Rotor] for the calculation results.
The Wet Rotor Active Fluid flow only option is intended for use when there is some flow down the
airgap but there is a significant amount of cooling in the endspaces due to rotational effects. This
option allows the endspace model to be used for cooling of the endspaces and the wet rotor cooling
When this option is selected then in the wet rotor cooling model only fluid flow down the active
section of the machine (airgap [and slot wedges and coil divider if this flow is enabled]) is
considered.
The airgap fluid flow path is coupled directly to the endspace nodes and the endspace model is used
for the cooling of the endspaces (endwindings, shaft and housing).
Housing water jacket cooling is when a fluid is passed down ducts in the housing as shown below:
In order to use a Housing Water Jacket a housing type that is suitable for a flow of liquid along ducts
in the housing should be chosen using the Housing Type selector this can be either the Water
Jacket (Axial) or Water Jacket (Spiral) housing type as shown below.
In addition there is often some extra dissipation from the outside of the motor housing by Natural
Convection and Radiation.
The fluid passing through the machine can be air or any other fluid.
Housing Water Jacket cooling of the machine is enabled by selecting the Housing Water Jacket
option from the Cooling Options.
The Housing Water Jacket Duct dimensions are set in the Radial Cross Section Editor and Axial
Cross Section Editor.
The model is then set up using the Housing Water Jacket Heat Transfer [Input Data Editor] and
Housing Water Jacket Fluid Flow [Input Data Editor).
The housing water jacket calculation is broken down into the following 2 steps:
1. The fluid flow through the machine is calculated. At the moment this is a fixed flow rate but
it is intended to include further flow options in the future.
2. Once the flow in the machine is known, the local velocities and heat transfer coefficients
(using the standard enclosed channel convection correlations) are calculated. (see
Enclosed Channel Convection Correlation for details).
See Housing Water Jacket [Output Data] for the calculation results.
Flow Definition
The flow rate can be defined using one of the following options:
Blower (Constant Flow Rate) There is an external blower that produces a constant specified
flow rate through the machine.
Blower (Constant Flow Rate There is an external blower that produces a constant flow rate
from Fan Characteristic through the machine. The flow rate is calculated from the fan
characteristic and the system resistance.
Shaft mounted Fan (Flow Rate The fan is mounted on the shaft. A Reference Shaft Speed and
Proportional to Speed) Reference Flow Rate are provided by the user. The flow rate
varies linearly with the shaft speed.
Shaft mounted Fan(User The fan is mounted on the shaft. A set of points of shaft speed vs
Specified Flow Rate Variation) flow rate are provided by the user. The flow rate varies non-
linearly with the shaft speed based on the set of points provided.
Editor for setting the Housing Water Jacket model Fan Characteristic Data and viewing the system
flow resistance curve and thus prediction of flow and pressure drop for the system.
The fan characteristic data is input using the editor shown above (Add Point and Remove Point
button functions). The minimum number of points is two. The Update Graph buttons sorts the data
into a logical order and plots the updated fan characteristic (in Red). Linear interpolation is used
between points. The system resistance characteristic is also calculated and plotted in blue. The
numerical values of intersection of the two characteristics is also indicted as shown above.
When housing water jacket cooling is used a fluid is passed down ducts in the housing (See
Housing Water Jacket Model).
The Housing Water Jacket checkbox should then be checked in Cooling Options [Input Data Editor]
to choose to pass a liquid down the ducts.
The Housing Water Jacket Fluid Flow editor is then available to set data such as:
Edit box for setting the housing water jacket volume flow rate.
Cooling Options
Channel Data
This gives an indication of flow area and channel dimensions.
The Housing Water Jacket Fluid Flow Table Editor is used to input/calculate/display the motors
water jacket fluid flow data. It has the following columns:
Component - description of motor component surface that can dissipate heat by convection
Flow Type - An indication of the Enclosed Channel Convection Correlation formulation used
to calculate h[water-jacket] of the particular motor component surface
Cross Section Area Calculated - Cross sectional area of the duct calculated from the
geometry.
Cross Section Area Adjustment - User adjustment factor if required.
Cross Section Area - Cross sectional area after adjustment taken into account. Used for
calculations.
N.B. - Parameters shown in Red are input, those in Black are output.
See also housing water jacket model, housing water jacket heat transfer, housing water jacket flow
options
See Units.
When housing water jacket cooling is used a fluid is passed down ducts in the housing (See
Housing Water Jacket Model).
The Housing Water Jacket checkbox should then be checked in Cooling Options [Input Data Editor]
to choose to pass a liquid down the ducts.
The Housing Water Jacket Heat Transfer editor is then available to set data such as:
Inlet Temperature:
Edit box for setting the housing water jacket fluid inlet temperature.
Edit box for setting the housing water jacket volume flow rate.
Fluid Properties:
Edit boxes for setting the housing water jacket fluid thermal conductivity, density, specific heat
capacity and kinematic viscosity.
Outputs showing calculated values of the housing water jacket fluid dynamic viscosity and Prandtl
Number.
The housing water jacket fluid properties can be either fixed values shown in the table or can have
properties that are dependent on temperature when the fluid is taken from the fluids database.
Cooling Options
Channel Data
The Housing Water Jacket Table Editor is used to input/calculate/display the motors water jacket
heat transfer data. It has the following columns:
Component - description of motor component surface that can dissipate heat by convection
Input h? - Check-box for choosing to input or calculate h[water-jacket] for the particular
motor component surface
Convection Correlation - An indication of the Enclosed Channel Convection Correlation
formulation used to calculate h[water-jacket] of the particular motor component surface
h[input] or h[adjust] [W/m2/C or dimensionless] - Depending upon the check-box in
column Input h?, the forced convection heat transfer coefficient (h[water-jacket]) can be input
directly or an adjustment multiplier input - this being used to modify the calculated value of h
[water-jacket]
Local Velocity Multiplier [pu] - Multiplier used in determining the local fluid velocity.
Local Fluid Velocity [m/s] - Local fluid velocity.
Re Number [dimensionless] - Reynolds number = (equivalent diameter) * (mean velocity) /
(kinematic viscosity) - used to determine which convection heat transfer mode is applicable
(laminar or turbulent).
%Re(crit) - percent of critical Reynolds number. At 100% the flow starts to move from being
laminar to turbulent.
h [W/m2/C] - Wet Rotor Convection Heat Transfer Coefficient, h[water-jacket], input in table
or calculated using the chosen convection correlation (Depending upon the contents of the
Input h? check-box)
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See also housing water jacket model, housing water jacket fluid flow, housing water jacket flow
options
See Units.
The Housing Water Jacket Type gives an indication of if a Spiral Groove or Axial Channel is used to
channel the fluid flow.
Spiral Groove when the Housing is set to Water Jacket (Spiral). The fluid is made to flow
along the spiral grooves in the housing
Axial Channel when the Housing is set to Water Jacket (Axial). The fluid is made to flow
axially along the channels in the housing. At the extreme end of each channel the fluid is
transferred to an adjacent channel via a u-bend (manifold).
Endcap Cooling
When this option is selected then the housing water jacket ducts are also in the endcaps. The ducts
can either be in a seperate cooling circuit or in series with the housing ducts. These ducts are not
currently drawn and the cross sectional area and length of the ducts needs to be entered using the
adjustment fields in the Housing Water Jacket Fluid Flow editor.
Flow Direction
Radio-button selector for selecting direction of fluid flow, i.e. from front of motor to rear or from rear of
motor to front.
When Parallel Flow Paths = 1 then all the flow passes through all the
ducts in series.
When Parallel Flow Paths = Number Flow Channels then all the ducts are in parallel.
When Parallel Flow Paths <> Number Flow Channels then there a groups of ducts connected in
series and these groups are connected in parallel.
The pressure requirement can be controlled to some extent by using parallel paths. As the number
of parallel paths is increased the volume flow rate and the effective channel velocity in each channel
is reduced. The flow paths can be seen in the Flow view.
The number of channels (ducts) in the housing is calculated from the geometry selected in Radial
Cross-Section Editor and Axial Cross-Section Editor. If the number of channels is small then there
can be a discrepancy between the actual number of channels in the motor and that in the model.
The Calculate or Input Flow Channels/Path selector can be set to Input in such cases and the
Number Flow Channels input directly - the calculated number of channels will be neglected in such
cases. If Calculate is selected then the calculated number of channels is shown. The number can be
a non integer.
In addition there is often some extra dissipation from the outside of the motor housing by Natural
Convection and Radiation.
This can be used at the same time as the housing water jacket if required.
The fluid passing through the machine can be air or any other fluid.
Shaft Spiral Groove cooling of the machine is enabled by selecting the Shaft Spiral Groove option
from the Cooling Options.
The Spiral Groove Duct dimensions (Shaft Groove Width, Shaft Groove Spacing, Shaft Groove Wall
and Shaft Groove Height) are set in the Axial Cross Section Editor.
The model is then set up using the Shaft Spiral Groove Heat Transfer [Input Data Editor] and Shaft
Spiral Groove Fluid Flow [Input Data Editor).
The shaft spiral groove calculation is broken down into the following 2 steps:
1. The fluid flow through the machine is calculated. At the moment this is a fixed flow rate but
it is intended to include further flow options in the future.
2. Once the flow in the machine is known, the local velocities and heat transfer coefficients
(using the standard enclosed channel convection correlations) are calculated. (see
Enclosed Channel Convection Correlation for details).
See Shaft Spiral Groove [Output Data] for the calculation results.
When shaft spiral groove cooling is used a fluid is passed down ducts in the shaft (See Shaft Spiral
Groove Model).
The Shaft Spiral Groove checkbox should then be checked in Cooling Options [Input Data Editor] to
choose to pass a liquid down the ducts.
The Shaft Spiral Groove Fluid Flow editor is then available to set data such as:
Edit box for setting the shaft spiral groove volume flow rate.
Cooling Options
Channel Data
This gives an indication of flow area, channel dimensions and inlet pressure
The Shaft Spiral Groove Fluid Flow Table Editor is used to input/calculate/display the motors water
jacket fluid flow data. It has the following columns:
Component - description of motor component surface that can dissipate heat by convection
Flow Type - An indication of the Enclosed Channel Convection Correlation formulation used
to calculate the heat transfer coefficient of the particular motor component surface
Cross Section Area Calculated - Cross sectional area of the duct calculated from the
geometry.
Cross Section Area Adjustment - User adjustment factor if required.
Cross Section Area - Cross sectional area after adjustment taken into account. Used for
calculations.
Length Calculated - Duct length calculated from the geometry.
Length Adjustment - User adjustment factor if required.
Length - Length after adjustment taken into account. Used for calculations.
R - Flow Resistance
Q - Flow Rate
P - Pressure Drop
Velocity - Fluid velocity
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See also Shaft Spiral Groove Model, Shaft Spiral Groove Heat Transfer[Input Data Editor]
When Shaft Spiral Groove cooling is used a fluid is passed down ducts in the shaft (See Shaft Spiral
Groove Model).
The Shaft Spiral Groove checkbox should then be checked in Cooling Options [Input Data Editor] to
choose to pass a liquid down the ducts.
The Shaft Spiral Groove Heat Transfer editor is then available to set data such as:
Inlet Temperature:
Edit box for setting the Shaft Spiral Groove fluid inlet temperature.
Edit box for setting the Shaft Spiral Groove volume flow rate.
The friction factor is normally automatically calculated by Motor-CAD using the formula given for
Turbulent Flow in Enclosed Channel Convection Correlation.
If required the user can enter their own values for the Shaft Spiral Groove ducts here.
Fluid Properties:
Edit boxes for setting the Shaft Spiral Groove fluid thermal conductivity, density, specific heat
capacity and kinematic viscosity.
Outputs showing calculated values of the Shaft Spiral Groove fluid dynamic viscosity and Prandtl
Number.
The Shaft Spiral Groove fluid properties can be either fixed values shown in the table or can have
properties that are dependent on temperature when the fluid is taken from the fluids database.
Cooling Options
Channel Data
This gives an indication of flow area, channel dimensions and inlet pressure.
The Shaft Spiral Groove Table Editor used to input/calculate/display the motors shaft spiral groove
heat transfer data. It has the following columns:
Component - description of motor component surface that can dissipate heat by convection
Input h? - Check-box for choosing to input or calculate h[spiral-groove] for the particular
motor component surface
Convection Correlation - An indication of the Enclosed Channel Convection Correlation
formulation used to calculate the heat transfer coefficient of the particular motor component
surface
h[input] or h[adjust] [W/m2/C or dimensionless] - Depending upon the check-box in
column Input h?, the forced convection heat transfer coefficient (h[spiral-groove]) can be
input directly or an adjustment multiplier input - this being used to modify the calculated
value of h[spiral-groove]
Local Velocity Multiplier [pu] - Multiplier used in determining the local fluid velocity.
Local Fluid Velocity [m/s] - Local fluid velocity.
Re Number [dimensionless] - Reynolds number = (equivalent diameter) * (mean velocity) /
(kinematic viscosity) - used to determine which convection heat transfer mode is applicable
(laminar or turbulent).
%Re(crit) - percent of critical Reynolds number. At 100% the flow starts to move from being
laminar to turbulent.
Surface Area - surface area of the ducts available for heat transfer.
h [W/m2/C] - Wet Rotor Convection Heat Transfer Coefficient, h[spiral-groove], input in table
or calculated using the chosen convection correlation (Depending upon the contents of the
Input h? check-box)
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See Units.
The Duct Type gives an indication that the Shaft Spiral Groove or Shaft Hole is used to channel the
fluid flow along the shaft.
Flow Direction
Radio-button selector for selecting direction of fluid flow, i.e. from front of motor to rear or from rear of
motor to front.
Flow Paths
Radio button selector for selecting if flow paths are in parallel or series.
When in series then the flow passes through Overhang -> Active -> Overhang.
When in parallel then the flow passes through the Overhang // Active // Overhang.
This form of cooling is where a fluid is passed down a duct in the shaft or the housing and then is
sprayed at the end-winding and/or rotor via nozzles. This fluid then drains out of the end-winding
and /or rotor and flows over various surfaces on the inside of the machine (giving some extra cooling)
and is drained out of the bottom of the machine.
The Spray Cooling model in Motor-CAD cannot be used at the same time as the Wet Rotor cooling.
The fluid passing through the machine can be air or any other fluid.
Spray Cooling of the machine is enabled by selecting the Spray Cooling option from the Cooling
Options.
If the cooling fluid enters via the shaft then the shaft should have a hole in it for the fluid flow. (See
Shaft Hole Diameter ).
If the cooling fluid enters via the housing then Housing Water Jacket cooling should be enabled and
the spray cooling inlet set to be the outlet temperature of the housing water jacket (see Cooling
Circuit Connections) .
The model is then set up using the Spray Cooling Heat Transfer [Input Data] and Spray Cooling Fluid
Flow [Input Data] pages.
The spray cooling calculation is broken down into the following 2 steps:
1. The fluid flow through the machine is calculated. At the moment this is a fixed flow rate but
it is intended to include further flow options in the future.
2. Once the flow in the machine is known, the local velocities and heat transfer coefficients are
calculated using different correlations for the flow down the shaft, onto the endwindings and
endcaps (see spray cooling convection correlation for details).
Calibration of the spray cooling models using test data is highly recommended if possible.
The spray cooling model accounts for the cooling of the spray hitting the inside bore of the end-
windings and also for the oil that splashed onto the other surfaces in the end-space, i.e. inside of
housing and endcaps and rotor axial surfaces. The oil is assumed to exit the motor via a sump hole
in the housing. The fluid flow path is shown below:
Another arrangement is to feed oil onto the rotor such that it is thrown off the axial ends by
rotation so hitting the end-windings and other surfaces in the end-space. This is the oil spray
cooling type that is used as a calibration example later in the document.
Spray cooling is when a fluid is passed down a duct in the shaft and then is sprayed at the end-
winding and/or rotor via nozzles. This fluid then drains out of the end-winding and/or rotor and flows
over various surfaces on the inside of the machine (giving some extra cooling) and is drained out of
the machine.
The spray cooling method cannot be used at the same time as Wet Rotor.
The Spray Cooling checkbox should be checked in Cooling Options [Input Data Editor] to choose to
use the Spray Cooling method.
The nozzle position and sizes are defined using the Spray Cooling Fluid Flow editor.
The Spray Cooling editor is then available to set data such as:
Inlet Temperature
Edit box for setting the spray cooling fluid inlet temperature.
Edit box for setting the spray cooling fluid volume flow rate.
Fluid Properties:
Edit boxes for setting the Spray Cooling fluid thermal conductivity, density, specific heat capacity
and kinematic viscosity.
Outputs showing calculated values of the Spray Cooling fluid dynamic viscosity and Prandtl Number.
The Spray Cooling fluid properties can be either fixed values shown in the table or can have
properties that are dependent on temperature when the fluid is taken from the fluids database.
Component - description of motor component surface that can dissipate heat by convection
Input h? - Check-box for choosing to input or calculate h[wet-rotor] for the particular motor
component surface
Convection Correlation - Indication of the Spray Cooling Convection Correlation
formulation used to calculate h of the particular motor component surface
h[input] or h[adjust] [W/m2/C or dimensionless] - Depending upon the check-box in
column Input h?, the forced convection heat transfer coefficient (h) can be input directly or
an adjustment multiplier input - this being used to modify the calculated value of h
Local Velocity Multiplier [pu] - Multiplier used in determining the local fluid velocity.
Local Fluid Velocity [m/s] - Local fluid velocity.
Re[Imp] [dimensionless] - Impingement Reynolds Number for spray target or Re for
channel/plate. See Spray Cooling Convection Correlation
Re[Wall] [dimensionless] - Wall Jet Reynolds Number for spray target or Re for channel/
plate. See Spray Cooling Convection Correlation
Nu[Imp] [dimensionless] - Impingement Nusselt Number for spray target or Nu for channel/
N.B. - Parameters shown in Red are input, those in Black are output.
The Spray Cooling Fluid Flow editor allows the user to:
define the position, number, sizes and flow through the nozzles.
define how much of the fluid passes over the internal surfaces inside the end spaces after it has
hit the target area but before it has drained out of the machine.
Edit box for setting the spray cooling fluid volume flow rate.
Nozzle Data
Several edit-boxes for defining the following spray cooling nozzle data.
Number of nozzles - Number of nozzles spraying onto each component. When set at 0 then there
is no nozzle.
Nozzle Diameter - Spray cooling Nozzle Diameter.
Nozzle Flow Rate ratio - the proportion of the total flow given that comes out of this set of nozzles.
Internal Flow Rate ratio - the proportion of the total flow given that flows over the internal surface.
The data is used to calculate the heat transfer due to the flow of fluid over various components
after it has hit the target area and before it is drained out of the machine.
Target Axial Length Calculated - Calculated value of spray cooling target area axial length.
Target Axial Length Adjustment - Adjustment factor for spray cooling target area axial length.
Target Axial Length - Value of spray cooling target area axial length used in calculation.
Flow Rate - the flow over the onto/over the surface of the component.
Local Fluid Velocity - the fluid velocity over the onto/over the surface of the component.
The nozzles are specified by giving Number of nozzles, Nozzle Diameter and Nozzle Flow Rate ratio
. In the example shown below there are nozzles pointed at the end winding bore surface at the front
and rear of the machine. See Spray Cooling Nozzles Positions for details of the nozzle positions.
The Nozzle Flow Rate ratio values are identical for each of the nozzles so the flow rate of 0.06 l/s is
split evenly between the front and rear nozzles.
Spray cooling is a complex mechanism with the spray that has impacted on the surface then flowing
over the internal surfaces inside the endspace.
This internal flow over the surfaces is modelled in Motor-CAD using the internal flow rate ratio.
In the example shown below half of the flow is over the endcap and the other half is over the rotor
pole and rotor core. The full flow then flows along the housing to the drain at the bottom of the
machine. In this case there is assumed to be no flow over the outer and end surfaces of the end
windings.
The data is used in the calculation of the spray cooling target heat transfer coefficient using the
Spray Cooling Convection Correlation.
Spray cooling is where a fluid is passed down a duct in the shaft and then is sprayed at the end-
winding via nozzles. This fluid then drains out of the end-winding and flows over various surfaces on
the inside of the machine (giving some extra cooling) and is drained out of the machine.
A simple Enclosed Channel Convection Correlation is used for the flow down the shaft and a Flat
Plate Convection Correlation for the fluid that passes over various surfaces before being drained from
the machine.
A special Impingement Jet Correlation is used for the cooling from the end windings. It is not known
how accurate this correlation is for this particular application so a special warning is given to warn
the user to calibrate the model if possible:
The user must set the Spray Fluid Volume Flow Rate and Spray Fluid Inlet Temperature. An
enclosed channel correlation is used for the shaft sections. The specific impingement jet correlations
are then used for the end windings.
A flat plate correlation is used for the endcap, housing and another other surfaces which do not have
nozzles directed at them which are both assumed to be in series with the end winding spray.
In the default case below the Internal Flow Rate ratio's are used to specify that for each end space
1/2 the fluid is flowing across the rotor and 1/2 is flowing across the endcap with the full flow across
the housing. More accurate values should be found from calibration.
The velocities of the fluid in the shaft and spray are calculated from the shaft hole diameter and
nozzle diameters respectively. The velocity of the housing fluid is set equal to the spray velocity
multiplied by the relevant Local Velocity Multipliers as shown below:
Default Local Velocity Multiplier values of 0.1 are used for the housing - more accurate values should
be found from calibration.
The Shaft Hole Diameter size is set within the Radial Cross-Section Editor:
Only the dissipation from the spray on the end windings are drawn on the schematic at present - the
Spray Cooling [Output Data] should be used to examine the results in more detail.
EW_Outer [Front]
EW_End [Front]
EW_Bore [Front]
EW_Outer [Rear], EW_End [Rear], EW_Bore [Rear], Rotor Pole [Rear], Rotor Core [Rear]
These nozzles are positioned on the rear endwindings in the same position to those on the front
endwindings shown above.
See also Spray Cooling Model, Spray Cooling Heat Transfer, Spray Cooling Fluid Flow [Input Data]
shown below:
The default spray cooling model assumes that nozzles are placed on the shaft to spray oil at
the end-winding bore. The spray targets selected are shown in [Geometry] – [Axial] if the
[Draw Cooling] checkbox is ticked in the lower right hand corner of the screen:
The user can energise further targets and set the number of nozzles and the nozzle flow ratio
(to have more flow hitting one target then the others) using the editor below:
The total flow rate for the oil is set in the editor above.
The [Internal Flow Rate Ratio] column is used to set the oil flow rate and subsequent velocity
of the oil passing over the other surfaces in the end-space (endcap, housing and axial ends of
rotor). It is input as a ratio of the total oil flow rate in the drive and non-drive (front/rear) end
of the motor. The default values are rough estimates that are as good as anything unless the
user has done previous testing or CFD on a similar motor.
Often it is worthwhile setting the units of flow to l/min using the [Options] – [Units] pull-
down menu:
In the [Spray Cooling] - [Heat Transfer] editor we can set the following:
Inlet temperature
Select fluid
The fluid thermal conductivity, specific heat, density and viscosity variation with temperature
can be input in the database.
The [Spray Cooling] – [Heat Transfer] editor also gives details of the correlation using to
calculate the heat transfer due to oil hitting the target:
The correlation for spray cooling was taken from the Electronics Cooling Magazine and is used
in this case to calculate the cooling from the spray target area. Due to the complexity of spray
cooling the red warning is given to remind the user that calibration should be done if a truly
accurate model is to be achieved, i.e. spray cooing being the most difficult convection model
that can be set in Motor-CAD due to the not very well defined flow circuit.
The [Spray Cooling] – [Fluid Flow] gives details of the Nozzle to Surface Distance and Target
Axial Length (from the target area) – these being parameters required by the spray cooling
correlation.
This correlation was first implemented in 2002. Since then the spray cooling model has been
used in several successful projects. As mentioned before, calibration using tests is always
recommended if possible for such cooling to obtain better accuracy.
The calculated heat transfer coefficient, surface area, fluid velocity, etc. are given in the
tables to help the user calibrate the model. Calibration adjustment factors for all the surfaces
heat transfer coefficients and local fluid velocity are by default set at 1:
Not only is data given for the surfaces sprayed with oil, but also for all other surfaces that see
the spray cooling fluid (before and after it has hit the end winding). An enclosed channel
correlation is used for the shaft sections. A flat plate correlation is used for the other surfaces
(assumed to be in series with the spray cooling targets).
The velocity of the fluid in the shaft and spray are calculated from the shaft hole diameter and
nozzle diameters respectively.
Note that if water jacket and spray cooling is used together then a problem that the housing
nodes that have both cooling types cannot be calculated at this time – we neglect the spray
cooling on such nodes.
Only the dissipation from the spray on the end windings are drawn on the schematic at
present to reduce the complexity of the diagram:
The cooling from the other surfaces can be seen in the [Circuit Editor] or [Output Data] pages:
This is an example of a more difficult spray cooling model than usual. In this case we did not
have a specific number of nozzles firing spray at the front and rear end windings, but oil
entering a void on the centre of the shaft and being thrown off the rotor by rotation induced
centrifugal forces to hit the end winding inner surface.
The motor cross-section is shown below with the spray cooling hitting the end-windings
highlighted by arrows in the axial cross-section editor:
The motor also has a water jacket. It is noted that the motor model dimensions used in this
example are not the actual ones used in the real machine that was tested due to
confidentially considerations. The method of setting up the calibration is however still valid.
In this case we will adapt the Motor-CAD model which assumes that the oil is passed down a
hole on the shaft and is sprayed directly at the inner surface of the end windings via a user
selectable number of nozzles set into the shaft. The spray coverage on the winding bore is
expected to be large so we can set a relatively large number of nozzles in the model to give
the same effect, i.e. 100 nozzles at end of the machine in this case:
In the model, which assumed nozzles are placed in the shaft, the velocity of the spray hitting
the end windings is calculated from the spray flow rate, number of nozzles and nozzle
diameter. The flow rate is set at 2 l/min and the nozzle diameter to 1mm in this case. This
results in a velocity of oil at 0.21m/s hitting the inside bore of the end-windings. However in
this more crude form of spray cooling, that does not use nozzles, the velocity is a function of
the rotational speed. We can adjust the [Local Velocity Multiplier] for the spray hitting the
end-winding bore to be a function of rotational speed to account for this, i.e. use this
parameter as a calibration constant:
In this case calibration was carried out using a set of 30 tests on a prototype machine (varying
speed, flow rates, etc.). The measured losses were fed into the model that had already been
calibrated for no oil cooling and the end-winding bore spray [Local Velocity Multiplier] varied
until the correct winding temperature was predicted in each case. This process was
automated using a MATLAB script. The figure below plots the values of spray velocity
multiplier required for each of the 30 tests:
1.6
1.4
1.2
Spray Velocity Muliplier
0.8
y = 0.0002x + 0.4141
R2 = 0.8448
0.6
0.4
0.2
0
0 1000 2000 3000 4000 5000 6000
Speed [rpm]
Plot showing va ria tion in End W inding Bore Spra y Velocity M ultiplier required to m a tch
m ea sured tem pera ture da ta for 30 test va lues
The graph shows a definite trend of an increase in spray velocity multiplier with speed. We
have plotted a liner line through the data points. The linear equation is given below:
More plots showing how the spray velocity, spray heat transfer coefficient, spray dissipation
and oil temperature rise vary with rotational velocity and oil flow rate are given in below. The
graphs show that the spray cooling is a primary function of rotational speed and a secondary
function of oil flow rate. If a correlation could also be identified relating to the flow rate then
the calibration equation shown above could be improved.
0.4
0.2
0.0
0 1000 2000 3000 4000 5000 6000
Speed [rpm]
700
flow rate = 1
600 flow rate =2
flow rate = 3
flow rate = 3.5
Spray h [W/m2/C]
300
200
100
0
0 1000 2000 3000 4000 5000 6000
Speed [rpm]
1000
flow rate = 1
flow rate =2
flow rate = 3
Spray Dissipation [W]
400
200
0
0 1000 2000 3000 4000 5000 6000
Speed [rpm]
flow rate = 1
flow rate =2
flow rate = 3
68 flow rate = 3.5
flow rate = 4.6
Spray Fluid Tout [C]
flow rate = 5
flow rate = 6.5
66
64
62
0 1000 2000 3000 4000 5000 6000
Speed [rpm]
We can implement an internal VBA script in Motor-CAD to automatically calibrate the spray
model using the equation below:
Note that we have set the [Run Before Analysis] option above so the script is run every time
we make a new thermal calculation. For instance below we have made a calculation at
2000rpm and the resultant calibration factors are shown below:
The spray cooling is dissipating around 1.7kW from each end winding in this case:
Again note that the above model is not the actual dimensions of the motor tested to create
the calibration data. So in this case the comparative values of dissipation from spray and
water jacket cooling should not be taken to be totally accurate but just an indication of what
type of output parameters are available from a calibrated model.
The submerged double jet impingement (SDJI) method for thermal testing of packages
By Evelien Driessens, IMEC and Fulip Chrisiaens, Alcatel
Excerpt from Electronics Cooling Magazine, May 2001.
Over the past decades, the functionality of electronic parts has improved considerably. Increasing power
requirements of semiconductor chips make it difficult to keep the temperature below the imposed limits.
Thermal management of the chip is one of the main functions of the package. To fully characterize the
package thermally, numerical simulations and experimental tests are performed. Numerical models offer
advantages because they are cost-effective and can be adjusted quickly. However, it is still necessary to
Thermal tests can be performed in different test environments. The standardized test environments used
at IMEC are the JEDEC box [1], which simulates a still air environment; a cold plate for obtaining higher
cooling conditions; and the fluid bath method. Some results for an arbitrary package tested in the three
Figure 1: An example of thermal resistance values in different test environments.The submerged double
jet impingement (SDJI) method has been developed and adjusted by F. Christiaens [2] to perform both
steady state and transient measurements. It was inspired by the need for accurate test methods to
validate thermal models. Very large heat transfer coefficients are obtained with SDJI, yielding a better
accuracy and repeatability of the thermal measurements. The original SDJI configuration was optimized by
determining the optimal working fluid and adjusting the geometry of the set-up. The heat transfer
coefficient of the new prototype SDJI method has been characterized. The practical implementation and
Nomenclature
DUT: Device-Under-Test
FOM: Figure-Of-Merit
p: Density (kg/m.sec)
The thermal performance of packages is traditionally expressed by the thermal resistance value:
Often an indirect measurement technique is needed to determine Tj, as the active chip surface is not
accessible for direct temperature sensing. In this case, a temperature sensing electrical parameter (e.g.,
the forward voltage of a diode) reacts to the applied power step and gives a direct correlation to the
junction temperature. Either a local or an average temperature can be measured with indirect techniques.
The measurements are nondestructive, very accurate and fast. Therefore, they can serve for both steady
The reference temperature is often a case or ambient temperature, denoted by Rjc and Rja respectively,
and depends of the environment in which the measurements are performed. The
It is difficult to measure the junction-to-case thermal resistance (Rjc) accurately with the fluid bath method
standardized by SEMI [3]. Rjc depends on the flow field around the package, the position of the test board
in the thermostatic bath and whether or not the fluid is stirred with an internal or external circulator. The
junction-to-ambient thermal resistance (Rja) is less sensitive to the flow field, but more sensitive to the
The existing standard method can be improved by increasing the heat transfer coefficient and defining,
around the package, a flow field that can easily be created and reproduced. The cooling performance of
the bath has been improved by directing a number of parallel impingement jets on the main surfaces of
the device and increasing the jet velocity. This leads to a reduction of the Rja values. The external
temperature difference between case and fluid is of the same order of magnitude as the temperature error
introduced by the thermocouple. Therefore, it is more accurate to measure the fluid temperature close to
Figure 2: Submerged double jet impingement configuration (a) and picture of the fluid bath with dielectric
oil (b).In practice, the component is submerged in a fluid bath with dielectric oil. The fluid in the bath is
continuously stirred with an internal circulation pump. Two dielectric liquid jets are directed on the main
surfaces of the component (Figure 2). An external pump steers the liquid jets through the nozzle outlets
and significantly increases the heat flow rate. A thermocouple is placed in the bath very close to the
component surface.
Improving the standard fluid bath method includes the selection of an appropriate fluid and the design of
an optimal nozzle configuration. To find a solution to these problems, first some hydrodynamic and
thermal issues of the method are considered. Next, the thermal resistance values obtained from
measurements in the adjusted test set-up were compared to numerically calculated values.
A submerged liquid jet can be subdivided into three flow regions (Figure 3): a free jet region, an
impingement zone, and a wall jet region. The orifice diameter, heat source length and nozzle-to-surface
separation distance are denoted by d, L and S respectively. The most relevant fundamental experiments
with respect to the SDJI configuration have been performed by Womac et al. [4]. They tried to correlate their
experimental data in water and FC-77 by considering the heat transfer in the impingement and wall region
separately. The following formula (the ‘Womac correlation’) yields the best fit for the two surface regions:
The cooling performance of the liquid jets is very sensitive to the nozzle geometry, the liquid velocity and
the type of fluid. Keeping the above-mentioned thermal and hydrodynamic considerations in mind, the
Next to good thermal properties, the working fluid should possess the following properties:
Non-toxic
Non-flammable
The use of water is precluded because of its chemical characteristics. De-ionized water rapidly looses its
Perfluorinated inert liquids, such as the Fluorinert liquids, made by the 3M Company, are suited for direct
liquid cooling because they fulfil the conditions mentioned above. However, some of their thermal
Other working liquids that can be considered for thermal measurements with the SDJI method are
Baysilone silicon oils manufactured by BAYER. They are clear, non-toxic and cheaper than the Fluorinert
liquids. The dielectric strength of the Baysilone and of Fluorinert liquids is comparable. Baysilone oils have
a relatively low density, and the thermal conductivity and specific heat are higher than for the Fluorinert
liquids. They are chemically compatible with most metals and plastics.
The primary requirement of chemical and electrical compatibility has narrowed the choice to working with
fluids with relatively inferior thermal transport properties. A careful fluid selection is of great importance in
order to minimize the external case-to-fluid thermal resistance. Therefore, an appropriate fluid thermal
figure-of-merit (FOM) is introduced, based on the above-mentioned empirical correlation for submerged
liquid jet impingement. A higher value of FOM means better cooling characteristics and, thus, lower
For small L/d, heat transfer predominantly occurs in the impingement region, represented by the following
figure-of-merit:
For large L/d, the wall jet region dominates the heat transfer:
From Table 1, it can be concluded that the heat transfer in the impingement region (denoted by FOMi) is
not very sensitive to the liquid type. However, when the wall jet region dominates heat transfer (FOMwj), a
significant difference is observed with the candidate liquids. In the actual SDJI cooling application, the
impingement and the wall jet region both dominate the heat transfer. Therefore the geometry of the nozzles
Pour poi nt (° 0 -90 -95 -95 -57 -50 -25 -60 -40 20
C)
Dens i ty (kg/ 998 1680 1730 1780 1870 1880 1940 900 920 960
m 3)
Therma l 0.598 0.057 0.060 0.063 0.066 0.066 0.070 0.105 0.116 0.150
conducti vi ty
(W/m°C)
Speci fi c 4180 1046 1046 1046 1046 1046 1046 1510 1510 1510
hea t
ca pa ci ty (J/
kg°C)
Dyna mi c 10×10-4 6.7×10-4 9.5×10-4 14×10-4 41×10-4 53×10-4 272×10-4 27.6×10-4 46×10-4 480×10-
vi s cos i ty 4
(kg/m s )
Di el ectri c 78 1.76 1.81 1.86 1.89 1.90 1.98 2.5 2.5 2.8
cons ta nt @
1 kHz
Vol umetri c 18×106 1×1015 1×1015 1.9×1015 4×1015 3.4×1015 2.3×1015 1×1014 1×1014 1×1015
res i s ti vi ty
(ohm/cm)
FOMi 1300 246 249 250 237 231 207 261 267 251
FOMwj 82000 20472 18795 16955 11800 10694 5906 11785 10380 4904
Table 1: Physical Properties of Fluorinert Liquids (3M) and Baysilone Oils (BAYER).
The hydrodynamic and thermal consideration discussed above only concern single liquid jets. To enhance
cooling and reach a higher degree of temperature uniformity, multiple jets can be used. The geometry of
Orifice diameter, d
Experiments in different cooling liquids were performed (water, FC-72 and HY-M3; Figure 4). DI-water has
better thermal properties than any other dielectric liquid. The higher liquid density of FC-72 causes an
increase in minimal orifice area and thus a decrease in the maximum orifice velocity obtainable at the
Figure 4: Convective heat transfer coefficient as a function of orifice diameter, number of orifices, and
nozzle-to-heater surface area ratio in three different working fluids (water, FC-72, HY-M3) (heater size = 25
In spite of the lower velocities, the thermal performance of FC-72 is still better than that of HY-M3. Three
nozzle geometries have been manufactured (Figure 5) for different package sizes.
Figure 5: Nozzle geometry for different package sizes (dimensions in mm): two 4×4 orifice grids for 25×25
mm2 (a) and 15×15 mm2 (b) packages, and one 7×7 orifice grid for 15×15 mm2 packages (c).In order to
characterize the convective heat transfer coefficient of the improved test set-up, thermal experiments have
been performed on dummy test structures. This is explained in the following section.
With the three nozzle geometries mentioned above, the heat transfer coefficients have been measured.
The surface temperature of some dummy heater components has been measured at different places and
averaged to find a uniform value of the heat transfer coefficient. The average convective heat transfer
References
1. JEDEC standard JESD51-2, "Integrated Circuits Thermal Test Method Environmental Conditions -
Natural Convection (Still Air)", Electronic
Rotor water jacket cooling is where a fluid is passed down ducts in the rotor
In addition there is often some extra dissipation from the outside of the motor housing by Natural
Convection and Radiation.
The fluid passing through the machine can be air or any other fluid.
Rotor Water Jacket Cooling of the machine is enabled by selecting the Rotor Water Jacket option
from the Cooling Options.
In order to use a Rotor Water Jacket the rotor must have Circular Rotor Ducts.
The model is then set up using the Rotor Water Jacket [Input Data].
The Rotor Water Jacket calculation is broken down into the following 2 steps:
1. The fluid flow through the machine is calculated. At the moment this is a fixed flow rate but
it is intended to include further flow options in the future.
2. Once the flow in the machine is known, the local velocities and the heat transfer coefficients
are calculated using the Enclosed Channel Convection Correlation for the flow down the
rotor ducts
See Rotor Water Jacket [Output Data] for the calculation results.
Model Details
An Enclosed Channel Convection Correlation is used to calculate the dissipation due to the fluid
passing down the rotor ducts. The amount of cooling of the rotor back iron is shown in the
Schematic:
A special form of cooling is if the Motor Type = BPM and the BPM Rotor = Embedded Radial and
the Rotor Duct Radius is such that the rotor ducts are placed between the lower and upper radius of
the magnets - as shown below:
In this case the cooling is from the node between the magnets:
See Rotor Water Jacket [Input Data] and Rotor Water Jacket Output Parameters
Rotor Water Jacket cooling is is where a fluid is passed down ducts in the rotor.
The Rotor Water Jacket checkbox should be checked in Cooling Options [Input Data Editor] to
choose to use the Rotor Water Jacket cooling method.
The Rotor Water Jacket editor is then available to set data such as:
Inlet Temperature
Edit box for setting the Rotor Water Jacket fluid inlet temperature.
Edit box for setting the Rotor Water Jacket fluid volume flow rate.
Fluid Properties:
Edit boxes for setting the Rotor Water Jacket fluid thermal conductivity, density, specific heat
capacity and kinematic viscosity.
Outputs showing calculated values of the Rotor Water Jacket fluid dynamic viscosity and Prandtl
Number.
The Rotor Water Jacket fluid properties can be either fixed values shown in the table or can have
properties that are dependent on temperature when the fluid is taken from the fluids database.
Radio-button selector for selecting direction of fluid flow, i.e. from front of motor to rear, from rear of
motor to front, from the two ends to the centre or the centre to the two ends.
Component - description of motor component surface that can dissipate heat by convection
Input h? - Check-box for choosing to input or calculate h[water-jacket] for the particular
motor component surface
Convection Correlation - An indication of the Enclosed Channel Convection Correlation
formulation used to calculate h[water-jacket] of the particular motor component surface
h[input] or h[adjust] [W/m2/C or dimensionless] - Depending upon the check-box in
column Input h?, the forced convection heat transfer coefficient (h[water-jacket]) can be input
directly or an adjustment multiplier input - this being used to modify the calculated value of h
[water-jacket]
Local Velocity Multiplier [pu] - Multiplier used in determining the local fluid velocity.
Local Fluid Velocity [m/s] - Local fluid velocity.
Re Number [dimensionless] - Reynolds number = (equivalent diameter) * (mean velocity) /
(kinematic viscosity) - used to determine which convection heat transfer mode is applicable
(laminar or turbulent).
%Re(crit) - percent of critical Reynolds number. At 100% the flow starts to move from being
laminar to turbulent.
Area - the surface area of the component.
h [W/m2/C] - Wet Rotor Convection Heat Transfer Coefficient, h[water-jacket], input in table
or calculated using the chosen convection correlation (Depending upon the contents of the
Input h? check-box)
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See Units.
Slot water jacket cooling is where a fluid is passed down ducts or in some cases between the
conductors in the slots
The fluid passing through the slot water jacket ducts can be air or any other fluid.
Slot Water Jacket Cooling of the machine is enabled by selecting the Slot Water Jacket option from
the Cooling Options.
Note: The Slot Water Jacket Ducts are displayed in the Motor-CAD cross section views.
The model is set up using the Slot Water Jacket [Input Data].
The Slot Water Jacket calculation is broken down into the following 2 steps:
1. The fluid flow through the machine is calculated. At the moment this is a fixed flow rate but
it is intended to include further flow options in the future.
2. Once the flow in the machine is known, the local velocities and the heat transfer coefficients
are calculated using the Enclosed Channel Convection Correlation for the flow down the slot
ducts
See Slot Water Jacket [Output Data] for the calculation results.
Model Details
In order to use a Slot Water Jacket ducts are be made in the slot to pass fluid (liquid or gas)
through.
The area for the water jacket competes with the area for the conductors so there is typically a
compromise between copper loss and cooling.
The duct sizes can be visually checked in the winding editor view as shown above.
The Slot Water Jacket [Input Data] editor is available to set slot water jacket cooling options.
Details of the amount of heat extracted due to the slot water jacket is indicated on the Schematic:
More details are also given in the Slot Water Jacket [Output Data]
Slot water jacket cooling is when cooling fluid is passed down the slots to cool the conductors.
Details of the Slot Water Jacket are given in Slot Water Jacket Model.
The Slot Water Jacket checkbox should be checked in Cooling Options [Input Data Editor] to
choose to use the Slot Water Jacket Cooling method.
The Slot Water Jacket editor is then available to set data such as:
Inlet Temperature
Edit box for setting the Slot Water Jacket fluid inlet temperature.
Edit box for setting the Slot Water Jacket fluid volume flow rate.
Fluid Properties:
Edit boxes for setting the Slot Water Jacket fluid thermal conductivity, density, specific heat
capacity and kinematic viscosity.
Outputs showing calculated values of the Slot Water Jacket fluid dynamic viscosity and Prandtl
Number.
The Slot Water Jacket fluid properties can be either fixed values shown in the table or can have
properties that are dependent on temperature when the fluid is taken from the fluids database.
Cooling Options
Flow Options
The fluid can either flow down ducts inside the slots or be flow in the space between the conductors.
Flow between conductors can be used in some large machines.
Duct Location
There can be 3 different duct positions in a slot. It is possible to have both the slot opening duct and
the slot bottom duct. If 2 ducts exist then they have the same size.
The duct locations are used in the cuboidal winding model to specify the faces that are cooled by
the slot water jacket duct. Note: If the winding layer model is used then the duct location is not
taken into account when cooling the winding.
Duct Width
The width of the duct placed in a slot to form a Slot Water Jacket
Note: In the present version no check is made to see if the duct area is available.
Duct Height
The height of the duct placed in a slot to form a Slot Water Jacket
There can be two ducts in a slot; the Upper Duct Exists and Lower Duct Exists. If both ducts exist
then they have the same size.
Note: In the present version no check is made to see if the duct area is available.
The wall thickness forms a conduction resistance whose value is a function of the thickness, duct
periphery and thermal conductivity. The thermal properties of the duct wall thickness should be
entered in the Material & Weights editor.
Insulation Thickness
An additional conduction resistance due to the Insulation Thickness can also be introduced to the
model.
The insulation thickness forms a conduction resistance whose value is a function of the thickness,
duct periphery and thermal conductivity. The thermal properties of the insulation thickness should be
entered in the Material & Weights editor.
Flow Direction
The coolant flow can be either from the front or from the rear.
Parallel Paths
The fluid velocity for ducts placed in a slot of a Slot Water Jacket is a function of the number of
Parallel Flow Paths and is set using this parameter.
If all the ducts are in series then the Parallel Paths should be set to 1.
If all the ducts are in parallel then the Parallel Paths should be set equal to Slot Number or two
times the Slot Number if there are two ducts in each slot.
Component - description of motor component surface that can dissipate heat by convection
Input h? - Check-box for choosing to input or calculate h[water-jacket] for the particular
motor component surface
Convection Correlation - An indication of the Enclosed Channel Convection Correlation
formulation used to calculate h[water-jacket] of the particular motor component surface
h[input] or h[adjust] [W/m2/C or dimensionless] - Depending upon the check-box in
column Input h?, the forced convection heat transfer coefficient (h[water-jacket]) can be input
directly or an adjustment multiplier input - this being used to modify the calculated value of h
[water-jacket]
Local Velocity Multiplier [pu] - Multiplier used in determining the local fluid velocity.
Local Fluid Velocity [m/s] - Local fluid velocity.
Re Number [dimensionless] - Reynolds number = (equivalent diameter) * (mean velocity) /
(kinematic viscosity) - used to determine which convection heat transfer mode is applicable
(laminar or turbulent).
%Re(crit) - percent of critical Reynolds number. At 100% the flow starts to move from being
laminar to turbulent.
Area - the surface area of the component.
h [W/m2/C] - Wet Rotor Convection Heat Transfer Coefficient, h[water-jacket], input in table
or calculated using the chosen convection correlation (Depending upon the contents of the
Input h? check-box)
Notes - User notes for table row
N.B. - Parameters shown in Red are input, those in Black are output.
See Units.
8.19 Submersible
Submersible Cooling of the machine is enabled by selecting the Submersible option from the Cooling
Options.
When the submersible option is not enabled then the machine is assumed to be operating with air
surrounding it.
By default the fluid outside the motor is air. This can however be changed to any other fluid by
selecting using the combobox shown above. The user can select whether to use fixed values of fluid
properties or to select a fluid from the database.
If the fluid material is selected from the fluid database then the fluid properties can vary with
temperature. In a transient calculation where the fluid temperature can change considerably over the
transient time period this is important to ensure the accuracy of the calculation.
8.20 Flooded
Flooded Cooling of the machine is enabled by selecting the Flooded option from the Cooling Options
.
When the flooded option is not enabled then the machine is assumed to be operating with air inside
it.
By default the fluid inside the motor is air. This can however be changed to any other fluid by
selecting using the combobox shown above. The user can select whether to use fixed values of fluid
properties or to select a fluid from the database.
If the fluid material is selected from the fluid database then the fluid properties can vary with
temperature. In a transient calculation where the fluid temperature can change considerably over the
transient time period this is important to ensure the accuracy of the calculation.
Thermal analysis for complex duty cycles is increasingly important in a variety of applications.
Motor-CAD allows the thermal modeling of simple and complex duty cycles.
Simple Transient - losses are fixed (unless loss variation with temperature is enabled see
below) as defined in the Losses [Input Data Editor]. Note that for a permanent magnet motor,
a rating test having constant torque or constant current over the entire length of the test can
be simulated using the function given in Losses Vary With Temperature and Load.
Duty Cycle Analysis - losses vary as defined in the Transient Duty Cycle [Input Data Editor].
Soak Back - same as Duty Cycle Analysis but machine has forced cooling that is removed
when power is switched off (see Soak Back) .
For each of the above transient calculation types the losses may also vary with temperature if the
models are enabled see Copper Loss Variation with Temperature, Loss Variation with Temperature &
Load.
1. Thermal capacitances are added to the motors thermal Schematic circuit so that the thermal
transient can be calculated. Each thermal capacitance is calculated from the Materials & Weights
[Input Data Editor] data for the particular motor component. Numerical values of the Thermal
Capacitance [Output Data] are also given and can also be plotted on the Schematic.
2. Iterative loop for each point in each period in each cycle of the transient to be calculated.
a) Losses are calculated based duty cycle data and temperature if required
c) Temperature and Loss values for this point are stored. (see Transient Results)
As the capacitances of the materials can normally be accurately calculated then good transient
thermal results can be achieved after a good steady state model has been created.
Motor-CAD has also options for modeling the transient behaviour of machines under fault conditions
(see Transient Fault).
Editor for setting the parameters required for the Transient Model and the calculation of the Transient
Graph.
The Transient Settings editor is allows the user to set the following parameters:
A radio button selector for selecting the transient calculation type - from one of the following:
Simple Transient - the losses defined in Input Data [Losses] for a set period of time
defined by Transient Period. The number of points to be calculated over the transient
period must also be defined - Number Points. Note that for a permanent magnet motor, a
rating test having constant torque or constant current over the entire length of the test can
be simulated using the function given in Losses Vary With Temperature and Load.
Duty-Cycle Analysis - the duty-cycle editor is provided to define the variation in torque
and speed (permanent magnet motors) or losses and speed over 1 cycle. The parameter
Number of Cycles defines the number of cycles of the defined duty cycle to be modelled.
Soak Back - this is a special form of duty-cycle analysis used with forced (air or liquid)
cooling to model the effect of a complete switch off in power. In this case when the power
is switched off the forced convection flow rate becomes zero. A switch off period is defined
by setting the torque, loss and speed all to zero for that period.
The External Air in Soak Back Calc and Internal Air in Soak Back Calc are used to set if the
External Fluid or Internal Fluid is replaced by air during the soak back period. The flow rates of any
cooling models active are also set to 0 during this period.
The transient start point option allows the starting point of the transient calculation to be specified.
This can be:
Ambient Temperature - all the machine components are at ambient temperature
Steady State Temperatures - the machine initial temperatures are those of the steady state
values. This can be useful when looking at a duty cycle that is present after the continuous steady
state temperatures have been reached.
Previous Transient Temperatures - to run another transient continuing with the end temperatures
from the previous transient calculation. The next transient calculation is appended to the current
data and plotted on the same graph. This can be used for instance to set some losses and then
calculate the thermal transient for a given time period. Then go back to the input data and set
other values of losses and continue the transient calculation. This can be very useful when
combining the thermal model in Motor-CAD with an external electromagnetic model. In such a
case we need to know the losses to calculate a thermal transient period - but the losses are
dependent upon the temperatures also (especially the copper losses). We can therefore just
calculate a short transient period and then re-calculate the losses knowing the new temperatures.
This can be fully automated using ActiveX technology.
Whole machine at specified temperature - when the machine starts at a temperature other than
ambient. This may be the case if have a coolant flowing through it that has a different temperature
than ambient.
Machine Components at specified temperatures - this can be useful for calibrating with test data
when the machine may not be at ambient.
Change in T ambient - Change in Ambient Temperature over Transient Time Period. It is assumed
that the ambient temperature changes linearly with time. Can be useful for matching transient test
data in which the ambient at the end of the test is different to that at the start of the test.
Number of Cycles - Number of full cycles (as defined in the Transient Duty Cycle editor) to be
modeled when the Transient Graph is calculated.
RMS Torque - Motor-CAD displays the calculated RMS torque of the duty cycle.
The RMS Torque of a duty cycle is calculated from the following equation:
Where
Ti is the torque for the period i
ti is the time duration of the period i
Average Speed - Motor-CAD displays the time averaged speed of the duty cycle.
Fault Analysis
See Transient Fault Analysis for details.
The Duty Cycle Editor allows the user to edit the duty-cycle waveform when Transient Calculation
Type = Duty-Cycle Analysis or Soak Back is used.
This editor is not visible when Transient Calculation Type = Simple Transient.
with Speed).
Note: To use the Loss Variation with Temperature & Load method with the transient solver the
Torque-Speed duty cycle data definition must be used.
Loss values are scaled from the Losses Editor Data using the pu value in the duty cycle.
The loss values may be scaled depending on speed if Speed Dependent Losses is enabled
(see Loss Variation with Speed).
The Copper losses may also vary with Temperature if have Copper Loss Variation with
Temperature or Loss Variation with Temperature & Load model enabled.
The calculated duty cycle losses can be viewed in the Transient Power Data.
Duty cycles can be loaded and saved to separated files if required. More details are given in External
Duty Cycle Data
Note: small duty cycles are normally also saved automatically in the .mot file.
Period - this is the sequential number of each period in the duty cycle
Elapsed Time - this is the cumulative elapsed time of the duty cycle.
Time - this is the time in this period of the duty cycle
Points - these are the number of points calculated during this duty cycle period.
Torque - value of unscaled torque during this period (only for BPM, BPMOR and PMDC
machines when using Torque-Speed Duty Cycle Definition)
Loss Columns - values of unscaled losses during this period.
Speed [Start] - the shaft speed at the start of this period
Speed [End] - the shaft speed at the end of this period
Fault - when check a fault is present during this period. (see Transient Fault Analysis)
End Ambient Temp - the temperature at the end of this period. Used when the ambient
temperature is varying during the duty cycle.
End Altitude - the altitude at the end of the period. This will have an effect on the fluid
properties in the model unless they are set to fixed values.
N.B. - Parameters shown in Red are input, those in Black are output.
Number of periods defining duty-cycle waveform over one cycle (BPM motors).
Losses can be read in from a duty cycle data file in watts or pu values (see Transient Duty Cycle
[Input Data Editor]).
The data format is defined in the first 4 lines of the datafile. It is assumed that the datafile is a text
file and has a name *.dat
The format of the duty cycle data file has changed as more functionality has been added. Previous
duty cycle formats should be able to be read by the current version of Motor-CAD.
To create a new duty cycle data file it is recommended to start from an existing file. This is easiest
done by setting the duty cycle type to that required and then selecting the save data option. The file
generated can then be edited in a text editor as shown below:
It is intended to add alternative formats that can be imported into Motor-CAD - so please contact
support if you have any special requirements.
See also Transient Duty Cycle Editor and Transient Settings Editor
Soak Back this is a special form of duty-cycle analysis used with forced (air or liquid) cooling to
model the effect of a complete switch off in power. In this case when the power is switched off the
forced convection flow rate becomes zero. At present this is applied to the blown over and through
ventilation cooling types. A switch off period is defined by setting the torque, loss and speed all to
zero for that period.
The External Air in Soak Back Calc and Internal Air in Soak Back Calc are used to set if the
External Fluid or Internal Fluid is replaced by air during the soak back period. The flow rates of any
cooling models active are also set to 0 during this period.
Below we see a typical Soak Back analysis. In this case it is for a small submersible used to drive a
camera around underwater. When in the water the cooling is in the form of forced liquid convection
with a flow of water along the axial length of the motor housing being at a velocity equal to the driven
speed of the submersible. Just before the submersible is taken out of the water the power is
switched off the losses become zero - the motor starts to cool down. But as it is taken out of the
water the cooling becomes much worse, i.e. just air natural convection and radiation. In such a case
the winding temperature falls but the housing temperature actually increases. In this case there is a
safe human operator handling temperature. Such analysis is also common in aerospace applications
to make sure that the housing temperature of a given motor does not rise above given limits when
switched off.
Motor-CAD has some capabilities for modeling the effect on the temperature rise of a fault occurring
on the winding.
The analysis used is very similar to the stall analysis except that the iron loss due to rotation is
present. As with the stall analysis we just model the worst slot (assume that all the slots have the
worst loss) and apply loss balance at the stator back iron node to reduce the power levels back to
that for the actual machine.
To enable a fault during a period of the duty cycle the fault is enabled as shown below:
The loss distribution during the fault period is defined using the loss distribution editor:
This editor gives a very flexible approach allowing many types of fault to be modelled.
See also Transient Settings Editor, Stall Analysis and Uneven Copper Loss Distribution
This type of fault can be analysed in a similar way to the 1-Phase open-circuit (square wave drive)
fault.
If we are to produce the same level of torque as in the machine with no open circuit phases we need
Phase/(Phases - Phases_o/c) more current in the good phases. The copper loss is proportional to
the current squared so the copper loss in the good phases is equal to [Phases/(Phases - Phases_o/
c)]2 more than without a fault. For the full machine the copper loss is equal to (Phases - Phases_o/
c)/Phases x [Phases/(Phases - Phases_o/c)]2 = Phases/(Phases - Phases_o/c) more than a non
faulty machine, i.e.
We then need to apply the stall analysis to account for the fact that the loss is not evenly distributed
in the machine. We can use the formula reported earlier:
[Stall Pcu Multiplier] = Worst Phase Loss / Phase Loss Assuming Even Loss Distribution
Example
The test case below illustrates the effects of such a fault in a 18 slot machine with 9 phases.
Below we see the thermal transient for the following duty cycle with no fault:
The copper loss in the machine is [9/(9-3)] x 180W = 270W when there is a fault (assume fixed level
of torque). This is concentrated in (9-3)/9 = 2/3 of the slots. So the amount of copper loss in the total
machine model assuming all slots have the same loss is 3/2 x 270 = 405W. We need to take 135W
from the stator back iron for loss balance in steady state.
Can see the increase in copper losses when the fault occurs. The stator back iron losses are also
reduced to take account of the uneven distribution of copper losses.
The loss adjustment at the stator back iron node is seen to be 60-39.2 = -20.8W in this case, i.e.
not the full 90W due to the storage term of the thermal capacitances.
The output data sheet also shows the copper loss multiplier being applied and the adjustment made
to the stator back iron losses.
Can see the power in stator back iron reducing over time. Now the power taken from the stator back
iron is -28.31 - 60 W = -88.31W which is close to the steady state prediction of -90W.
See also 1 Phase open circuit (square wave drive), Transient Settings Editor and Stall Analysis,
Uneven Copper Loss Distribution
8.22 Settings
8.22.1 Settings [Input Data Editor]
Models
Geometry
Winding
Cooling
Losses
Editor
End Space
Convergence
Transient
Miscellaneous
Airgap Model
The airgap heat transfer model can be made to be solely based on conduction or to include
convection as well. The convection model is recommended so it is the default setting.
In the original Motor-CAD model the thermal resistance of the endcaps was neglected. The main
reason why they were not included before was the fact that end-cap material thickness is not well
defined. Most often the endcaps have a very complex shape that does not lend itself to a simple
geometric definition using a minimum of parameters as required in Motor-CAD. Also the endcap
resistances are in series with the not well known bearing resistance, endcap-housing, endcap-
bearing and bearing-shaft interface gaps. There is no need to have too many thermal resistances in
series that are hard to define accurately - it is better to select one and calibrate the model using it, i.
e. bearing interface gap.
In the future more parameters (such as the end-cap thickness) may be added to allow a better
model for the endcaps. At present the user can alter thermal conductivity of the endcap material to
adjust their resistance values if required.
The Fin Efficiency parameter is calculated within the program to indicate the effectiveness of a fin in
transferring a given quantity of heat,
i.e. Fin Efficiency = Actual Heat Transferred / Heat that would be transferred if entire fin area were
at base temperature.
The Fin Efficiency can be left out of the steady-state/transient performance calculation if desired
using the checkbox shown above.
This option create a node in the thermal model for the coil divider when the Winding Type = Solid
Divider.
This node is create with a capacitance calculated from the weight of the coil divider and is connected
to the central winding node by a resistance that is calculated from the coil divider geometry.
The Coil Divider node is not connected automatically to any other part of the circuit. The node is
intended to be used with the circuit editor to connect an external circuit to model the complex
cooling path from the divider.
There are three options for the number of nodes used to model the heat transfer along a stator tooth:
1 Node
3 Nodes
Tooth + Tooth Tip
The selection is made using the [Stator Tooth Nodes] parameter in Models [Settings].
The [1 Node] model is the default. If the teeth become very long and the user is worried about the
fact that there is not sufficient nodes to model the large temperature rise along the tooth then three
nodes [3 Nodes] can be used. If a very large tooth tip is used then the [Tooth + Tooth Tip] model
may be more appropriate. The [Tooth + Tooth Tip] model id also useful for cases where there is
some excessive loss in the tooth tip region, i.e. when field weakening is used in a Non Overlapping
winding. The loss in the tooth is distributed evenly throughout the tooth and tooth tips if the [Tooth +
Tooth Tip] model is not used. When the [Tooth + Tooth Tip] model is used the loss in the tooth tip is
set using Loss [Stator Tooth Tip].
The changes in the schematic for the three cases are shown below:
Due to space constraints we do not plot the new nodes when the [3 Nodes] model is used. But the
we do give values for the power flow, temperature data for the three nodes and the three new thermal
resistances. In the new model three tooth temperatures are listed - the upper one is for the node
closest to the back iron and the lower one is closest to the stator bore. Similarly, for the power
values indicated near the winding to tooth resistance, the upper value is for the power to the node
closest to the stator back iron. Multiple power values are indicated next to the tooth resistances.
The power value closest to the bore node is for the resistance closest to the bore. The power value
closest to the back iron node is for the resistance closest to the back iron.
The multiple axial slice model is designed for some larger machines where there may be a
significant axial temperature gradient along the active section of the machine, particularly if there is a
cooling fluid flowing through the machine.
The user can select the number of axial slices to use in Models [Settings]:
The single slice is normally used unless there is a significant axial temperature gradient along the
active section. Different number of slices can be used to check if any difference is seen. The
advantage of the single slice is the model is easier to understand and quicker to calculate.
The axial model circuit and results can be view in the Schematic as shown below. The slice shown
can be selected using the slider.
The graphs show how the temperatures vary throughout the length of the machine:
Test results can also be shown the axial slice graph. Temperature values can be entered in the
temperature validation section as shown below:
These results are then displayed in the axial slice graph to help in comparing test data against the
model results as shown below:
Use Separate EndRing Nodes - This option is only available for Induction Machines. This option is
provided to allow backward compatiblity but should usually be enabled.
In Motor-CAD v6 an improved end ring model has been added. Previously there was only a single
node for modelling the induction machine cage.
When this option is enabled there are 3 nodes for modeling the rotor cage, 1 node for the active
section (bars) and a node each for the front and rear endrings.
The rotor copper losses are split between the rotor bars and the rotor endrings using the Rotor Cage
Loss Distribution formula.
Use Separate Front and Rear Rotor Nodes - This option is available for other than Induction
Machines. This option is provided to allow backward compatibility but should usually be enabled.
In Motor-CAD v6 an improved rotor end model has been added. Previously there was only a single
node for modelling the rotor back iron and a single node for the magnet.
The default magnet model in Motor-CAD is for a single node so that the losses are all placed on this
node and so the node temperature may not be the average temperature of the magnet.
The Distributed loss model takes account of the distributed nature of the losses (assumed even) and
calculates the average temperature based on these losses.
New definitions for the endcap and end winding expannsion have been added to the Axial Editor in
v4.
The end winding specification option specifies how the gaps are specified. You get the same model
using either specification. The original specification of the endwinding expansion was by specifying
the gap between the outer surface of the endwinding and the housing (this is often measurable).
This has now been changed by default to specify directly the expansion of the endwinding.
By default the new methods Endcap Thickness and Winding Expansion are now used in Motor-CAD
although existing Motor-CAD .mot designs will use the old methods.
In the original version of Motor-CAD the dimensions of fins placed on the housing were defined using
the following parameters:
Fin Thickness
Fin Pitch/Thick
The advantage of the above parameters was that as the machine diameter was scaled up and down
then the number of fins would be increased or decreased automatically. The disadvantage was that
some thought is required when putting in the dimensions for an existing finned housing. In such
cases often the number of fins and the fin spacing is known - this is why the new option to input the
fin dimension using these parameters was added.
Fin Thickness
Fin Number [Quadrant]
Fin Spacing [Fin Base]
Component Colour:
This defines how the components are coloured for visualisation purposes only.
When Parts with same colours is selected then components are assumed to be a single
component if they have the same material and there is no interface gap between them. In this case
the component is given the same colour in the cross sectional views.
For example the endspace and front of the housing have the same material and the interface gap
between them is set to 0 then the front of the housing has the endcap colour as is assumed to be a
single component.
Flow Arrows:
The flow arrows scaling option scales the flow arrows based on the proportion of the flow in the
different paths. This can be particularly useful for the through ventilation model. This scaling can be
enabled and disabled here.
The flow arrows sizes displayed in the radial and axial cross sectional views (see Machine Cooling
Flow Paths) can be set using the above interface.
The head width is the relative width (based on airgap size) of the arrow head.
The tail width is the relative width (based on airgap size) of the arrow tail.
The no. per flow and min length specifies the desired arrow length. If the requested number of arrows
and minimum arrow length is not suitable for the length available then the number of arrows will be
reduced.
The Spacing Factor is the distance between the arrows.
OpenGL 3D Settings
The 3D view is drawn using an OpenGL interface. The 3D curve precision setting allows the precision
of the curves to be increased or decreased if the computer graphics card is having problems
rendering the 3D image.
The following options are available for specifying the graphs displayed:
Torque Graph:
Points per cycle - the number of points to calculate for each electrical cycle
Number of cycles - the number of electrical cycles to run
The torque graph of the different torque calculation methods can be selected.
Current Graph:
The phase or Line to Line current values can be graphed.
Harmonic Graph:
Maximum Order - this is the maximum harmonic order to display in the harmonic graphs. The
maximum harmonic order will also be limited to half of the number of data points.
Amplitude values - the harmonics graphs can either show the magnitudes or the normalised
magnitudes
Losses:
Stator Iron Loss Build Factor - this multiplication factor is used to take account of differences in the
stator iron losses due to manufacturing
Rotor Iron Loss Build Factor - this multiplication factor is used to take account of differences in the
rotor iron losses due to manufacturing
Magnet Iron Loss Build Factor - this multiplication factor is used to take account of differences in the
magnet losses due to manufacturing
Iron Loss Calculation method - Bertotti or Steinmetz iron loss calculation methods can be used
Cores:
Rotor Iron - either laminated or solid rotor core
Stator Iron - either laminated or solid rotor core
Manufacturing Factors:
Stator EWdg length multiplier - the stator end winding length is calculated automatically based on
the winding pattern. The endwinding length calculated and used for the stator copper loss calculation
are shown in the winding output data sheet.
Magnet Br multiplier - this multiplication factor is used to adjust the magnet remanence from the
material datasheet to take account of any manufacturing factors.
D axis flux linkage multiplier - this multiplication factor will adjust the D axis flux linkage calculated.
This will impact on the inductance and flux linkage torque values calculated.
Q axis flux linkage multiplier - this multiplication factor will adjust the Q axis flux linkage calculated.
This will impact on the inductance and flux linkage torque values calculated.
Mesh Quality:
This controls the mesh density created. The default value of 1 is intended to give good results. This
value can be increased if a higher mesh density is required.
Model Size:
This specifies whether symmetry is used to reduce the size of the model. Note: symmetery is not
used if a custom winding pattern is used.
Air Divider Layer Connection - can set whether the endspace nodes are connected to the inner or
middle winding layer node.
End winding enamel model - when enabled then the wire enamel is taken into account in the thermal
resistance calculation of the endwinding to endspaces. This option should normally be used,
however this option is included for backward compatibility.
Liner Layers
Winding model - can set to use the original layer winding model or to use the cuboidal winding
model
End Winding Area Calculation - the old calculation method option is available here for backward
compatibility. The improved method should normally be used.
Insulation Resistance Calculation - the old calculation method option is available here for backward
compatibility. The improved method should normally be used.
Conductor Positions - custom conductor positions can be saved or loaded from a separate datafile or
saved in the mot file.
Winding Termination
This allows the winding model type to be selected for the Sync machine rotor in the winding settings
page.
Original model - this is the original (old) winding model for the Type 1 Synchronous machine rotor.
The original model does not have the multilayer winding model so is not expected to be as accurate,
This option was introduced in Motor-CAD v5 is available for both the stator and rotor winding
models.
It allows the user to specify a target number of layers for the winding model rather than to have the
number of layers defined by wire diameter. It is particularly useful to use the target winding layers if
the wire diameter is large with few conductors per slot as the original method will give few layers in
the thermal model as shown below.
Generally the larger number of winding layers then the higher the maximum winding temperature
calculated. To get an accurate winding model the aim is to have the material distribution in the slot
the same as that in reality. This can be done visually or using a FEA model.
The winding layer model can be calibrated well using the Flux link to calculate the average and
maximum winding temperature and then by adjusting the number of winding layers to calibrate the
winding model to give these values.
With a target number of layers set to 10 the winding model shown below is produced. This winding
model has the same quantities of material in the slot distributed over more layers.
The EWdg Active Layer Connect option in Input Data [Calculation Options] allows the user to make
the connection between the end winding node and the active winding at the mid layer node (Mid
Active Layer) rather than the default of the outer layer node (Outer Active Layer). This is to try and
help overcome any modelling deficiency due to the lack of nodes available in the present model to
give an accurate representation of the winding active and end-winding heat transfer. The new
connection can help give better results for a Non Overlapping winding where conductors furthest from
the tooth sides in the active section tend to be those showing at the extremity of the end-windings
(the EWdg Active Layer Connect = Outer Active Layer can give better results in this case). We
could have made the connection from the end-winding node to the inner most active layer - but it was
decided that this may over predict any cooling due to the end-windings. The end turn length will tend
to be longer for the end turns at the extremity of the end winding than the internal ones - Motor-CAD
uses an average end turn length:
The original default tends to give good results for traditional Overlapping windings where the
conductors closest to the tooth sides in the active section tend to be those showing at the extremity
of the end-windings (winding outer layer). The increased axial layer model that is under development
will make this option redundant and it will give finer resolution of temperatures throughout the active
and end-windings.
When the EWdg Active Layer Connect = Mid Active Layer option is used the Schematic does not
draw the connections between the end-winding and active node correctly. The connections are as in
the default calculation of EWdg Active Layer Connect = Outer Active Layer. This is because it is
difficult to find space to draw the new connections. Also, only the outer most few layers are drawn
so the mid layer node may not be shown on the schematic. To try to assist the user to spot that the
Mid Active Layer connection is being used, the MN (mid node) symbol is drawn together with two
extra copper nodes:
Liner Layers is a pull-down list selector to set whether the liner is modelled as:
a single material (set in the Materials & Weights [Input Data Editor])
For multi layer liners the Liner Layer Data required is:
Number of Layers
Layer Thickness
Layer Thermal Conductivity
Layer Specific Heat
Layer Density
Most slot liners are a composite of different layers - but often the designer does not have data for
each layer and it is easier to model it as a single layer with some form of average material data.
The multiple layer model is more accurate as the layers make up an inhomogeneous mixture where
we cannot make a simple average of the layers thermal properties. The effective properties should be
calculated according to the following (this is done in the multiple layer model):
Effective specific heat capacity - Cp(eff) = [th(a)/th(total) x (a) x Cp(a) + th(b)/th(total) x (b)
mass dependant: x Cp(b) + th(c)/th(total) x (c) x Cp(c)] / (eff)
The insulation lifetime is a very important aspect of machine design and is dependent on the winding
temperature.
The general rule of thumb is that the insulation lifetime is halved for every 10oC rise in the winding
temperature.
There are 2 different lifetime estimation methods available. These module are both approximations
and produce similar results.
Montsinger Model:
Lifetime L = a x e-bT
where :
a = insulation lifetime at 0oC
b = a material constant
T = temperature in oC
Dakin Model:
Lifetime L = A x e(-B/T)
where :
A = insulation lifetime at 0K
B = a material constant
T = temperature in K
Example of Use
The estimated lifetime for a winding insulation system can be obtained from the Motor-CAD model
an example of its use is given below.
The estimated lifetime for 2 different points needs to be entered by the user. The default values are
used in this example with a the lifetime of 20,000 hours at 180oC and 10,000 hours at 190oC. (i.e the
lifetime halved for every 10oC temperature rise).
Calculated lifetime values are shown below. Can see that the lifetime of the insulation has been
reduced from 20,000 hours to 15,000 hours due to the temperature of the winding.
The lifetime of the winding insulation system can also be estimated for complex duty cycles where
the maximum winding temperature will vary over time an example is shown below:
In this example have a machine running with constant losses for 5 hours. The maximum winding
temperature gets to 210C.
If the machine were operating with a maximum winding temperature of 180C then would expect that
the lifetime would reduce by 5 hours during the 5 hours of running.
In this case as the winding temperature gets to 210C then the insulation lifetime loses about 25
hours during this 5 hours of running as can be seen in the graph below.
It can be seen that the rate of loss of lifetime (the slope of the graph) increases as the temperature
of the winding increases.
This setting specifies how the winding in the winding editor view is displayed and how the conductors
are positioned.
Show Conductors
This specifies whether the layered winding model or the conductors will be shown in the Winding
Editor as shown below.
Automatic Separation
When enabled the horizontal, vertical conductor separation between the conductors and to the slot
walls are taken from the Conductor Separation parameter (see Winding Editor).
When disabled then there is more control of the conductor placement with separate parameters for
Conductor - Base Gap, Conductor - Tooth Gap, Vertical Conductor Gap and Horizontal Conductor
Gap.
Conductor Placement
With Interleaved placement the conductors are nestled into each other as shown below.
With Side by Side placement the conductors sit next to each other as shown below:
Conductor Positions
With automatic positioning the conductors are placed using a collision detection algorithm starting
from the bottom corner of the slot between the slot wall and slot bottom.
With custom positioning the conductors are placed in a regular grid from the bottom corner of the
slot between the slot wall and slot bottom. This allows the conductors to be precisely positioned in
the slot and is particularly useful when the winding in the slot is not symmetrical. Note: There is no
collision detection made so the user must ensure that the conductors are correctly positioned inside
the slot.
Position Checking
The conductors are placed in the slot using collision detection. The Position Checking parameter
controls the precision of this collision detection. 1 = fast, rough check, 10 = slow, precise check.
The default level of 5 should be suitable for most cases however in some cases it may be necessary
to increase this if the geometry is being exported to another package e.g. to a DXF.
The above parameters are used to account for any extra cooling due to heat extraction by the
winding termination. The user would usually define the termination to be at the drive (front) or non-
drive (rear) end of the machine using the Termination [F] and Termination [R] checkboxes.
The conduction resistance between the relevant end winding node and the housing node (front or
rear) would then be calculated using the formula:
Front:
R = Termination Wire Length [F] / [k[copper] x (Termination Wires [F] x pi/4 * (Termination Wire
Diameter [F])2]
Rear:
R = Termination Wire Length [R] / [k[copper] x (Termination Wires [R] x pi/4 * (Termination Wire
Diameter [R])2]
If a Termination [F] or Termination [R] is defined then the resistance will be shown in the Schematic:
The cuboidal winding model is a new winding model that gives more accurate temperature prediction
of the winding. The model uses cuboids[1] to describe the 3 dimensional heat transfer of the winding.
The winding is split into 6 cuboids. 2 cuboids inside the slot (1 at slot bottom the other at slot
opening), 2 cuboids for the front end windings and 2 cuboids for the rear end windings.
The cuboids have the advantage that the cooling surfaces of the winding are all considered and
different cooling methods can be examined. For example slot water jacket cooling with cooling from
the slot opening, slot base or slot centre.
The cuboidal model relies on the correct size and effective thermal conductivity values in the 3
directions. These values are calculated automatically based on the conductor locations, wire sizes
and slot fill used in the design.
The values calculated can be checked in the output data sheet as shown below:
If required the effective thermal conductivity and cuboid dimensions can be set by the user using the
options shown below:
The cuboidal model circuit can be displayed in the circuit editor tab as shown below. Each cuboidal
element has 6 nodes, 1 on each face and also a central average temperature node.
The copper losses are placed on the central node that has the effect of distributing the losses
throughout the cuboidal volume.
The positions of the cuboidal nodes in the slot are shown below:
[1] A General Cuboidal Element for Three-Dimensional Thermal Modelling. R. Wrobel, P.H Mellor,
IEEE Transactions on Magnetics Vol 46 No. 8, August 2010.
Include End Space Rotational Effects. This option should normally be selected.
Cooling Circuit Connections.This allows different cooling systems to be connected in series and
parallel.
Airgap Heat Transfer Calculation. The improved calculation method should normally be used.
This model attempts to take account of the rotational movement of air in the machine endspaces.
When this option is selected then the endspace rotational velocity is taken into account when
calculating the cooling from the surfaces in the endspaces in the through ventilation model.
The rotational air velocities calculated in the endspace model shown above are used to calculate the
heat transfer coefficients in the through ventilation model as shown below:
The heat transfer coefficient is calculated using the mixed convection formulation
The Cooling Circuit Connections options allow the different cooling systems to be coupled together
so that the outlet temperature and flow rate of one cooling system becomes the inlet temperature
and flow rate of another cooling system.
Note: not all connection options are currently available. Please contact support if you require another
connection possiblity.
The Original Calculation method in Motor-CAD was based on the work of G.I Taylor [1].
The work of G.I. Taylor [1] is used to determine the increase in heat transfer coefficient over that for
pure conduction due to convection induced between two concentric cylinders rotating relative to each
other. He found that the heat transfer is by pure conduction when the flow is laminar but there is an
increase in heat transfer when the airgap Reynolds number (Re[Airgap]) increases above a critical
value of Re[Airgap] Critical - Vortex as the flow takes on a regular vortex pattern. Above a higher
critical value of Re[Airgap] Critical - Turbulent the flow becomes turbulent and the heat transfer
increases further.
Equations Used:
Nu = 2 (conduction only)
h = Nu x k / (2 x Airgap)
When there is axial flow down the airgap then the effective velocity is used for calculating the
Reynolds number.
The Improved Airgap Heat Transfer Calculation method is based on the work of Chong [2].
When the rotational flow in an annular gap between concentric cylinders is superimposed with an
axial flow, Kaye and Elgar [3] revealed the existence of four possible modes of flow in an annular
gap. They can be distinguished and categorised using the axial Reynolds number and Taylor
number. These four modes are:
a) Purely laminar flow - when Taylor number is less than Non-Vortex to Vortex limit and Reynolds
number is less than laminar to turbulent limit.
b) Laminar flow with Taylor vortices - when Taylor number is greater than Non-Vortex to Vortex limit
c) Purely turbulent flow
d) Turbulent flow with Taylor vortices
In Motor-CAD the airgap heat transfer calculation is based on the Airgap flow type that is based on
the Taylor number and the Airgap Axial Flow Reynolds number as detailed below.
No Axial Flow
If Airgap Axial Flow Reynolds number = 0 then there is no axial flow so use Taylors number to
calculate airgap heat transfer as in the Original Calculation method shown above .
The transitions between different flow types are determined from the Critical values for Taylor number
and Reynolds number and blending the calculated values in the transition regions.
[1] Taylor, G.I.: 'Distribution of Velocity and Temperature between Concentric Cylinders', Proc Roy
Soc, 1935, 159, PtA, pp 546-578
[2] Chong, Y, C : The Study of Flow Resistances in Through Ventilated Electrical Machines.
Edinburgh University research work not yet published.
[3] J. Kaye and E. Elgar, “Modes of adiabatic and diabatic fluid flow in an annulus with an inner
rotating cylinder,” Transaction ASME, vol. 80, pp. 753-765, 1958.
[4] F. Tachibana and S. Fukui (1964), “Convective heat transfer of the rotational and axial flow
between two concentric cylinder,” Bulletin of JSME, 7 (26), pp. 385-391.
[5] D. Simmers and J. Coney, “A Reynolds analogy solution for the heat transfer characteristics of
combined Taylor vortex and axial flows,” International Journey of Heat and Mass Transfer, vol. 22,
pp. 679-689. (1979)
[6] T. M. Kuzay and C. J. Scott (1977), “Turbulent heat transfer studies in annulus with inner
cylinder rotation,” Trans. ASME, J. Heat Transfer, vol. 99, pp. 12-19.
[7] P. Childs and A. Turner (1994), “Heat transfer on the surface of a cylinder rotating in an annulus
at high axial and rotational Reynolds numbers,” Proc. 10th International Heat Transfer Conference,
Brighton, UK, vol. 3, pp. 13-18.
[8] C. Gazley (1958), “Heat transfer characteristics of the rotational and axial flow between
concentric cylinders,” Trans. ASME, Vol. 80, pp. 79-90.
[9] S. I. Kosterin and Y. P. Finatev (1962), “Heat transfer in turbulent airflow in the annular space
between rotating coaxial cylinders (in Russian),” Inzh. Fiz. Z. H., vol. 8, pp. 3-9.
Slotting:
Testing by Gazley [1] suggests that with laminar flow the slot opening can decrease the heat
transfer due to the longer effective gap length. He suggests values of around 10% for rotor slotting
and 20% for rotor and stator slotting. He suggests that with vortex flow there is no change in heat
transfer and for turbulent flow there can be a significant increase in heat transfer.
Saliency:
CFD analysis on large cavities (salient poles) by Hayase [2] suggests that there is a slight reduction
in heat transfer for rotor saliency with laminar flow. With vortex/turbulent flow he suggests a 20%
increase in heat transfer for rotor saliency and a 10% increase in heat transfer for stator saliency. No
data is given for rotor & stator saliency. There findings are in line with those of Gazley [1].
In Motor-CAD there is no automatic account of slotting and saliency on heat transfer as the findings
are rules of thumb rather than mathematical formulations. However, the user can increase/decrease
the airgap heat transfer by making use of the Airgap Conductivity Multiplier. Its default value is 1. If
the value of 1 is kept then it is thought that an over prediction of rotor temperature will result if the
flow is turbulent which is on the safe side.
[1] Gazley, C.: 'Heat Transfer Characteristics of rotating and axial flow between concentric
cylinders', Trans ASME, Jan 1958, pp.79-89.
[2] Hayase, T., Humphery, J.A.C., Greif, R. : Numerical calculation of convective heat transfer
between rotating coaxial cylinders with periodically embedded cavities, Journal of Heat Transfer
Transaction of the ASME, Vol 114, Aug 1992, pp 589-597
The Laminar to Turbulent limit is the Reynolds number at which flow is fully turbulent.
The Non-Vortex to Vortex limit is the Reynolds number at which flow is fully vortex.
This editor provides the settings pages for the loss calculation methods used in Motor-CAD.
See:
General Losses
Bearing Losses
Proximity Losses
End Winding Losses - the stator end winding losses can either be split evenly between front and
rear of machine or split in relation to the endwinding copper volumes. See also Stator Copper Loss
Distribution
Iron Loss Flux to Power Ratio - Iron loss flux density raise to power coefficient - used in
formulation given in Data for Loss Variation with Temperature & Load for calculating the variation
in copper & iron loss with temperature load. Used in conjunction with Magnet Temp Coeff Br to
find iron loss at any temperature. Only used with BPM-Therm, BPMOR-Therm and PMDC-Therm
motors.
The bearing loss interface allows the user to specify the using the Location option whether the
losses are split between the inner and outer race or placed on the central bearing node. See Bearing
Loss Location
When the automatic calculation method is used then the bearing loss values for different
temperatures and speeds must be input in the table.
The graph then shows the losses entered for the different speeds and temperatures.
The calculation speed value allows the user to see the points calculated for a certain speed. This
can be useful for checking that the bilinear interpolation is producing a good result.
The loss multiplier value allows the losses input in the table to be scaled without having to re-enter
each loss value.
The windage losses are the losses produced by the fluid movement in the airgap.
If the direct user input is selected then the user must input the windage loss in the Losses [Input
Data Editor]
A calculated value of the windage loss is however also calculated and displayed in Losses [Output
Data]
If the automatic calculation method is used then the windage losses are calculated automatically.
This is most useful when carrying out a transient calculation as some applications have viscous
fluids in the airgap and this fluid tends to have a much larger viscosity at low temperatures. This
makes it necessary to calculate the variation in windage with temperature.
Windage Calculation:
The temperature coefficient of resistivity are used to adjust the stray load loss of induction machines
based on the temperature of the components.
The values for Iron and Copper Stray Load Losses are (for an Induction Machine) in the data given in
Losses [Input Data Editor]:
Rules of thumb regarding which nodes the iron loss should be assigned are:
induction motor - ¼ to stator tooth, ¾ to rotor tooth
slip ring motor - ¼ to stator tooth, ¼ to rotor, ½ to internal air
The user can always set the Stray Load Losses in Losses [Input Data Editor] to zero and increase
the other loss components. In this way the user has full control of where the stray load loss is
distributed.
Proximity losses are commonly taken account of in transformer design but are often neglected in
electric machines. The proximity effect is the increase in ohmic loss due to high frequency currents
in adjacent conductors. These give rise to eddy currents which cause a non uniform current
distribution over the conductors cross section. The skin effect leads to a increase in the ac
resistance and so and increase in the copper loss. The increase can be surprising large in high
speed machines, values of Rac/Rdc of up to 17 being reported in machines with rating of around
12000rpm [1], [2].
FEA analysis such that of Flux (www.cedrat.com) should be used to calculate the proximity effects:
Motor-CAD has several different methods for modeling the effect of proximity losses.
The added loss can be accounted for in the Motor-CAD thermal model by:
firstly increasing the level of copper loss using a Proximity Loss Model
secondly by using the Winding Temp Variation with Proximity Loss Distribution factors to
model the tendency for the proximity effect to concentrate the loss towards the centre of a
slot.
or
using the uneven distribution of proximity losses within the winding layers to place the
proximity losses in the slot
Variation with Speed - the proximity losses are input using the Losses [input Data Editor]
as shown below:
Variation with Speed and Load - the proximity losses vary with speed and load based on
the variation of winding resistance with frequency. If have known winding resistance at high
frequency then can put in the ratio of (winding resistance at high speed / winding resistance at
0 speed) .
Motor-CAD will then calculate the proximity losses from 0 proximity losses at 0rpm to (Rac/Rdc
– 1) x dc copper losses at the specified speed.
To illustrate by example for a machine with 120W copper losses and Rac/Rdc = 2 at 6000 rpm:
The screenshot below shows how the proximity losses vary with speed (starting at the origin and
passing through 6000rpm, 120W):
The Uneven distribution of Proximity Losses option models the position of the proximity losses in the
slot.
Proximity Losses tend to not only increase the amount of copper loss but also concentrate the loss
towards the centre of a slot. This second effect is accounted for using the Uneven distribution of
Proximity Losses option.
When not enabled the proximity losses (100W in this case) are spread evenly throughout the slot as
shown below:
Uneven distribution of Proximity Losses with a value of 0.5 will put the proximity losses in the centre
of the slot as shown below:
Uneven distribution of Proximity Losses with a value of 0.1 will put the proximity losses towards the
outside of the slot:
Uneven distribution of Proximity Losses with a value of 0.95 will put the proximity losses towards the
centre of the slot:
The Winding Temp Variation with Proximity Loss Distribution option is an older model and
suberceded by the Variation with Speed and Load Model.
The concentration of the proximity losses at the centre of a slot leads to a larger winding hotspot
compared to the case where the same loss was evenly distributed across the winding cross section.
Calibration of the model can be performed by carrying out Finite Element Analysis (FEA) on two
models, one with the loss evenly distributed and the other with the effect of proximity loss taken into
account. Previous FEA seems to indicate that the percentage temperature increase is a linear
function of speed (hence the linear model proposed) - although more work is required on this
subject.
In the model the thermal resistances associated with the winding active section are artificially
increased by the factor:
The resistance multiplier is given in the Losses [Output Data] as the Proximity Winding Resistance
Multiplier parameter.
[1] Rafal Wrobel, Phil Mellor, Neville McNeil, 'Analysis of Proximity Losses in a Brushless
Permanent Magnet Motor, ISEF 2005, Baiona, Spain, 15-17 September, 2005
[2] Jack McCabe, Beating the Heat: Temperature Control of a High-Performance Spindle, MMS
Online, www.mmsonline.com/articles/120104.html
The stator copper losses are by default automatically split between the active and end winding
sections based on the length of the active section and end turns.
The endwinding losses are then split between the front and rear endwindings based on the
endwinding volumes shown in the axial cross section editor.
The stator copper losses location and Endwinding losses location options allow the split of the
copper losses to be specified by the user if required.
The rotor cage losses distribution can be either specified by the user or calculated automatically
using the formula shown below.
The Induction machine rotor cage losses are distributed in proportion with the rotor cage resistance
values.
For the automatic calculation method the Rotor Cage Resistance calculations are based on the
following:
{Inputs}
RhoBar := resistivity of the bar material [ohm/m]
RhoER1 := resisitivity of the end-ring left material [ohm/m]
RhoER2 := resisitivity of the end-ring right material [ohm/m]
TCCBar := thermal coefficient for the bar material at 20C [1/C]
TCCER1 := thermal coefficient for the left end-ring ER1 material at 20C [1/C]
TCCER2 := thermal coefficient for the right end-ring ER2 material at 20C [1/C]
NOTE: In MotorCAD we don't know directly kw1 - fundamental factor of the winding. However, all
components are multiplied with Sqr(Turns_phase *kw1) so we can use the ratio of Rbar/Rendring
where Sqr(T_ph*kw1) will vanish.
8.22.9 Settings
8.22.9.1 Shaft Speed [RPM]
Note: The shaft speed has implications on the forced convection model when a shaft mounted fan is
used.
Stall:
It is possible to model a complete stall (Shaft Speed [RPM] = 0) using the Motor-CAD BPM-Therm
module - see Stall Model for details. Under a worst case stall (3-phase sinewave star connected
motor) - one phase will carry i amps and the other two phases with carry i/2 amps. Thus, one phase
will have 4 times the loss of the other two.
See Cooling Options [Input Data Editor], Losses [Input Data Editor] and Forced Convection Data
[Input Data Editor].
The duct wall friction resistance R is calculated using the following equation:
R=k / (2 A2)
where:
k = k factor
fluid density
A = channel cross section area
k = fL/D
where:
f = friction factor
L= duct length
D = hydraulic diameter
In Motor-CAD we use the Colebrook-White equation [1] below which forms a good estimate
of the turbulent flow area of the Moody chart:
where:
k = surface roughness
D = hydraulic diameter
Re = Reynolds number
Allow editing of radiation view factor: The option to allow the editing of the radiation view factor is
given here. When the tick box is selected then the radiation view factor shown in the Radiation Data
[Input Data Editor] table can be edited.
Automatic File Backup: When this option is enabled then the previous version of the mot file is
saved in a backup folder located in the current folder/backup the files are labeled sequentially
filename_1.mot, filename_2.mot etc.
The variation with temperature options allow the user to select whether the heat transfer coefficient
scaling will be done for a fixed fluid temperature specified by the user or to be calculated based on
the front and rear endspace temperatures. Note: Using the Variable - Calculated @ Endspace Node
Temperatures option may result is long transient calculation times due to non linearity in the model.
See k1, k2, k3 When not air for full details of the heat transfer coefficient scaling calculation.
Default Correlation
Pull down selection box for choosing which authors correlation data to base the heat transfer
calculation on. The default is the Schubert [EW] model - see End Space Convection Correlation
The cooling of these surfaces is taken to the ambient node in the model.
This cooling is not inside the end space however the endspace correlation suits it well as it allows
for both natural and forced convection to be taken into account for the cooling of the surfaces of the
rotating shaft.
Before Motor-CAD v7 this cooling was not considered. This cooling should normally be included in
the model but can be disabled if required.
This cooling model can be enabled and disabled in the End Space [Settings] page.
When enabled the extra fields are shown in the end space editor as shown below:
The cooling paths from the shaft to the ambient node can be seen in the circuit editor as shown
below:
The calculated values can also be seen in the output data sheets:
See also Rt [Shaft - Amb (Front)] and Rt [Shaft - Amb (Rear)] in Miscellaneous Settings.
Options for the setting of the convergence criteria for the solver are given here. These do not normally
need to be changed from the default settings but there are occasions for complex problems where
they may need to be altered.
Transient Convergence
Duty cycle point insertion - a point will automatically be added to the duty cycle when the speed
varies by the value specified.
Transient Iterations - This is useful for scripting to allow the script to stop the transient calculation
if the specified max number of iterations is exceeded.
End space fluid flow - this is seldom used but allows the max rate of change of the endspace fluid
to be limited to help solving.
Sometimes convergence is difficult to achieve. This seems to be mainly if the ambient and inlet
temperature of through ventilation or liquid cooling are very different. A warning is given of the error
and which node if this is the case.
Steady State Max Iterations: The maximum number of iterations to allow for reaching steady state.
Steady State Min Iterations: The minimum number of iterations to be run before allowiing
convergence at reaching steady state.
If convergence is a particular problem it may be better to use the option to base convergence on an
allowable temperature difference of nodes between iterations and set a lower value for this figure
using the Steady State Max Convergence Error [dT] parameter.
The number of iterations taken to reach steady state and the convergence error is shown in the
Miscellaneous Output Data sheet (Other) .
Steady State iterations before averaging: After the specified number of iterations if convergence has
not been reached then results are average to try to aid convergence.
The transient convergence settings allow the convergence criteria to be modified. It is not normally
necessary to modify the default values, however for some more unstable models adjusting these
values can help to get the transient to solve correctly.
The convergence succeeds when solver error < Absolute Error + Relative Error x MaxDt
where:
MaxdT = maximum temperature difference for each node from previous iteration.
If there are large variations of speed during the duty cycle then it is sometimes necessary to add
extra calculation points during the changes of speed.
These points can be inserted automatically using the Duty Cycle Point Insertion option.
If the model fails to converge then the Transient Iterations option allows the user to specify the point
at which the calculation will be aborted.
The England 4th of 5th order method is the default Runge-Kutta embedded formula used in Motor-
CAD. This does not normally need to be changed.
Runge-Kutta 2nd/3rd order methods and Prince-Domand 4th/5th order methods are also available if
required.
The tempererature variation option enable within Integration Step allows the temperature variation of
properties to be disabled between each point calculated. Normally the fluid properties are taken into
account during the solving between data points. When disabled this means that the fluid properties
are only calculated once for each data point and between data points the fluid properties are fixed at
the value calculated. This can help significantly for convergence problems. However you must ensure
that the data points are close enough together so that the fluid properties do not change significantly
between points.
Transient Solver
2 different transient solvers are available:
Ordinary Differential Equations - this is the GEAR method that was originally used in Motor-
CAD. It is a 1 step solver.
Advantages:
1. should give accurate results.
Disadvantages:
1. can be unstable.
2. need to specify the calculation points in duty cycle.
3. slow to solve if have small capacitances.
4. Will not solve if any nodes have no capacitance.
Differential Algebraic Equations (DAE) - this is a new method introduced in Motor-CAD based
on numerical differentiation formulas. It is a multi step solver.
Advantages:
1. Faster solving than GEAR. Especially if have small capacitances (stiff equations).
2. More stable than GEAR.
3. Will solve with smaller number of specified points as is multi step solver and will vary step size to
converge.
Disadvantages:
1. May be less accurate than GEAR.
2. Unstable with some fluid cooling methods. Note fluid paths method is always used for the DAE
solver. (see Fluid Temperature Calculation method)
The compromise is between stability of the solution and solving time. This type of change is not
needed if the solver is stable and solver instability is easily seen in the results by large variations in
the temperatures.
Fluid flowing through the machine is heated up or cooled down as it passes through the machine.
The different methods of calculating the fluid temperatures and linking them with the thermal circuit
are described below:
Length Segmentation:
This method splits the channels into 10 segments For each segment the inlet temperature is known
and the outlet temperature is calculated. The inlet for the (n+1)th segment is the outlet of the (n)th
segment. This is good for the temperature calculation when the flow rates are low so that the
temperature of the fluid nearly reaches that of the surrounding duct. The fluid nodes in the model
have power injected to match the average temperature calculated.
Fluid Paths:
This is the latest method. In this method an equivalent thermal resistance is calculated based on the
fluid properties and the flow rate. The thermal resistance calculated is put directly into the solver
circuit and will give a more robust solution. The disadvantage of this method is that the fluid node
temperatures used in the thermal circuit will be the outlet temperatures rather than the average
temperatures so may give a higher machine temperature.
a fluid. See Cooling Options for more details of the cooling options available.
1. Original Method adding power sources. In order to fix the temperature of a node the power
input/output to/from the node is varied within the solver to maintain the required temperature.
2. Equation adjustment method. This is the new method for fixing temperatures in the model and
is the preferred method. The equations are adjusted to ensure that the node temperature is fixed.
This has the advantage that there is 1 less equation to solve for every fixed temperature and so
the solution is faster to converge.
set manually.
Capapitance Adjustment, Transient Speed Increase Factor
Data Logging
Transient Results File Analysis
Special integration techniques are used in Motor-CAD to give stable calculations in circuits having
vastly different sizes of thermal capacitance (termed a stiff set of equations). The problem is that
the thermal transient calculation speed is critically related to the component with the lowest thermal
capacitance and some nodes within the model have very low capacitance values, i.e. the end-space
air nodes. If we were to use such low thermal capacitances in the analysis then unrealistically long
calculation periods may result. There are two general techniques that can be used to speed up the
calculation speeds - both however lead to slight inaccuracies:
1. Artificially increase the thermal capacitance of nodes that have very low thermal
capacitance. This is the techniques used in Motor-CAD at present when [Transient Fluid
Nodes] = [Include Capacitance] (default). More details are given in Transient Speed Increase
Factor.
2. Make all nodes that have low thermal capacitance arithmetic nodes with fixed temperature
over the integration time period. This has been implemented for transient calculations with a
fluid cooling system (wet rotor, water jacket, through ventilation) if the [Transient Fluid Nodes]
= [Neglect Capacitance]. If selected then any heating of fluid nodes with time is neglected.
This may give a faster solution and in some cases may be more accurate as the capacitance
effects of the fluid should only be substantial if the flow rate is small.
The above radio button selector is used to decide which method is used.
Adjustment factor used to try to increase the transient calculation speed. It works by artificially
increasing the thermal capacitances of nodes that have low values (i.e. Cap [End Space (F)], Cap
[End Space (R)], etc). The transient calculation speed is related to the component with the lowest
thermal capacitance, i.e. usually the end space air. It has been found that this can be increased
significantly without altering the transient accuracy significantly.
When Transient Speed Increase Factor <> 0 [Default = 1], any node that has a capacitance less
than a Critical Capacitance value is artificially increased according to the following:
IM, SYNC & CLAW Machines: Rotor Critical Capacitance = MIN(Cap Rotor Tooth,
Cap Rotor Copper) x 0.1
Combine inner two winding layers if smallest layer volume is less than 10% of the next
smallest
Winding Layers = 1 - Winding Critical Capacitance = 0.1 x Cap Winding Inner Layer
Winding Layers = 2 - Winding Critical Capacitance = 0.2 x Cap Winding Inner Layer
Winding Layers = 3 - Winding Critical Capacitance = 0.3 x Cap Winding Inner Layer
Winding Layers = 4 - Winding Critical Capacitance = 0.4 x Cap Winding Inner Layer
Winding Layers = 5 - Winding Critical Capacitance = 0.5 x Cap Winding Inner Layer
Winding Layers = 6 - Winding Critical Capacitance = 0.6 x Cap Winding Inner Layer
Winding Layers = 7 - Winding Critical Capacitance = 0.7 x Cap Winding Inner Layer
Winding Layers = 8 - Winding Critical Capacitance = 0.8 x Cap Winding Inner Layer
Winding Layers = 9 - Winding Critical Capacitance = 0.9 x Cap Winding Inner Layer
Winding Layers > 9 - Winding Critical Capacitance = 1.0 x Cap Winding Inner Layer
Critical Capacitance:
1st term makes sure that if [Transient Speed Increase Factor] = 0 then the
actual values of Nodal Capacitance are used at each node
During the transient calculation useful integration information (step size, number of integrations
steps, etc) is given on the status line:
This is especially useful if the transient calculation is very slow or if the calculation fails.
To help judge the limit on [Transient Speed Increase Factor] a table showing the values of nodal
capacitance before and after the speed increase algorithm is implemented is displayed in Speed
Increase Data when a Transient Graph is calculated. The information given includes:
critical capacitance values and number of nodes that have been increased
It is best to not use such a large value of [Transient Speed Increase Factor] that unrealistic
capacitances are used. You can always look at the variation in transient graph with varying [
Transient Speed Increase Factor]
This allows for other parameters to be stored in a file during the transient analysis and also viewed in
a graph if required.
When this option is enabled then before a transient analysis is run the options window below is
displayed:
The values are stored in the text file and can also be viewed as shown below:
Transient results can be saved to file by enabling the transient results file option in Default File
Locations as shown below:
When this option is enabled then whenever a transient calculation is made the results will
automatically be stored in the file.
This file can then be view in Motor-CAD by selecting the file as shown below:
The temperatures, power values and capacitances in the file can then be viewed in Motor-CAD in
Schematic, Raidal, Axial views, Circuit Editor and the Transient Graphs. The time period to view is
selected using the arrow buttons as shown below:
Caution: The file only holds temperatures, capacitances and power sources. The resistance values
and power flow in the schematic have still got the previous values from the last calculation so may
not be valid for the transient results.
The airgap thermal resistance is calculated using either a heat transfer model based on convection
or conduction only - choice depends upon Airgap Model selected. It is suggested that peripheral
speed induced convection heat transfer can enhance conduction heat transfer by 0% to 100% [1-3].
It has been found that slotting increases gap heat transfer by around 10% due to added air
disturbances [1,2] - when the motor is running. However, at stall the conduction heat transfer is
reduced due to the extended gap [4]. The Airgap Conductivity Multiplier parameter is used to enable
the user to adjust the level of airgap conductivity to accommodate such affects.
The present thermal resistance model assumes that the heat transfer is by conduction only. The
Encoder - EncCase Conductivity Multiplier parameter is used to enable the user to adjust the level of
conductivity to accommodate convection and radiation enhancements. See also Airgap Conductivity
Multiplier.
Weighting of natural convection fin side/fin base models used in axial-fin horizontally mounted fin
model. See Natural Convection Correlation
Thermal resistance between front of shaft and connected device [ambient]. See also External Shaft
Cooling.
Thermal resistance between rear of shaft and connected device [ambient]. See also External Shaft
Cooling.
Altitude
Temperature
Editors for managing and setting the Solids material database data, i.e.:
The solid material data is used in the Materials & Weights [Input Data Editor] and
Editors for managing and setting the Fluids material database data i.e.:
Editor for setting the temperature and altitude at which the air property data in conduction (no
convection) regions of the model is calculated.
Steady-state & transient performance calculations can be made with constant air property data
(calculated at Air Property Temperature) or fully variable properties. The thermal conductivity of air
increases by around 25% for a 100C increase in temperature. The disadvantage of allowing the air
properties to vary with nodal temperature is an increased calculation time - see Transient Progress
Form.
A indication of the change is property values with temperature is given in the two tables shown above
- i.e. one at Ambient Temperature and one at Air Property Temperature.
A radio button selector is provided to set the option of whether to use a fixed temperature at which
the air properties are calculated. This option can be used to speed up the calculation process at the
The variation in the following air properties with temperature are modelled within the program:
Steady-state & transient performance calculations can be made with constant air property data
(calculated at Air Property Temperature) or fully variable properties - see Air Property Options. It is
noted that the thermal conductivity of air increases by around 25% for a 100C increase in
temperature. The disadvantage of allowing the air properties to vary with nodal temperature is an
increased calculation time - see Transient Progress Form.
SI units are listed above, however, British units are also available via the Units Selection Form.
The standard atmosphere is based on mathematical formula that reduce temperature and pressure
by certain amounts as altitude is gained
Editor for setting the altitude at which the following air parameters are calculated:
Air Temperature [US Standard Atmosphere]
Air Pressure [US Standard Atmosphere]
Air Density [US Standard Atmosphere]
The standard atmosphere is based on mathematical formula that reduce temperature and pressure
by certain amounts as altitude is gained. The results are close to averages of balloon and airplane
measurements at various altitudes. The variation in pressure with altitude is shown below:
The pressure used in any air convection calculations is set according to that calculated using the
US Standard Atmosphere. However, the Ambient Temperature is not directly set equal to that
calculated by the standards - this is because the motor may a completely different local ambient.
Thus the local air density is not equal to that in the standard atmosphere but equal to that calculated
using the following formula:
= P / (TR)
The user can also adjust the values of Ambient Temperature and Air Property Temperature to match
that given in Air Temperature [US Standard Atmosphere] if required (if the local ambient is equal to
the global ambient at the particular altitude).
The Altitude is also set in Natural Convection [Input Data Editor], Forced Convection [Input Data
Editor] or Calculation Options [Input Data Editor].
The solids database contains material data for solid materials typically used in electric machines, i..
e electrical properties, magnetic properties, loss properties, thermal conductivity, specific heat and
density. The information is stored in the database file in a text format so can be opened and modified
in a text editor if required. The database properties for the materials that are displayed depends on
the whether the electromagnetic or thermal context is visible. The thermal context view is shown
above. The electromagnetic context view is shown below.
The default database is stored in the location specified in the Default File Locations.
When Motor-CAD is started if the Solids.mdb file is not in existence then it is created and a copy of
several standard materials are added to it. Care should be taken if using these values to ensure that
these values are accurate for the materials being used.
The database can be updated to have the latest default materials from Motor-CAD by selecting the
Create new Database option and then selecting the existing open database.
The values in the database can be modified directly in the table. The database is saved
automatically whenever an entry has changed.
Adding a new material entry is done by entering the details on the bottom row of the table (where the
* is).
After the data has been put into the table and the user comes back to the editor it will be seen that
the new material has been put in alphabetical order:
The user can select the new material in the Materials & Weights [Input Data Editor]:
Deletion of a material is done by selecting the material in the table and then clicking on the Delete
Solid button.
The default materials database can be created by clicking on the Create New Database button.
For more details of setting the magnetic material properties also Solids Database -
Electromagnetics
The solids database holds the property details for the different materials. There is a large set of
default materials provided in the database and new materials can be added as required.
Now that we have our steel, we need to enter in some information about the steel, such as
Thermal Conductivity, specific heat capacity and density:
Next we need to add BH curve values. To do this, change to the Steel BH tab, and select the
show ‘All materials’ as shown below. Note: the steel will not appear in the steel list until it
has some BH values.
We can see that our material has no data. We can enter our data by adding BH Data Points in
one at a time:
Once we have done this, we can switch back to steel materials and see that our material is
now listed as a steel:
Steel Losses
First of all we switch to the Steel Properties tab inside the Steel Losses tab as shown below.
We can see that there is no loss coefficients, as expected.
This particular steel has a lamination thickness of 0.65mm, so we enter this, as well as a
standard hysteresis loss value of 0.65:
The Steinmetz Coefficients are by default all 0. We can now calculate some new values from
the Iron Loss Coefficients tab.
We can either add our data one row at a time to the table, or we can simply copy and paste
the data from a spread sheet into the table. The loss density values should be results from a
datasheet or from test for the particular steel:
Clicking the Update Graph button will then display the graph of values. In the graph below
there are 2 data series plotted. Actual Data is the data provided in the table. The Current
Model series is the loss density values calculated using the current iron loss coefficients. At this
point the iron loss coefficients have not been set so there is not match with the Actual Data.
We can then find the Iron Loss Coefficients by clicking the ‘Find Coefficients’ button. Note: in
this case we are using the Steinmetz method.
Once the coefficients are found, they are used to plot a curve with the data, and we can see
how good the plot is. We also have a message displaying the values of our coefficients (seen
on the right of the screenshot).
If we are happy with these values, we can click the ‘Update Database values’ to add these
values to our solids database for later use, and save the database values in the .mot file. We
can also see these values have been updated in the ‘Steel Properties’ tab:
Bertotti
For Bertotti coefficient calculations we must also make sure that we have values for the
Electric Resistivity before trying to calculate the iron loss coefficients.
The calculation of the coefficients and the updating of the solids database is done in the same
way as for the Steinmetz coefficients.
The method extends just as easily to multiple frequency data. The plots below show the loss
density values for different frequencies for the datasheet data and the model data.
Above we have data across 5 different frequencies. As before, we can calculate and fit curves
to this data using our curve fitting algorithm:
We see that the curves fit very well, as in the single frequency case, all with the same loss
coefficients.
As before, if we are happy with the calculated values, we can add them to our database by
clicking the Update Database values button.
See also Solids Database - Electromagnetics, Iron Loss Calculations and Iron Loss Coefficients.
The fluid database allows the full characterisation of fluid properties with temperature.
The fluid database contains material data for fluid materials typically used in electric machines, i..e
thermal conductivity, specific heat, density and Kinematic Viscosity. The information is stored in the
database file in the form shown below:
The default database is stored in the location specified in the Default File Locations.
When Motor-CAD is first run there are several fluids in the default database. Care should be taken if
using these values to ensure that these values are accurate for the fluids being used.
The values in the database can be modified directly in the table. The database is saved
automatically whenever an entry has changed.
Adding a new fluid entry is done by entering the details on the bottom row of the table.
Then add the fluid conductivity, specific heat, density and viscosity properties to the table on the left
of the screen as shown below.
The user can select the a fluid from the database for the internal, external, wet rotor, housing water
jacket, rotor water jacket, slot water jacket, shaft spiral groove and spray cooling. The fluid
properties used in the calculation will then be automatically used for the fluid temperature.
9 Electromagnetics
The Electromagnetic module options are visible when there is a valid BPM-EMag module licence and
the E-Magnetic model type is selected.
The E-Magnetic module allows the electromagnetic performance and loss calculation of a machine
to be modeled.
9.1 Control
The control page is only visible in the electromagnetic context this is set using the model type
option.
Current Definition - define the current in terms of peak current, RMS current or RMS current
density in the slot.
Phase advance (in electrical degrees) of the phase current waveform to the back EMF waveform.
DC Bus voltage - the voltage of the DC supply to the drive . Motor-CAD will check to ensure have
sufficient supply voltage for the specified speed and current.
Drive Mode - the machine can be driven with sinusoidal, square wave or custom current waveforms
Stator Skew - stator skew is taken into account when calculating the torque waveforms and mean
lenght per turn (MLT) of the windings.
The temperatures options below are not editable when the E-Magnetic + Thermal model is enabled.
In this case the values shown are calculated by Motor-CAD thermal model from the loss values
calculated in the electromagnetic model.
Winding Temperature - this temperature is used to calculate the winding electrical resistivity that is
used in the calculation of the copper losses
Magnet Temperature - this temperature is used to calculate the effective magnet remanence
Stator lamination Temperature - this temperature is used for calculating the electrical resistivity of
the stator laminations that is used in the stator eddy current loss calculation.
Rotor lamination Temperature - this temperature is used for calculating the electrical resistivity of the
rotor laminations that is used in the rotor eddy current loss calculation.
Stator sleeve Temperature - this temperature is used for calculating the electrical resistivity of the
stator sleeve that is used in the sleeve loss calculation.
Rotor banding Temperature - this temperature is used for calculating the electrical resistivity of the
rotor banding that is used in the banding loss calculation.
Performance Tests
The tests to be performed are selected using the checkboxes shown below.
Open Circuit:
Back EMF - the machine is rotated through 360 electrical degrees with no current to calculate the
voltage induced in the windings. Note: this uses a multi-static FE solution
Cogging Torque - the machine is rotated through a multiple of the cogging torque period with no
current and dense airgap mesh to calculate the cogging torque of the machine. Note: this uses a
multi-static FE solution
Loss Calculation (open circuit) - the machine is rotated through 360 electrical degrees with no
current to calculate the losses induced in the machine. Note: this uses a transient FE solver.
On Load:
Torque - the machine is rotated through 360 electrical degrees with a drive current to calculate the
torque waveforms of the machine. Note: this uses a multi-static FE solution so does not calculate
magnet or iron losses.
Loss Calculation (on load) - the machine is rotated through 360 electrical degrees with drive currents
to calculate the losses in magnet, stator and rotor iron (also banding and sleeve if present). Torque
waveforms of the machine are also calculated. Note: this uses a transient FE solution.
Torque Speed Curve - the machine performance for different angle of phase advance is calculated to
give torque speed and power speed envelopes.
The model is automatically solved when the data has been changed by viewing the results sheets.
The model solved can also be started by clicking on the solve button shown below.
When the electromagnetic + thermal model is has been selected then the Solve E-Magnetic and
Thermal Model button will also be enabled. When this is clicked then the combined
electromagnetic and thermal model is solved.
During the solving the status bar at the bottom of the Motor-CAD window will indicate the progress of
the calculation. The calculation can be stopped at any time by clicking on the Cancel Solving
button.
The Back EMF waveforms are calculated when the Back EMF performance test option is selected in
the control page.
Back EMF waveforms can be customised using the Back EMF graph options in the Graphs settings
page.
The cogging torque waveforms are calculated when the Cogging Torque performance test option is
selected in the control page.
Cogging torque waveforms using Co-Energy and Virtual works methods are calculated. See torque
calculation methods for details.
Cogging torque waveforms can be customised using the cogging torque graph options in the Graphs
settings page.
Open circuit losses calculated when the Open Circuit losses option is selected in the control page.
The open circuit loss shading results can be viewed by opening the loss result file using the option
highlighted above.
Different shading options for total losses, hysterisis losses, eddy current losses and magnet losses
can be selected.
The loss values calculated are shown in the losses output data sheet.
On Load losses are calculated when the On Load losses option is selected in the control page.
The On Load loss shading results can be viewed by opening the loss result file using the option
highlighted above.
Different shading options for total losses, hysterisis losses, eddy current losses and magnet losses
can be selected.
The loss values calculated are shown in the losses output data sheet.
9.6 Torque
The torque waveforms are calculated when the Torque or On Load Loss performance test option is
selected in the control page.
The FE Torque waveform is the average of torque calculated using Maxwell stress tensor and Virtual
works methods.
Flux linkage torque, reluctance and alignment torques are also displayed.
See torque calculation methods for details of the methods used to calculate the torque values.
Torque waveforms can be customised using the torque graph options in the Graphs settings page.
The torque speed curves are calculated when the Torque Speed performance test option is selected
in the control page.
The torque speed curves show the operating range of the machine based on the limitation of the
supply voltage.
The Torque speed curve can be customised using the torque speed graph options in the Graphs
settings page.
The power speed curves are calculated when the Torque Speed performance test option is selected
in the control page.
The power speed curves show the operating range of the machine based on the limitation of the
supply voltage.
The power speed curve can be customised using the torque speed graph options in the Graphs
settings page.
When the E-Magnetic(interface) + Thermal model type is selected then the control page has the
option to Solve E-Magnetic and Thermal Model enabled and the temperature edit boxes are disabled
as shown above.
After clicking on the Solve E-Magnetic and Thermal Model button the interface shown below will be
displayed to right hand of the screen.
Allowable convergence error [%] - convergence is assumed when winding temperature, magnet
temperatue and copper losses have changed by less than this percentage specified.
Maximum Number Iterations - this is the maximum number of iterations to run before stopping if it
has not converged.
The text box is updated during the solving showing the progress as shown below:
Can see in the text box above that the Average torque calculated has reduced from 13.9Nm with
stator winding temperature of 71.1C and magnet temperature of 70.2C to 13.2Nm with stator winding
temperature of 129.1C and Magnet temperature of 123C.
The final temperature are shown in the main user interface as shown below:
The electromagnetic results, losses and material properties for this machine operating at this
temperature can be viewed in the output data sheets.
Similarly the thermal results, losses and temperatures can be viewed by changing the model type to
E-Magnetic + thermal(interface).
When the Custom Drive Mode option has been selected in the Control page then this option will be
enabled.
The custom drive cycle file needs to have 4 columns separated by tabs or spaces.
Column 1 = the electrical angle of phase 1
Column 2 = the current in phase 1
Column 3 = the current in phase 2
Column 4 = the current in phase 3
There can be as many points in the waveform as required however the final point must be a multiple
of 360.
The phase currents are scaled to have a peak value equal to that of the Peak Current specified in
the Control page. The waveforms produced are shown below.
The waveform can over multiple electrical cycles allowing different startup and fault conditions to be
modeled.
Where
Where:
Iron loss data from steel suppliers is almost always for a sinewave current, and may be
characterized by the equation (1) with separate terms for hysteresis, classical and excess loss:
(1)
Kh , α and Kexc are determined using curve fitting techniques in Motor-CAD from one or two specific
core losses curves.
Another approach is to characterise the iron-loss variation by the equation (2) with separate terms for
hysteresis and eddy-current:
(2)
Kh , α, β and Keddy are to be determined using curve fitting techniques in Motor-CAD from three or
more specific core losses curves.
NOTE: Both Bertotti and Steinmetz methods make use a loss coefficient for minor hysteresis loops,
KC that takes into account the non-sinusoidal supply. This coefficient cannot be calculated as it’s
more an empirical value. Therefore based on previous studies it is set to 0.65 by default. The user
can change this coefficient. In extracting the other loss coefficients we neglect this.
[1]. G. Bertotti, “General properties of power losses in soft ferromagnetic materials,” IEEE Trans. on
MAG, vol. 24, no. 1, pp. 621–630, Jan. 1988
[2]. D.M., Ionel, M. Popescu, M.; MI, McGilp, TJE Miller, S.J. Dellinger, R.J. Heidemann [2007]
“Computation of Core Losses in Electrical Machines Using Improved Models for Laminated Steel”,
IEEE Transactions on Industry Appl. – Vol. 43, No. 6, Nov/Dec 2007 – pp. 1554-1564
[3]. Allen N, Lai HC, Rodger D and Leonard PJ, IEEE Trans., Vol. MAG-34, No. 5, September 1998,
pp. 2535-2538
[4]. Hameyer K and Belmans R, Numerical modelling and design of electrical machines and devices,
WIT Press, 1999
[5]. Salon S.J. Finite element analysis of electrical machines, Norwell, MA: Kluwer Press, 1995
Iron loss Coefficients for Steinmetz and Bertotti iron loss calculation methods are stored in the
Motor-CAD solids database.
Motor-CAD has a coefficient calculator to allow the iron loss coefficients to be calculated from new
material data and the existing coefficients to be verified.
For details of adding new iron loss coefficients and the automatic calculation method see Adding
Steel data to the solids database.
(1)
Eddy-currents are assumed to flow in conducting regions with conductivity σ. Such regions include
certain types of permanent magnet, for which Br is non-zero, the remanent flux-density vector.
Eddy-currents also flow in conducting cans and any solid conducting components such as the rotor
shaft. For such unmagnetized components, Br = 0.
The term σ(v curl A) represents motion-induced eddy-currents which arise only on the rotor. The
local linear velocity v = ω r, where r is the radius and ω is the angular velocity. In regions
representing conducting components on the stator, this term is omitted.
The induced eddy-current density can be deduced from the time variation of the magnetic vector
potential A :
(2)
Consequently, the power loss/volume density is
(3)
Two-dimensional finite-element solutions must satisfy the zero net current condition
(4)
The 3D end-effect and axial segmentation of the magnets is considered via a 3D/2D factor:
(5)
[1]. Ruoho et al, Modeling Magnet Length In 2-D Finite-Element Analysis of Electric Machines,
IEEE Trans. On Magnetics, 2009
Motor-CAD uses several methods to estimate the electromagnetic torque. All these methods are
based on the solution obtained from embedded 2D finite-element solvers (magneto-static or
transient) and are described below:
(1)
In the two-dimensional model of an electrical machine, the three-dimensional surface integral in (1)
may be replaced by a volume integral over a hollow shell in free space surrounding the moving rotor.
As the true torque is independent of the radius, we obtain by integrating the expression (2):
(2)
(3)
where We represents the magnetic energy in an element, and Ai represents the magnitude of the
vector potential at each node of the element
(4)
where the dq flux linkages yd;q and currents id;q have been determined at each rotor position by
applying the Park transformation to the phase quantities.
(5)
where w’ in principle consists of three components: soft-magnetic materials (i.e. steel laminations),
hard-magnetic material (i.e. permanent magnets) and air.
(5)
(6)
Note that for air regions, the co-energy is equal to the energy.
(7)
In Fig. 1 the magnetic energy of a soft-magnetic material is illustrated as vertical hatched area. In
hard magnetic materials the magnetic energy density w’m is computed as (vertical hatched area in
Fig. 1b):
(8)
(9)
where the index 1 and 2 refer to two consecutive solutions.
The virtual work method implementations previously described need an energy or co-energy
calculation as accurate as possible. Therefore it is necessary to refine the discretization on the
entire analysed area, especially in those areas where there are great changes in the magnetic flux
density and where deformation of the mesh elements occurs due to the special movement. When
the iron parts saturate it is important to have a magnetization curve of the material as close as
possible to the real curve. Every deviation between real magnetization curve and the implemented
one is integrated and this leads to important errors in the torque (force) calculation.
Fig 1. Magnetic energy for a soft-magnetic material (iron) (a) and for a permanent magnet with rigid
magnetization (b)
Loop torque
(10)
Where W Is the area of the flux-linkage vs current loop, m the number of phases, p the pole pairs.
1. S.J. Salon, Finite Elem ent Ana lysis of Electrica l Ma chines, Kluwer Academic Publishers, Boston, 1995, ISBN 0-7923-9594-8.
2. K.J. Binns, P.J Lawrenson, C.W. Trowbridge, The a na lytica l a nd num erica l solution of electric a nd m a g netic fields, John
Wiley and Sons, Chichester (UK), 1992, ISBN 0 471 92460 1.
3. J.L. Coulomb, G. Meunier, “Finite element implementation of virtual work principle for magnetic or electric force and
torque computation” – IEEE Tra nsa ctions on Ma g netics, Vol. MAG-20, No.5, Sept. 1984, pp. 1894-1896.
4. D.M. Ionel, M. Popescu, S. Dellinger, TJE Miller, M.I. McGilp: “Assessment of Pulsating Torque Components in Brushless
Permanent Magnet Machines through Numerical Analysis of the Electromagnetic Field”, Conf. Rec. IEEE-IA S 2 0 0 4, October
2004, Seattle, USA., Vol. 3, pp. 1715-1722
5. M. Popescu “Prediction of the Electromagnetic Torque in Synchronous Machines through Maxwell Stress Harmonic Filter
(HFT) Method” – Electrica l Eng ineering , Spring er, 2005
6. M. Popescu, D.M. Ionel, S. Dellinger, TJE Miller, M.I. McGilp: “Improved finite element computations of torque in brushless
permanent magnet motors”, IEE Proceeding s Electrica l Power A ppl.- Vol. 152, No.2, April 2005, pp. 271-276
7.TJE Miller, M. Popescu, C. Cossar, M.I. McGilp [2006]: “Performance Estimation of Interior Permanent-Magnet Brushless
Motors Using the Voltage-Driven Flux-MMF Diagram” – IEEE Tra nsa ctions on Ma g netics. – Vol. 42, No. 7, July 2006, pp. 1867-
1873
9.16 Harmonics
The FEA window allows the user to view the finite element solving of the electromagnetic model.
Note: viewing this solving window during solving may slow down the solver due to the time taken to
display the shading and contours.
Solve E-Magnetic model - this will start the solving and is equivalent to pressing the Solve E-
Magnetic model button on the Control page.
Shading - this specifies the shading function as region, flux density, vector potential, current
density or eddy current density.
Shading Function - this allows other shading functions to be specified.
Vectors - the vectors are displayed or hidden and the function can be specified
Debug - these debug options allow the geometry and mesh to be viewed if there is an error with the
geometry
Open - this allows the previous result files to be opened and played back
(1)
2. THD total harmonic distortion for EMF, voltage, current (square wave drive only).
Two formulas are used – one for the so called distortion factor:
(2)
(3)
(1)
If abs(Phi_V) > abs (Phi_I) we have leading power factor, if abs(Phi_V) < abs (Phi_I) we have
lagging power factor
2. Influence of THD on power factor – use Total Harmonic Distortion eq. (3) and (1) above
(2)
The DXF geometry to be imported should be in the format as shown below. The DXF should be for 1
slot and 1 pole of the machine positioned with the shaft centre point on the origin and lying on the
positive x axis as shown below. The DXF units should be in mm.
The Motor-CAD model should be set up so as to match as closely as possible the DXF geometry.
For the moment the region locations must be the same in the Motor-CAD geometry and the DXF
geometry however in the future it is intended to give the user the option to select the regions
manually.
The DXF geometry is imported using the window shown below (see geometry import for more details)
After selecting the DXF file the imported DXF will then be displayed overlayed on top of the Motor-
CAD geometry as shown below:
The model can then be solved using this custom geometry by clicking on the solve button in the
Control Page.
The effect of the custom geometry on the Flux pattern can also be seen as shown below:
10 Temperatures
Motor-CAD allows the results to be displayed on the Schematic, the Radial and Axial Cross
sections and the Slot.
The Motor-CAD results can also be compared against test results for validation of the model.
10.1 Schematic
Schematic diagram of the thermal circuit used to aid in the analysis of steady-state and transient
thermal resistance, power flow and node temperature data. The top left hand corner of the screen will
indicate if the data is for steady state or the final time step of a transient calculation.
If the Schematic diagram is gone to from one of the input data screens (cross-section, winding or
input data) then the steady state results will calculated and shown. If the Schematic diagram is gone
to directly after a Transient Graph then the data will be of the final time step of the transient.
Note: The results can be locked if necessary using the Lock Results menu option
The components are colour coded to match those shown in the Radial Cross-Section and Axial
Cross-Section Editors. Alternatively, the user can view the output data in tabular form using the
Output Data viewer.
Two Plot Option radio-button selectors are used to select the parameters to be displayed, i.e.
Resistance:
Thermal Resistance Label
Thermal Resistance Value
Power Flow through the resistance.
Temperature Drop across the resistance
Node:
Node Label
Node Temperature
Node Thermal Capacitance
Node Delta Temperature from ambient
(Convection): (Radiation):
The node temperature display shows the main nodes used for the thermal calculations. This display
shows an alternative way of displaying the nodes to the Schematic and the Circuit Editor views.
The node display options allow the user to select which nodes to display and whether to display the
node labels, temperatures or delta temperatures.
The radial tab allows the user to view the names and temperatures of the nodes on the radial view.
The node temperature display shows the main nodes used for the thermal calculations. This display
shows an alternative way of displaying the nodes to the Schematic and the Circuit Editor views.
The node display options allow the user to select which nodes to display and whether to display the
10.4 FEA
Motor-CAD has an integrated finite element module that allows finite element analysis to be
performed. This allows to power of finite element analysis to be performed to model features in detail
so that the analytical models in Motor-CAD can be compared (and if necessary calibrated) with the
FE model. The results are then fed into the complete machine model in Motor-CAD. This then
harnesses the power of the finite element and analytical models using each to their best effect.
Finite element analysis (FEA) can be useful for calibrating areas of the Motor-CAD lumped circuit
model.
See Slot Temperatures for details for the slot finite element calibration.
The winding layer model is used in Motor-CAD to model the slot and its components. The slot
temperatures view allows the winding layer model to be verified and calibrated if necessary using a
2D finite element model. The 2D finite element model is set up with the the temperature of the
housing calculated in Motor-CAD used as the boundary condition on the finite element model. The
active copper and stator losses are used in the finite element model.
To run the finite element solution click on the Solve Slot Model button. Motor-CAD then produces
as thermal solution for the slot with average, minimum and maximum winding temperatures
displayed.
These results can then be compared with the winding layer model and the winding layer model
calibrated if required. Calibration of the internal Motor-CAD model can be done by setting the Max
Calibration and Min Calibration values. Note: as the FE model is only 2D and does not take into
account interface gaps then it is advised to set the 2D Calibration Model in Motor-CAD so that the
Motor-CAD model best matches the FE model.
Shading:
Temperature - shows the temperatures
Gradient - shows the temperature gradient
Region - shows the model regions
Shading function: allows different shading functions to be selected e.g. J = Current Density, PG =
loss density etc.
Options:
Legend - display the legend
Boundaries - show the region boundaries
Equipotential lines - show the equipotential lines
Mesh - show the mesh
Vectors - show the temperature gradient vectors
The debug options allow the user to identify any problems with the model geometry and should not
normally be required.
The status panel highlighted in the in the bottom corner gives details of the region properties when
the mouse pointer is moved over them. The properties shown are:
The model region
T - The temperature at the point
K - The thermal conductivity at the point
LD - The loss density at this point
L - The loss for this region. Note this is for the proportion of the machine modelled.
The finite element module is a separate functionality module inside Motor-CAD and has to be leased
or purchased separately.
Details of the finite element solving process can be viewed in the message display window if
required:
The mesh generation is done automatically. The mesh can be viewed by selecting the mesh option
and the results can be displayed as shown below:
After running the slot FE analysis it can be seen that there is a slight difference in the average and
maximum winding temperatures predicted. The Motor-CAD layered winding model assumes that
there are a large number of conductors in the slot that are evenly distributed. In this case there are a
reasonable number of conductors and they are evenly distributed so the results are close, this may
not be the case if the assumptions for the layered winding model are not met.
There are 2 calibration parameters that can be used to match the analytical winding model to the FE
results.
Min Calibration : this multiplier is used to increase / decrease the thermal resistance of the winding
to the lamination. A value > 1 will increase this resistance and so the minimum winding temperature.
Max Calibration : this multiplier is used to increase / decrease the thermal resistance of the winding
inside the slot. A value > 1 will increase this resistance and so the maximum winding temperature.
To show the calibration would be better to have a slot that has a more uneven conductor distribution
as shown below. You can see that the conductor distribution in the slot is not even and so the
winding layer model is not that accurate.
Can calibrate the winding layered model using the Min and Max Calibration parameters. It is
usually best to start with the max calibration to get the difference between the maximum and
average temperatures of the analytic model the same as the FE results. In this case have set a
value of 0.55. This takes account of the fact that the conductors are towards the side of the slot.
Now adjust the min calibration value to get the average temperature in the slot correct.
With the 2 calibration values set the analytic model is now calibrated to the conductor positions in
the slot and can run the full machine model knowing that the winding model is good.
10.5 Validation
The validation interface allows the user to compare the model results against another set of results.
These results may be from test measurements or from another Motor-CAD model.
The validation temperatures entered by the user are saved in the Motor-CAD model.
The user can load in and save the results data to a separate file if required. The file format is csv
(comma separated values). The easiest way to create a file is to edit a file saved from Motor-CAD.
The results can then be displayed as a bar graph as shown below. The colours are the colours used
for the different components within Motor-CAD (e.g. Yellow = Winding, Red = Stator, Blue = Housing
etc..).
The difference between the results can also be displayed as shown below:
It is planned in the future to have an auto-calibration feature to allow a model to be quickly calibrated
against test results.
11 Transient Graph
11.1 Thermal Transient Calculation
Thermal capacitances are added to the motors thermal Schematic circuit so that the Transient
Graph can be calculated. Each thermal capacitance is calculated from the Materials & Weights
[Input Data Editor] data for the particular motor component. Numerical values of the Thermal
Capacitance [Output Data] are also given. The numerical values can also be plotted on the
Schematic.
The title and axes labelled on the can be altered using the above editor:
The title and axes labelled on the Temperature Graph can be altered by unselecting the automatic
titles option and using the above editor . When the automatic titles option is selected then the titles
are set automatically by Motor-CAD:
Temperature time graph comparing the Temperature Graph for the current calculation with the Test
Temperature Graph for External Test Temperature Data imported into Motor-CAD.
Use Test Data to plot test temperature results on the calculated Temperature Graph, i.e.
the Combined Temperature Graph
Use Motor-CAD results to plot previous model temperature results on the calculated
Temperature Graph, i.e. the Combined Temperature Graph and Difference Temperature
Graph.
The Full and reduced node comparison allows the comparison of transient results for the full
thermal network and the reduced node network.
The Custom Graph Settings feature allows the user to save and load customised graph settings. The
settings for the Legend, Point Type, Line Colour and whether a node is plotted are can be saved and
loaded from .gsf files (graph settings file).
The default location for the file is a sub-folder called graphsettings under the folder where the current
.mot datafile is located.
The Use Custom Graph Settings check box allows the user to activate / deactivate the Custom
Graph Settings. When the check box is unchecked altering any of the custom graph settings has no
effect on the Temperature Graph displayed.
The Drawing Options check boxes allow the customisation of the graph display. The Draw Lines
checkbox controls whether lines are drawn between the points. The Draw Points checkbox controls
whether the points are drawn. The Show Legend checkbox controls whether the legend is shown on
the graph.
The Select/Deselect all nodes checkbox allows the selection or deselection of the Graphed
checkboxes in the custom graph settings table.
The user can modify the graph legend, whether a point is graphed, the point type and line colour in
the custom graph settings table.
The custom graph settings table can be saved in a gsf file by clicking on the Save Settings button.
The user is then prompted to provide a name for the settings file as shown below:
The custom graph settings table can be loaded from a previously saved gsf file by clicking on the
Load Settings button.The user is prompted to select the graph settings file to open as shown below:
Examples:
If we calculate a transient then by default curves for all the nodes are plotted as shown below:
This can be too much data to get a good understanding of the thermal behaviour of the machine. The
user may set up standard graphing settings such as housing_winding.gsf and housing_winding_etc.
gsf, etc. that contain setting of how they would like to plot the graph, i.e. some settings with few
curves and others with a medium amount of curves. These setting can then be used at any time in
the future:
Grid showing data points comparing the Temperature Data with the Saved Temperature Data, i.e. for
each node the temperature difference. The resulting Diff Temperature Graph can then be viewed.
Data and settings can be changed in the grid which will effect how the Diff Temperature Graph is
plotted. The settings that can altered are:
Plot Curve - tick-box to define which curves will be included in the graph
Offset - It is possible to add an offset to each curve time and temperature data
Multiplier - It is possible to multiply each curve time and temperature data by a set amount
Temperature time graph for data held in the Diff Temperature Data chart.
It is usual to compare the Temperature Graph for the current calculation with the Saved Temperature
Graph using the Diff Temperature Graph.
The name of the Saved Temperature Data file is given in the graph title for future reference.
The graph axes limits and increments used to plot the following graphs can be altered by
unselecting the automatic scale option and using the above editor. When the automatic scale option
is selected then the graph limits are set automatically by Motor-CAD:
Temperature Graph
Test Temperature Graph
Saved Temperature Graph
Combined Temperature Graph
All the above graphs use the same axes scaling to make comparisons easier.
Below we see the graph that uses the setting set in the editor above:
Grid showing the power at each node in the schematic and its variation with time.
The power data can also be graphed showing the variation over the duty cycle.
A pop-up form used to indicate the progress of the Transient Graph calculation:
The current status of the graph will be shown if the 'Update Graph Now' button is clicked.
When the 'Enable Automatic Graph' is selected then the graph will be updated periodically during the
transient calculation. Warning: This will slow down the calcuation time. The automatic graph update
can also be enabled and the frequency of the update changed as shown below:
The following warning is displayed when the air property data is made a function of the schematic
nodal temperature (see Air Property Options). The results will be more accurate but at the expense
of increased execution times.
Execution can be terminated by pressing the Cancel Transient Calculation button. If terminated then
the results up to that point in time already calculated are displayed.
The Reduced node modelling functionality is enabled using the option shown above.
- there is an (automatically) generated list of our current nodes with their numbers as
shown in the Motor-CAD circuit, their names, and more importantly whether or not they
are enabled. When enabled a node is included in the reduced node model. For advice on
choosing the best nodal combinations, see Best Practices.
- Save Load List and Load node list allow the node list to be saved or loaded from a text file.
After saving or loading a node list, the file location will be shown in the box next to
‘Latest Node List’.
- Automatic Locking - locks the selection of nodes once a model has been run.
- Maximum Resistance - When the Reduced node model algorithm is run, due to the way it
manipulates the resistance network, sometimes there are some resistances which are
theoretical representations of the resistances between nodes, but sensibly we need not
always need to model them. Lowering the maximum resistance with this allows us to
simplify our nodal network further than simply removing nodes, by putting a ceiling on
the maximum resistance, and removing non-thermally relevant resistances. For Example,
with the default model, using only 8 nodes with a maximum resistance of 1E8 we see that
the circuit can appear fairly messy, and still quite complex (we note that between the
ambient node
- and rotor node, 12, that there is a resistance of 3204, in reality we would not consider this
as a resistance path):
- However, when we set the maximum resistance to be, say 100, we get a more simplified
circuit which is easier to read and will have a simpler resistance network if we export the
model:
This has eliminated a few resistance paths, as we knew, but has only effected the
temperature of one node, node 50, and only by 1C, probably less when we consider
rounding errors.
- When we look at this reduced circuit representation, we can also compare its
temperatures to those of the full model, and see that we have a fairly good
representation of the full circuit (for this motor):
We have some error which is higher than we would like. See best practices for methods of
increasing the accuracy of a reduced node model.
Steady State
Performing a steady state calculation with the reduced node modelling is done by changing to
the circuit editor tab, choosing the nodes, and then clicking the Run reduced node model
button on the reduced node window. Nodes can be added and removed from the model by
checking and unchecking them in the list as shown in the example below:
We have chosen only a few nodes as visible in both the list, and reflected in the circuit shown.
However if we would like to know more information about the winding layers. We simply
check more nodes in this part of the list until we have all the nodes we want, then run the
model:
Now we have selected a few more winding nodes, we can analyse the winding temperatures
without worrying about other nodes in the system too much. Note however that we have
tried to keep some resemblance in our model to the full model; we still have a housing node,
a stator node, winding nodes, and a magnet node. Generally keeping some level of
resemblance to the full motor helps to get understandable results.
Transient
Now that we have a model, we will attempt to run a transient calculation on it.
After selecting the nodes to have in our model in the circuit editor tab, then click on the
transient graph tab to run our transient:
To compare this to our full model, we change to the setup tab as shown below:
Once we have done this, we see that is has been loaded into the Transient model comparison
tab:
Now we simply click on the Full and reduced node comparison button to compare our model
to the full model:
We note that here we have chosen, in the Transient Graph setup tab, to use a difference
graph. We see that our difference is high on the winding nodes.
To improve the model there is a setting on the Reduced node model window, where we are
able to scale the capacitances in the model in order to fit the temperatures better. We see
that the temperatures in the reduced graph are generally lower than the full model, so it
would be good to reduce the capacitance values to match the temperature rise of the full
model better.
A good initial choice is to use automatic scaling, which scales the capacitance based on how
many nodes we are using:
We see that using the automatic straight line scaling has greatly improved our result, but it is
still a little low.
We see that the straight line scaling has suggested a capacitance of around 0.63, but this must
still be too high, so we can disable automatic scaling and choose our own scaling:
Trying a value of 0.5 we bring the temperatures fairly close to the full model temperatures.
Iterating through this we can try and bring temperatures close to the full model
temperatures.
See also Matrix Export for details of exporting the reduced model results.
- If a node has high losses, it is best to include it as this node can be considered ‘thermally
important’. In fact if there is comparatively large error on a node in your model, first
compare it to the full model, and see if there are any adjacent nodes with losses. If there
are, it may be best to consider including this node, or simply swapping the node you have
chosen for the node with losses. With comparison in mind, a print out or print screen of
the full model can be good for this comparison.
- Another way to counter error is to try and increase the maximum resistance, if it has been
set fairly low (by default, it will include all resistance paths the reduced node algorithm
calculates). For example, if the maximum resistance has been set to 100, but there are
enough paths with resistances in the range of 100-120, they could also be having a
physical effect on the temperatures in the model, and increasing the maximum
resistance is a good way to combat error. Gradual increases in this maximum resistance
along with comparison to the full model will allow integration of important resistance
paths into the model. With this in mind a good plan can be to start with the resistance
very high, and slowly decrease it until it has an effect on the model, to eliminate
unimportant paths, and retain important ones.
- When using complex cooling models, such as through ventilation (below), it is important
to consider the fact that we have added a fixed temperature node to the model.
Removing this node will usually result in catastrophic error, so it is advised to leave it in
the model. With this and the earlier points in mind, we see other points introduced by
the through ventilation model add nodes with negative losses. It is sensible to leave
these nodes in the model, or compare the model with and without them before
proceeding to remove them from the model completely.
See also Reduced Node Model for details of the reduced model setup.
Note: The results can be locked if necessary using the Lock Results menu option.
Once the transient has been calculated the user has the following options for analysing the data (set
of tab-sheets):
Setup:
Transient Control
Power Graph Setup
Temperatue Graph Setup
Data:
Temperature Data
Test Temperature Data (only available if External Test Temperature Data set)
Saved Temperature Data (only available if Saved Temperature Graph set)
Diff Temperature Data (only available if Saved Temperature Graph set)
Power Data
Speed Increase Data
Graphs:
Power Graph
Temperature Graph
Test Temperature Graph (only available if External Test Temperature Data set)
Saved Temperature Graph (only available if Saved Temperature Graph set)
Combined Temperature Graph (only available if External Test Temperature Data set)
Diff Temperature Graph (only available if Saved Temperature Graph set)
Some of the Tab-Sheets will not be available if certain options are not set.
If the user goes back to the Schematic then the temperature, resistance and power data for the last
transient is displayed (indicated by a symbol in the top left hand corner of the screen). If
now data is changed then the transient graph can still be viewed (not re-calculated) by pressing the
button again.
Grid showing data points of Saved Temperature Data imported into Motor-CAD. This data can be
compared with the current transient calculation by plotting the difference between the two sets of
data (Diff Temperature Graph). The imported data can be viewed alone on the Saved Temperature
Graph.
Data and settings can be changed in the grid which will effect how the Saved Temperature Graph is
plotted. The settings that can altered are:
Plot Curve - tick-box to define which curves will be included in the graph
Offset - It is possible to add an offset to each curve time and temperature data
Multiplier - It is possible to multiply each curve time and temperature data by a set amount
The currently calculated Temperature Graph data can be saved to a transient results file (.trf) so that
it can be compared results for future calculations, i.e. make a change to the design and see how
much the temperatures increase or decrease.
Clicking on the Save Current Results button allows the user to save the current Temperature Data.
The default location for the file is a sub-folder called transientResults under the folder where the
current .mot datafile is located. The format of the datafile is shown below:
Clicking on the Load Previous Results button allows the user to load in a previously saved set of
Temperature Data.
After loading a previous set of Transient Temperature Data results four additional tabs are displayed
in the Transient Graphsection as shown below:
The Saved Temperature Data and Saved Temperature Graph tabs show the previously saved
temperature data loaded from the file.
The Diff Temperature Data and Diff Temperature Graph tabs show the difference between the current
results (Temperature Data tab) and the previously saved results (Saved Temperature Data tab).
The scaling used for the Saved Temperature Graph is the same as that used for the Temperature
Graph to allow for easy comparison of values. If necessary the scaling can be adjusted manually
using the Graph Limits settings of the Graph Setup tab.
Example:
To illustrate the use of the Saved Temperature Data method we shall make a simple example. We
will calculate a transient duty cycle for a BPM motor and then reduce the interface gap between the
stator lamination and housing (Stator Lam - Housing) from 0.03mm to 0.003mm.
The transient with the original 0.03mm gap is shown below, i.e. the Saved Temperature Graph:
The new transient with the new 0.003mm gap is shown below, i.e. the Test Temperature Graph:
Both seem to be very similar. The Diff Temperature Graph shown below however highlights the
differences, i.e. that the housing temperature is reduced in the larger gap machine while the other
temperatures are reduced:
The data used to draw the graph is held in the Saved Temperature Data chart. It is usual to compare
the Temperature Graph for the current calculation with the Saved Temperature Graph using the Diff
Temperature Graph.
The name of the Saved Temperature Data file is given in the graph title for future reference.
Special integration techniques are used in Motor-CAD to give stable calculations in circuits having
vastly different sizes of thermal capacitance (termed a stiff set of equations). The problem is that
the thermal transient calculation speed is critically related to the component with the lowest thermal
capacitance and some nodes within the model have very low capacitance values, i.e. the end-space
air nodes. If we were to use such low thermal capacitance's in the analysis then unrealistically long
calculation periods may result. There are two general techniques that can be used to speed up the
calculation speeds - both however lead to slight inaccuracies:
1. Artificially increase the thermal capacitance of nodes that have very low thermal
capacitance. This is the techniques used in Motor-CAD at present when [Transient Fluid
Nodes] = [Include Capacitance] (default). More details are given in Transient Speed Increase
Factor.
2. Make all nodes that have low thermal capacitance arithmetic nodes with fixed temperature
over the integration time period. This has been implemented for transient calculations with a
fluid cooling system (wet rotor, water jacket, through ventilation) if the [Transient Fluid Nodes]
= [Neglect Capacitance]. If selected then any heating of fluid nodes with time is neglected.
This may give a faster solution and in some cases may be more accurate as the capacitance
effects of the fluid should only be substantial if the flow rate is small. This method is selected
using the Transient Fluid Nodes selector in Calculation Options [Input Data Editor]
This screen gives data on the number of nodes that have had their thermal increased and by what
amount.
Grid showing Transient Graph data points calculated and used to plot the Temperature Graph).
Data and settings can be changed in the grid which will effect how the Temperature Graph) is
plotted. The settings that can altered are:
Plot Curve - tick-box to define which curves will be included in the graph
Offset - It is possible to add an offset to each curve time and temperature data
Multiplier - It is possible to multiply each curve time and temperature data by a set amount
The data in the table is also put into a file called datafile_name_Trans.dat
For changing the graph settings the transient temperature graph setup is more a powerful method of
setting up the graph and allows the graph settings to be stored and retrieved.
Transient Graph, i.e. temperature time graph for the current transient calculation.
The data used to draw the graph is held in the Temperature Data chart.
Custom Graph Settings can be used to draw the graph according to setting saved by the user.
Chart Titles
Graph Limits
Set Custom Graph Settings to plot the graphs in just the way you require (symbols, colours,
labels, etc.) and save the setting for later calculations.
Grid showing data points imported into Motor-CAD using the External Test Temperature Data option.
This will usually be tested transient data that is to be compared with the current transient calculation
. The test data alone can be viewed on the Test Temperature Graph. Both the calculated and test
data can be viewed together on the Combined Temperature Graph.
Data and settings can be changed in the grid which will effect how the Test Temperature Graph and
Combined Temperature Graphs are plotted. The settings that can altered are:
Plot Curve - tick-box to define which curves will be included in the graph
Offset - It is possible to add an offset to each curve time and temperature data
Multiplier - It is possible to multiply each curve time and temperature data by a set amount
This option is used to select the file in which transient data exists that is to be plotted on the
calculated transient graphs, i.e. the Combined Temperature Graph as shown below:
The test data alone can also be plotted using the Test Temperature Graph
The file format is as shown below, i.e. 1st time column(seconds) followed by temperature columns:
This data is put in the grid - Test Temperature Data where it can be viewed and labels, colours,
symbols, etc edited.
If the test data is very large then the user can opt to import every nth point.
In v4 a new file format has been added to allow the user to save the test data and the plot settings
for external test data.
The user can load external test data in either the .dat format with no plot settings or in the .tst format
which includes plot settings as shown below.
The .tst file can be generated from a .dat file in Motor-CAD by loading the .dat file and then saving it
as a .tst file.
Temperature time graph for External Test Temperature Data imported into Motor-CAD.
The data used to draw the graph is held in the External Test Temperature Data chart. It is usual to
compare the Temperature Graph for the current calculation with the Test Temperature Graph using
the Combined Temperature Graph
The name of the External Test Temperature Data file is given in the graph title for future reference.
To zoom into a section of a graph hold down the left-hand mouse button and drag out the area to be
zoomed (from top-left to bottom-right), letting go of the key once the area to be zoomed has been
highlighted. To zoom back out just hold down the left-hand mouse button and drag out any top-right
to bottom-left area within the graph area, letting go of the key once the area has been highlighted.
Temperature Graph
Power Graph
Test Temperature Graph
Saved Temperature Graph
Combined Temperature Graph
Diff Temperature Graph
Sensitivity Graph
12 Circuit Editor
The circuit editor allows the user to view the circuit that is automatically generated and solved by
Motor-CAD. This view provides more detail than the Schematic view.
The options boxes highlighted above on the left of the view allow the user to select which items to
display. The different option are given below.
The user can choose to select to display the resistance labels, resistance values, the power
For each power source the user can select to view the power source labels, the power source values
or nothing.
For each node the user can select to view the node label, the node temperature, the node
capacitance that is used for transient analysis or nothing. The user can also select whether to
display the node numbers. The node numbers are used when editing the circuit.
The user can also select how the grid is displayed, the grid can be displayed with labels and no grid
lines, with labels and grid lines, with lines and no labels or with no labels or lines. The labels are
used when editing the circuit for positioning the component in the circuit.
The user can zoom and scroll the circuit editor display as it is not always possible to see in detail
the whole circuit. Zooming is done by clicking and holding down the left mouse button on the view
while dragging the zoom rectangle. Releasing the mouse button will complete the selection of the
zoom area. The screen shot below shows an example of how the zoom allows the user to see in
detail part of the circuit.
Scrolling of the display is done by holding down the right mouse button and moving the mouse.
Releasing the mouse button will stop the scrolling.
The display can be reset to the initial zoom and scroll settings, that show the whole of the circuit
with the maximum possible zoom, by either clicking on the 'Reset View' button on the left of the
screen or clicking with the left mouse button on the view area.
The Editor view is selected from the Editor tab of the Circuit Editor view.
The Editor screen is shown below.
Resistance - allows the user to add a new resistance or to remove or modify an existing
resistance.
The user must give the resistance a name and value. The Node1 and Node2 drop down boxes should
be used to specify the resistance terminals. A description can also be given. If a resistance already
exists between the specified nodes then its value will be changed to the value specified by the user.
If the user specifies a value of infinite resistance e.g. 1e9 Ohms then the resistance will no longer be
present in the circuit. Note the user is not allowed to specify a resistance of 0 Ohms.
Power Source - allows the user to add a new power source or to remove or modify an existing
power source.
The user must specify the power source name and value. The user must the specify the node that
the power source connects to and can give a description. If a power source already exists
connected to this node then its value will be changed to the value specified by the user. If the user
specifies a value of 0 then the power source will be removed.
Capacitance - allows the user to add a new capacitance or to remove or modify and existing
capacitance.
If any of the data for the component is invalid then the line will be highlighted in red as shown below.
The Edited circuit is saved in the Motor-CAD .mot file with the rest of the model specification. It is
also possible to save, load and clear the circuit using the buttons highlighted below. The circuit file is
saved in a external circuit file '.ecf'
An example of a circuit modification for some external control electronics bolted to the housing is
shown below.
The new and modified components are displayed in the Circuit with the modified lines in red as can
be seen below.
Transient results that include any new nodes can be produced as shown below.
The values of Power sources added in the circuit editor can also be varied with rotational speed as
for other loss values as shown below:
The values of Power sources added in the circuit editor can also be varied in the duty cycle as for
other loss values as shown below:
13 Sensitivity Study
Sensitivity analysis allows the user to assess the impact of parameter changes to the thermal
model. The analysis tool allows the user to vary a single or multiple parameters. The main uses of
the sensitivity tool are seen to be in the sensitivity of a design to tolerances and for running ranging
calculations to optimise a design.
The parameters are found by selecting the tabsheet that they are found from the drop down box as
shown below.
When the parameter to be modified has been found the user then enters the minimum and maximum
values that they wish for the parameter. The user can then either set the step size or the number of
values as shown below.
The variation value is validated when the user selects another line. The number of values and the
total number of runs have been automatically calculated as shown below
Where you wish to vary several parameters with the same values then the linked variable option
shown below allows several parameter values to be linked. You can also specify a multiplier by
which the the linked parameter will vary.
If the run data entered by the user is not valid then the line will be highlighted in red as shown below.
The parameter study is then run by clicking on the 'Run Parameter Study' button at the bottom of the
screen.
The status of the parameter study is shown in the status line at the bottom of the screen as shown
below.
During the parameter study it is possible to view the results that have already been generated. The
steady state results sheet shows the results that have been generated as shown below.
It is also possible to view the graph of the results while the analysis is still in progress by going to
the Graph view as shown below.
Once the parameter study has completed the mouse pointer will change back from the hour glass to
the normal cursor.
For more details on the analysis of the results see Sensitivity Results.
14 Scripting
Motor-CAD is ActiveX enabled and allows scripting from other applications to automate the design
process. (Note: For the internal script to run then Motor-CAD must have ActiveX enabled see ActiveX
for details of enabling ActiveX in Motot-CAD).
The Motor-CAD scripting interface allows the user to create and run scripts using VBS (Visual Basic
Script). These scripts can run Motor-CAD and any other ActiveX enabled programs from the script.
There is an example script loaded in Motor-CAD by default. This script shows a general outline of a
script that can be used as a basis for a script.
The script can be edited in the main text window. Cutting and pasting of a text selection is done by
clicking the right mouse button.
The scripting controls highlighted below allow the user to load, save, run, pause and stop a script.
The script runs Motor-CAD using ActiveX calls. To be able to stop the script using the user interface
it is necessary to have the following line in the innermost loop of your script as shown below:
The Script Control option allows the script to be automatically run before or during the solving
process.
The "Run before Analysis" option sets the script to run so that model can be set up prior to running
a calculation. The script can be set to run automatically before a steady state or transient
calculation by selecting the check boxes as shown below. Note: care must be taken when using
this to ensure the script is valid as once the script is run the user controls will not allow the script to
be stopped.
The "Run during Analysis" option sets the script to run every step of the solving process. For a
steady state calculation the script will be run every iteration. For a transient calculation the script will
be run before calculating each point.
The "Run during Analysis" option gives a lot of flexibility and customization in the use of Motor-
CAD for an advanced user. Possible uses are:
Calculation and use of losses based on user equations.
Calculation and use of heat transfer coefficients based on user equations to override those
calculated by Motor-CAD.
Coupling of cooling systems together so that the outlet of one system is the inlet of another
Coupling of the inlet and outlet of a cooling system so that the inlet temperature is a function of
the outlet temperature and flow ( to model a heat exchanger)
Variation of other input parameters as functions of speed, time or another parameter
The status of the script is shown in the Script Status window and messages are displayed in the
Message Display screen.
15 Flow Circuit
The Flow Circuit sheet allows the visualization of the flow circuits.
At present this is only available for Through Ventilation and Housing Water Jacket flow. Other
cooling types will be added to this feature in the future.
Series of tab-sheets used to display numerical output parameters. The output is divided into the
following categories:
Spray Cooling
Slot Water Jacket
Housing Water Jacket
Shaft Spiral Groove
Wet Rotor
The drive sheet gives the calculated values for the following machine parameters:
DC Bus Voltage
RMS Line-Line Supply maximum available at the motor terminals based on the control strategy
Voltage
RMS Phase Supply maximum available at the motor terminals based on the control strategy
Voltage
RMS Line-Line Terminal the drop voltage over the winding resistance plus the time derivative of the
Voltage flux
RMS Phase Terminal the drop voltage over the winding resistance plus the time derivative of the
Voltage flux
Harmonic Distortion Line-
Line Terminal Voltage
Harmonic Distortion
Phase Terminal Voltage
Peak Current
RMS Current
Phase Current
Power Factor (leading) See Power Factor for details
Phase Advance or torque angle, refers to the angle between the current and back EMF
waveforms
The reference 0 phase advance is when current and back EMF waveforms
are aligned
Drive Offset Angle (Open angle used to align the S-pole axis of the rotor with magnetic axis of
Circuit) phase 1
Drive Offset Angle (On angle used to align the S-pole axis of the rotor with magnetic axis of
load) phase 1
Drive angle to give phase angle corresponding to the maximum torque, assuming constant
maximum torque parameters
Phasor Angle angle used to align the MMF phasor of phase 1 with x-axis - see Winding
pattern
D axis inductance
Q axis inductance
End Winding Inductance
D axis RMS current phase value
Q axis RMS current phase value
Torque Constant (Kt) ratio between the airgap torque (average FE) and the peak line current
Back EMF Constant (Ke) ratio between peak line-line EMF and the rotational speed in rad/s
Electrical Constant ratio between phase inductance and phase resistance
Mechanical Constant
The drive sheet gives the calculated values for the following machine parameters:
Maximum torque the maximum torque possible from this machine based on the parameters at the
possible operating point (Ld, Lq, magnet flux, current)
Load point torque the torque for the current operating point based on the parameters at the operating
(flux linkage) point (Ld, Lq, current)
Average torque Airgap torque. See Torque calculation methods.
(virtual work)
Average torque Airgap torque. See Torque calculation methods.
(Maxwell stress)
Average torque The average airgap torque from virtual work and Maxwell stress finite element
(FE) methods
Average torque based on DQ axes theory. See Torque calculation methods.
(flux linkage)
Average torque with squarewave drive - based on sum of the products between phase back EMFs
D Axis flux
linkage (Open
circuit)
Q Axis flux
linkage (Open
circuit)
D Axis flux
linkage (On load)
Q Axis flux
linkage (On load)
Torque Constant ratio between the airgap torque (average FE) and the peak line current
(Kt)
Back EMF ratio between peak line-line EMF and the rotational speed in rad/s
Constant (Ke)
Stall Current ratio between maximum supply voltage and equivalent resistance between two
motor phases
Stall Torque Stall Current x Torque Constant
Cogging Period cogging torque period = 360 / slots / (poles / GCD (poles, slots)) [mdeg]
Cogging cogging torque frequency = 6 * speed/cogging period
Frequency
Fundamental shaft speed [rpm] / 60 x pole pairs
Frequency
Mechanical shaft speed [rpm] / 60
Frequency
Magnet Br at 20CMagnet remanence at 20C
Magnet Br Magnet remanence at the magnet operating temperature
Magnet Operating temperature of the magnets
temperature
Magnetic Factor by which the full machine can be reduced to increase calculation time =
symmetry factor Slots / GCD (Slots, Poles pairs)
Airgap flux Peak flux density in the airgap
density (peak)
Stator Tooth flux Peak flux density in the stator tooth
density (peak)
Stator Yoke flux Peak flux density in the stator back iron
density (peak)
Rotor Yoke flux Peak flux density in the rotor back iron
density (peak)
Set of 3 Grids showing numerical Output Data for the Temperature values.
Note that the user can set separate values for Ambient Temperature [Convection] and Ambient
Temperature [Radiation]. This is required if the local fluid ambient is different to that of the
surrounding surfaces with a view of the machine. The hot/cold surface will give a different amount of
radiation than if the fluid ambient was used in its calculation.
T [Housing - Overhang (F)] Temperature of housing overhang [Front] Schematic node [ Hoh
[F]]
T [Flange Mounted Plate] Temperature of flange mounted plate Schematic node [ Plate]
T [Housing - Overhang (R)] Temperature of housing overhang [Rear] Schematic node [ Hoh
[R]]
T [Stator Lamination (back iron)] Temperature of stator lamination (back iron) Schematic node [
Yoke]
T [Stator Lamination (tooth)] Temperature of stator lamination (tooth) Schematic node [ Tooth
]
T [WJ Fluid (A)] Temperature of housing water jacket active fluid Schematic
node [ wj[A]]
T [WJ Fluid (F)] Temperature of housing water jacket front fluid Schematic
node [ wj[F]]
T [WJ Fluid (R)] Temperature of housing water jacket rear fluid Schematic
node [ wj[R]]
T [SG Fluid (A)] Temperature of shaft spiral groove active fluid Schematic
node [ sg[A]]
T [SG Fluid (F)] Temperature of shaft spiral groove front fluid Schematic
node [ sg[F]]
T [SG Fluid (R)] Temperature of shaft spiral groove rear fluid Schematic
node [ sg[R]]
T [Wet Rotor (F)] Temperature of wet rotor front fluid Schematic node (end
winding bore).
T [Wet Rotor (R)] Temperature of wet rotor rear fluid Schematic node (end
winding bore).
T [Wet Rotor (EW End)] Temperature of wet rotor end winding end section fluid
Schematic node (front and rear nodes)
T [Active Winding Maximum] Maximum winding temperature of the active section of the
windings
T [Active Winding Minimum] Minimum winding temperature of the active section of the
windings
Set of 2 Grids showing numerical Output Data for the following Losses:
Loss [Stator Copper] Stator Winding copper loss. See also Stator
Copper Loss Distribution.
Loss [Stator Copper Freq Comp] Second Stator Winding copper loss component -
used for proximity losses.
Loss [Stator Copper](Active) Stator Winding copper losses in the Active
section.
Loss [Stator Copper](EWdg Front] Stator Winding copper losses in the front
endwindings.
Loss [Stator Copper](EWdg Rear] Stator Winding copper losses in the rear
endwindings.
Loss [Stator Back Iron] Stator lamination iron loss - back iron
component.
Loss [Stator Tooth] Stator lamination iron loss - tooth component.
Loss [Stator Tooth Tip] Stator lamination iron loss - tooth tip component.
Loss [Rotor Cage] Total losses in the rotor cage. See also Rotor
Cage Loss Distribution.
Loss [Rotor Cage](Active) Losses in the rotor bars.
Loss [Rotor Cage](EndRing Front) Losses in the front endring.
Loss [Rotor Cage](EndRing Rear) Losses in the rear endring.
Loss [Rotor Back Iron] Rotor lamination iron loss - back iron
component.
Loss [Rotor Tooth] Rotor lamination iron loss - tooth component.
Loss [Rotor Back Iron]@Tm(i/p) Rotor iron loss (back iron) at Magnet
Temperature - Tm(i/p).
Shaft Torque
Shaft Torque [Required] Required torque in a Losses Vary With
Temperature and Load model.
The drive sheet gives the calculated values for the following machine parameters.
Note: the losses are calculated for the last electrical cycle run, if 5 cycles are run then only the 5th
cycle will be used for the loss calculation. See also Number of cycles in graph settings.
These are the surface areas used for radiation and convection heat transfer.
Area [Housing - Active] Surface area of main housing [non-finned section] available for
cooling to ambient by convection/radiation.
Area [Housing - Active Fin Base] Surface area of main housing [fin base section] available for
cooling to ambient by convection/radiation.
Area [Housing - Active Fin Sides] Surface area of main housing [fin side section] available for
Area [Housing - Rear Corner Cutout] Surface area of rear overhanging housing section [corner
cutout section] available for cooling to ambient by convection/
radiation.
Area [Endcap - Front Radial] Surface area of front endcap available for cooling to ambient by
convection/radiation (radial section).
Area [Endcap - Front Axial] Surface area of front endcap available for cooling to ambient by
convection/radiation (axial section).
Area [Endcap - Rear Radial] Surface area of rear endcap available for cooling to ambient by
convection/radiation (radial section).
Area [Endcap - Rear Axial] Surface area of rear endcap available for cooling to ambient by
convection/radiation (axial section).
Area [Flange Mounted Plate] Surface area of flange mounted plate (Mounting Type) available
for cooling to ambient by convection/radiation.
Area [Foot Mounted Base] Surface area of foot mounted plate (Mounting Type) available
for cooling to ambient by convection/radiation.
Area [Encoder Case - Radial] Surface area of encoder case (Feedback Type) available for
cooling to ambient by convection/radiation (radial section).
Area [Encoder Case - Axial] Surface area of encoder case (Feedback Type) available for
cooling to ambient by convection/radiation (axial section).
Area Airgap Inner Surface Surface area of inner surface of airgap.
Area Airgap Outer Surface Surface area of outer surface of airgap.
Set of 3 Grids showing numerical Output Data for the following heat-transfer coefficient and surface
area parameters:
Convection Heat Transfer Coefficient Data (If Cooling Type = Natural Convection(TENV) [Left
Hand Grid]:
See Natural Convection Heat Transfer Coefficients
Convection Heat Transfer Coefficient Data (If Cooling Type = Blown Over (TEFC)) [Left Grid]:
(See also Heat Transfer Coeff - 2 [Output Data] and Mixed Convection)
See Mixed Convection Heat Transfer Coefficients
Radiation Heat Transfer Coefficient Data (If Cooling Type = Blown Over (TEFC)) [Left Grid]:
See Radiation Heat Transfer Coefficients
See Units.
N.B. This tab-sheet is only displayed if Cooling Type = Blown Over [TEFC]. See also Heat Transfer
Coeff - 1 [Output Data]
Set of 3 Grids showing numerical Output Data for the following heat-transfer coefficient parameters:
Natural Convection Heat Transfer Coefficient Data [Left Hand Grid]: See Natural Convection
Heat Transfer Coefficients
Forced Convection Heat Transfer Coefficient Data [Central Grid]: See Forced Convection Heat
Transfer Coefficients.
Mixed Convection Heat Transfer Coefficient Data [Right Hand Grid]: See Mixed Convection
Heat Transfer Coefficients
See Units.
hnc [Housing - Active Fin Base] Convection Heat Transfer Coefficient [Natural Convection]
of active housing [fin base section] available for cooling to ambient.
hnc [Housing - Active Fin Sides] Convection Heat Transfer Coefficient [Natural Convection]
of active housing [fin sides section] available for cooling to ambient
hnc [Housing - Active Fin Tips] Convection Heat Transfer Coefficient [Natural Convection]
of active housing [fin-tip section] available for cooling to ambient.
hnc [Housing - Active Fin Channel] Convection Heat Transfer Coefficient [Natural Convection]
of active housing [fin-tip section] available for cooling to ambient.Default is zero due to covered
nature of fins.
hnc [Housing - Active Corner Cutout] Convection Heat Transfer Coefficient [Natural Convection]
of active housing [corner cutout section] available for cooling to ambient.
hnc [Housing - Front Fin Base] Convection Heat Transfer Coefficient [Natural Convection]
of front housing [fin base section] available for cooling to ambient.
hnc [Housing - Rear Fin Sides] Convection Heat Transfer Coefficient [Natural Convection]
of rear housing [fin sides section] available for cooling to ambient.
hnc [Housing - Front Fin Sides] Convection Heat Transfer Coefficient [Natural Convection]
of front housing [fin sides section] available for cooling to ambient.
hnc [Housing - Front Fin Tips] Convection Heat Transfer Coefficient [Natural Convection]
of front housing [fin-tip section] available for cooling to ambient.
hnc [Housing - Front Fin Channel] Convection Heat Transfer Coefficient [Natural Convection]
of front housing [covered fin channel section] available for cooling to ambient. Default is zero due to
covered nature of fins.
hnc [Housing - Front Corner Cutout] Convection Heat Transfer Coefficient [Natural Convection]
of front housing [corner cutout section] available for cooling to ambient.
hnc [Housing - Rear Fin Base] Convection Heat Transfer Coefficient [Natural Convection]
of rear housing [fin base section] available for cooling to ambient.
hnc [Housing - Rear Fin Sides] Convection Heat Transfer Coefficient [Natural Convection]
of rear housing [fin sides section] available for cooling to ambient.
hnc [Housing - Rear Fin Tips] Convection Heat Transfer Coefficient [Natural Convection]
of rear housing [fin-tip section] available for cooling to ambient.
hnc [Housing - Rear Fin Channel] Convection Heat Transfer Coefficient [Natural Convection]
of rear housing [covered fin channel section] available for cooling to ambient. Default is zero due to
covered nature of fins.
hnc [Housing - Rear Corner Cutout] Convection Heat Transfer Coefficient [Natural Convection]
of rear housing [corner cutout section] available for cooling to ambient.
hnc [Endcap - Front Radial] Convection Heat Transfer Coefficient [Natural Convection]
of front endcap [radial section] available for cooling to ambient.
hnc [Endcap - Front Axial] Convection Heat Transfer Coefficient [Natural Convection]
of front endcap [axial section] available for cooling to ambient.
hnc [Endcap - Rear Radial] Convection Heat Transfer Coefficient [Natural Convection]
of rear endcap [radial section] available for cooling to ambient.
hnc [Endcap - Rear Axial] Convection Heat Transfer Coefficient [Natural Convection]
of rear endcap [axial section] available for cooling to ambient.
hnc [Flange Mounted Plate] Convection Heat Transfer Coefficient [Natural Convection]
of flange mounted plate available for cooling to ambient.
hnc [Foot Mounted Base] Convection Heat Transfer Coefficient [Natural Convection]
of foot mounted plate available for cooling to ambient.
hnc [Encoder Case - Radial] Convection Heat Transfer Coefficient [Natural Convection]
of encoder case [radial section] available for cooling to ambient.
hnc [Encoder Case - Axial] Convection Heat Transfer Coefficient [Natural Convection]
of encoder case [axial section] available for cooling to ambient.
hfc [Housing - Active Fin Base] Convection Heat Transfer Coefficient [Forced Convection]
of active housing [fin base section] available for cooling to ambient.
hfc [Housing - Active Fin Sides] Convection Heat Transfer Coefficient [Forced Convection]
of active housing [fin sides section] available for cooling to ambient.
hfc [Housing - Active Fin Tips] Convection Heat Transfer Coefficient [Forced Convection]
hfc [Housing - Active Fin Channel] Convection Heat Transfer Coefficient [Forced Convection]
of active housing [covered fin channel section] available for cooling to ambient.
hfc [Housing - Active Corner Cutout] Convection Heat Transfer Coefficient [Forced Convection]
of active housing [corner cutout section] available for cooling to ambient.
hfc [Housing - Front Fin Base] Convection Heat Transfer Coefficient [Forced Convection]
of front housing [fin base section] available for cooling to ambient.
hfc [Housing - Front Fin Sides] Convection Heat Transfer Coefficient [Forced Convection]
of front housing [fin sides section] available for cooling to ambient.
hfc [Housing - Front Fin Tips] Convection Heat Transfer Coefficient [Forced Convection]
of front housing [fin-tip section] available for cooling to ambient.
hfc [Housing - Front Fin Channel] Convection Heat Transfer Coefficient [Forced Convection]
of front housing [covered fin channel section] available for cooling to ambient.
hfc [Housing - Front Corner Cutout] Convection Heat Transfer Coefficient [Forced Convection]
of front housing [corner cutout section] available for cooling to ambient.
hfc [Housing - Rear Fin Base] Convection Heat Transfer Coefficient [Forced Convection]
of rear housing [fin base section] available for cooling to ambient.
hfc [Housing - Rear Fin Sides] Convection Heat Transfer Coefficient [Forced Convection]
of rear housing [fin sides section] available for cooling to ambient.
hfc [Housing - Rear Fin Tips] Convection Heat Transfer Coefficient [Forced Convection]
of rear housing [fin-tip section] available for cooling to ambient.
hfc [Housing - Rear Fin Channel] Convection Heat Transfer Coefficient [Forced Convection]
of rear housing [covered fin channel section] available for cooling to ambient.
hfc [Housing - Rear Corner Cutout] Convection Heat Transfer Coefficient [Forced Convection]
of rear housing [corner cutout section] available for cooling to ambient.
hfc [Endcap - Front Radial] Convection Heat Transfer Coefficient [Forced Convection]
of front endcap [radial section] available for cooling to ambient.
hfc [Endcap - Front Axial] Convection Heat Transfer Coefficient [Forced Convection]
of front endcap [axial section] available for cooling to ambient.
hfc [Endcap - Rear Radial] Convection Heat Transfer Coefficient [Forced Convection]
of rear endcap [radial section] available for cooling to ambient.
hfc [Endcap - Rear Axial] Convection Heat Transfer Coefficient [Forced Convection]
of rear endcap [axial section] available for cooling to ambient.
hfc [Flange Mounted Plate] Convection Heat Transfer Coefficient [Forced Convection]
of flange mounted plate available for cooling to ambient.
hfc [Foot Mounted Base] Convection Heat Transfer Coefficient [Forced Convection]
of foot mounted plate available for cooling to ambient.
hfc [Encoder Case - Radial] Convection Heat Transfer Coefficient [Forced Convection]
of encoder case [radial section] available for cooling to ambient.
hfc [Encoder Case - Axial] Convection Heat Transfer Coefficient [Forced Convection]
of encoder case [axial section] available for cooling to ambient.
hmix [Housing - Active Fin Base] Convection Heat Transfer Coefficient [Mixed Convection] of
active housing [fin base section] available for cooling to ambient.
hmix [Housing - Active Fin Sides] Convection Heat Transfer Coefficient [Mixed Convection] of
active housing [fin sides section] available for cooling to ambient.
hmix [Housing - Active Fin Tips] Convection Heat Transfer Coefficient [Mixed Convection] of
active housing [fin-tip section] available for cooling to ambient.
hmix [Housing - Active Fin Channel] Convection Heat Transfer Coefficient [Mixed Convection] of
active housing [covered fin channel section] available for cooling to ambient.
hmix [Housing - Active Corner Cutout] Convection Heat Transfer Coefficient [Mixed Convection] of
active housing [corner cutout section] available for cooling to ambient.
hmix [Housing - Front Fin Base] Convection Heat Transfer Coefficient [Mixed Convection] of
front housing [fin base section] available for cooling to ambient.
hmix [Housing - Front Fin Sides] Convection Heat Transfer Coefficient [Mixed Convection] of
front housing [fin sides section] available for cooling to ambient.
hmix [Housing - Front Fin Tips] Convection Heat Transfer Coefficient [Mixed Convection] of
front housing [fin-tip section] available for cooling to ambient.
hmix [Housing - Front Fin Channel] Convection Heat Transfer Coefficient [Mixed Convection] of
front housing [covered fin channel section] available for cooling to ambient.
hmix [Housing - Front Corner Cutout] Convection Heat Transfer Coefficient [Mixed Convection] of
front housing [corner cutout section] available for cooling to ambient.
hmix [Housing - Rear Fin Base] Convection Heat Transfer Coefficient [Mixed Convection] of
rear housing [fin base section] available for cooling to ambient.
hmix [Housing - Rear Fin Sides] Convection Heat Transfer Coefficient [Mixed Convection] of
rear housing [fin sides section] available for cooling to ambient.
hmix [Housing - Rear Fin Tips] Convection Heat Transfer Coefficient [Mixed Convection] of
rear housing [fin-tip section] available for cooling to ambient.
hmix [Housing - Rear Fin Channel] Convection Heat Transfer Coefficient [Mixed Convection] of
rear housing [covered fin channel section] available for cooling to ambient.
hmix [Housing - Rear Corner Cutout] Convection Heat Transfer Coefficient [Mixed Convection] of
rear housing [corner cutout section] available for cooling to ambient.
hmix [Endcap - Front Radial] Convection Heat Transfer Coefficient [Mixed Convection] of
front endcap [radial section] available for cooling to ambient.
hmix [Endcap - Front Axial] Convection Heat Transfer Coefficient [Mixed Convection] of
front endcap [axial section] available for cooling to ambient.
hmix [Endcap - Rear Radial] Convection Heat Transfer Coefficient [Mixed Convection] of
rear endcap [radial section] available for cooling to ambient.
hmix [Endcap - Rear Axial] Convection Heat Transfer Coefficient [Mixed Convection] of
rear endcap [axial section] available for cooling to ambient.
hmix [Flange Mounted Plate] Convection Heat Transfer Coefficient [Mixed Convection] of
flange mounted plate available for cooling to ambient.
hmix [Foot Mounted Base] Convection Heat Transfer Coefficient [Mixed Convection] of
foot mounted plate available for cooling to ambient.
hmix [Encoder Case - Radial] Convection Heat Transfer Coefficient [Mixed Convection] of
encoder case [radial section] available for cooling to ambient.
hmix [Encoder Case - Axial] Convection Heat Transfer Coefficient [Mixed Convection] of
encoder case [axial section] available for cooling to ambient.
hr [Housing - Active Fin Base] Radiation Heat Transfer Coefficient of active housing [fin
base section] available for cooling to ambient.
hr [Housing - Active Fin Sides] Radiation Heat Transfer Coefficient of active housing [fin
sides section] available for cooling to ambient.
hr [Housing - Active Fin Tips] Radiation Heat Transfer Coefficient of active housing [fin-tip
section] available for cooling to ambient.
hr [Housing - Active Fin Channel] Radiation Heat Transfer Coefficient of active housing
[covered fin channel section] available for cooling to ambient. Default is zero due to covered nature of
fins.
hr [Housing - Active Corner Cutout] Radiation Heat Transfer Coefficient of active housing
[corner cutout section] available for cooling to ambient.
hr [Housing - Front Fin Base] Radiation Heat Transfer Coefficient of front housing [fin
base section] available for cooling to ambient.
hr [Housing - Front Fin Sides] Radiation Heat Transfer Coefficient of front housing [fin
sides section] available for cooling to ambient.
hr [Housing - Front Fin Tips] Radiation Heat Transfer Coefficient of front housing [fin-tip
section] available for cooling to ambient.
hr [Housing - Front Fin Channel] Radiation Heat Transfer Coefficient of front housing
[covered fin channel section] available for cooling to ambient. Default is zero due to covered nature of
fins.
hr [Housing - Front Corner Cutout] Radiation Heat Transfer Coefficient of front housing [corner
cutout section] available for cooling to ambient.
hr [Housing - Rear Fin Base] Radiation Heat Transfer Coefficient of rear housing [fin
base section] available for cooling to ambient.
hr [Housing - Rear Fin Sides] Radiation Heat Transfer Coefficient of rear housing [fin
sides section] available for cooling to ambient.
hr [Housing - Rear Fin Tips] Radiation Heat Transfer Coefficient of rear housing [fin-tip
section] available for cooling to ambient.
hr [Housing - Rear Fin Channel] Radiation Heat Transfer Coefficient of rear housing [covered
fin channel section] available for cooling to ambient. Default is zero due to covered nature of fins.
hr [Housing - Rear Corner Cutout] Radiation Heat Transfer Coefficient of rear housing [corner
cutout section] available for cooling to ambient.
hr [Endcap - Front Radial] Radiation Heat Transfer Coefficient of front endcap [radial
section] available for cooling to ambient.
hr [Endcap - Front Axial] Radiation Heat Transfer Coefficient of front endcap [axial
section] available for cooling to ambient.
hr [Endcap - Rear Radial] Radiation Heat Transfer Coefficient of rear endcap [radial
section] available for cooling to ambient.
hr [Endcap - Rear Axial] Radiation Heat Transfer Coefficient of rear endcap [axial
section] available for cooling to ambient.
hr [Flange Mounted Plate] Radiation Heat Transfer Coefficient of flange mounted plate
available for cooling to ambient.
hr [Foot Mounted Base] Radiation Heat Transfer Coefficient of foot mounted plate
available for cooling to ambient.
hr [Encoder Case - Radial] Radiation Heat Transfer Coefficient of encoder case [radial
section] available for cooling to ambient.
hr [Encoder Case - Axial] Radiation Heat Transfer Coefficient of encoder case [axial
section] available for cooling to ambient.
Set of 3 Grids showing numerical Output Data for the following thermal resistance parameters:
Rt [Yoke Inner (Slot)] Thermal resistance inner half of stator yoke above slot -
radial direction. Labelled Yoke(Back Iron) in Schematic.
Rt [Yoke Inner (Tooth)] Thermal resistance inner half of stator yoke above tooth -
radial direction. Labelled Yoke(Tth) in Schematic.
Rt [Stator Lam - Housing] Thermal resistance between stator lamination and housing
(interface resistance). Labelled Lam-House in Schematic.
Rt [Housing - Housing OHang (R)] Thermal resistance between active section of motor and
housing overhang [rear of motor]. Used to model frameless motors that have Housing Dia = Stator
Lam Dia and -ve value of Housing Step [R]. Labelled H-Hoh[R] in Schematic.
Rt [Housing - Housing OHang (F)] Thermal resistance between active section of motor and
housing overhang [front of motor]. Used to model frameless motors that have Housing Dia = Stator
Lam Dia and -ve value of Housing Step [F]. Labelled H-Hoh[F] in Schematic.
Rt [Housing OHang/2 (R)] Thermal resistance of housing in axial direction along 1/2
overhang at rear of motor. Labelled Housing[R]/2 in Schematic.
Rt [Housing OHang/2 (F)] Thermal resistance of housing in axial direction along 1/2
overhang at front of motor. Labelled Housing[F]/2 in Schematic.
Rt [Liner - Lam (Slot Bottom)] Thermal resistance of gap between slot liner and stator
lamination (bottom of slot). Labelled Liner-Lam in Schematic.
Rt [Liner - Lam (Tooth Side)] Thermal resistance of gap between slot liner and stator
lamination (side of tooth). Labelled Liner-Lam in Schematic.
Rt [Housing - Endcap (F)] Thermal resistance between housing and endcap [Front of
motor]. Labelled House-Ecap in Schematic.
Rt [Housing - Endcap (R)] Thermal resistance between housing and endcap [Rear of
motor]. Labelled House-Ecap in Schematic.
Rt [Endcap - Flange Plate] Thermal resistance between endcap and flange mounted
plate. Labelled Ecap-Plate in Schematic.
Rt [Bearing - Endcap (F)] Thermal resistance between bearing and endcap [Front of
motor]. Note that the bearing thermal model is a very complex matter and the bearing internal
thermal resistance is at present included in with the bearing interface gaps (Rt [Bearing - Endcap
(F)], Rt [Bearing - Endcap (R)], Rt [Bearing - Shaft (F)], Rt [Bearing - Shaft (R)])
Rt [Bearing - Endcap (R)] Thermal resistance between bearing and endcap [Rear of
motor]. Note that the bearing thermal model is a very complex matter and the bearing internal
thermal resistance is at present included in with the bearing interface gaps (Rt [Bearing - Endcap
(F)], Rt [Bearing - Endcap (R)], Rt [Bearing - Shaft (F)], Rt [Bearing - Shaft (R)])
Rt [Bearing - Shaft (F)] Thermal resistance between bearing and shaft [Front of
motor]. Note that the bearing thermal model is a very complex matter and the bearing internal
thermal resistance is at present included in with the bearing interface gaps (Rt [Bearing - Endcap
(F)], Rt [Bearing - Endcap (R)], Rt [Bearing - Shaft (F)], Rt [Bearing - Shaft (R)])
Rt [Bearing - Shaft (R)] Thermal resistance between bearing and shaft [rear of
motor]. Note that the bearing thermal model is a very complex matter and the bearing internal
thermal resistance is at present included in with the bearing interface gaps (Rt [Bearing - Endcap
(F)], Rt [Bearing - Endcap (R)], Rt [Bearing - Shaft (F)], Rt [Bearing - Shaft (R)])
Rt [Sleeve - Stator Lam] Thermal resistance across stator bore sleeve to stator
lamination interface gap. Labelled Sv-Lam in Schematic. Not displayed if Sleeve Thickness = 0.
Rt [Rotor Lam (Outer)] Thermal resistance across rotor lamination radial length
(Outer Half). Labelled Rotor (Outer) in Schematic.
Rt [Rotor Lam (Inner)] Thermal resistance across rotor lamination radial length
(Inner Half). Labelled Rotor (Inner) in Schematic.
Rt [Rotor Tooth (Outer Half)] Thermal resistance across rotor lamination tooth (outer
half). Labelled Tooth(Outer) in Schematic.
Rt [Rotor Tooth (Inner Half)] Thermal resistance across rotor lamination tooth (inner
half). Labelled Tooth(Inner) in Schematic.
Rt [Cage - Rotor Lam] Thermal resistance across cage - rotor lamination interface
resistance. Labelled Cage-Lam in Schematic.
Rt [Stator Pole (Outer Half)] Thermal resistance across SRM stator pole (outer half).
Labelled Tooth/2 in Schematic.
Rt [Stator Pole (Inner Half)] Thermal resistance across SRM stator pole (inner half).
Labelled Tooth/2 in Schematic.
Rt [Rotor Pole (Outer Half)] Thermal resistance across SRM rotor pole (outer half).
Labelled Pole(outer) in Schematic.
Rt [Rotor Pole (Inner Half)] Thermal resistance across SRM rotor pole (inner half).
Labelled Pole(inner) in Schematic.
Rt [Shaft Ohang (F)] Thermal resistance axially along shaft section at front of
motor. Labelled Shaft[F] in Schematic.
Rt [Shaft Ohang (R)] Thermal resistance axially along shaft section at rear of
motor. Labelled Shaft[R] in Schematic.
Rt [Shaft Ext (F)] Thermal resistance axially along shaft between bearing
and connected device (front of machine). Set with Shaft Extension [F] Used if Fixed Temperatures
model used for front of shaft.
Rt [Shaft Ext (R)] Thermal resistance axially along shaft between bearing
and connected device (rear of machine). Set with Shaft Extension [R] Used if Fixed Temperatures
model used for rear of shaft.
Rt [Feet - Base] Thermal resistance between feet and base [Foot mounted
motor only]. Labelled Feet-Base in Schematic.
Rt [Impreg - Wdg Outer (Slot Bottom)] Thermal resistance across outer winding impregnation
layer (slot bottom component) Labelled Imp[slot bottom] in Schematic.
Rt [Ins - Wdg Outer (Slot Bottom)] Thermal resistance across outer winding copper insulation
layer (slot bottom component) Labelled Ins[slot bottom] in Schematic.
Rt [Impreg - Wdg Outer (Tooth Side)] Thermal resistance across outer winding impregnation
layer (tooth side component) Labelled Imp[tooth side] in Schematic.
Rt [Ins - Wdg Outer (Tooth Side)] Thermal resistance across outer winding copper insulation
layer (tooth side component) Labelled Ins[tooth side] in Schematic.
Rt [Housing (Act) - Amb] - Radn Thermal resistance [Radiation] between main housing
section and ambient. Labelled Housing-Amb in Schematic.
Rt [Housing (Act) - Amb] - Conv Thermal resistance [Convection] between main housing
section and ambient. Labelled Housing-Amb in Schematic.
Rt [Housing (Act) - Amb] Thermal resistance [Total] between main housing section
and ambient. Labelled Housing-Amb in Schematic.
Rt [Housing (F) - Amb] - Radn Thermal resistance [Radiation] between front overhanging
housing section and ambient. Labelled House[F]-Amb in Schematic.
Rt [Housing (F) - Amb] - Conv Thermal resistance [Convection] between front overhanging
housing section and ambient. Labelled House[F]-Amb in Schematic.
Rt [Housing (R) - Amb] - Radn Thermal resistance [Radiation] between rear overhanging
housing section and ambient. Labelled House[R]-Amb in Schematic.
Rt [Housing (R) - Amb] - Conv Thermal resistance [Convection] between rear overhanging
housing section and ambient. Labelled House[R]-Amb in Schematic.
Rt [Endcap (F) - Amb] - Radn Thermal resistance [Radiation] between front endcap and
ambient. Labelled Endcap[F]-Amb in Schematic.
Rt [Endcap (F) - Amb] - Conv Thermal resistance [Convection] between front endcap and
ambient. Labelled Endcap[F]-Amb in Schematic.
Rt [Endcap (F) - Amb] Thermal resistance [Total] between front endcap and
ambient. Labelled Endcap[F]-Amb in Schematic.
Rt [Endcap (R) - Amb] - Radn Thermal resistance [Radiation] between rear endcap and
ambient. Labelled Endcap[R]-Amb in Schematic.
Rt [Endcap (R) - Amb] - Conv Thermal resistance [Convection] between rear endcap and
ambient. Labelled Endcap[R]-Amb in Schematic.
Rt [Endcap (R) - Amb] Thermal resistance [Total] between rear endcap and
ambient. Labelled Endcap[R]-Amb in Schematic.
Rt [Flange Plate - Amb] - Radn Thermal resistance [Radiation] between flange mounted
plate and ambient. Labelled Plate-Amb in Schematic.
Rt [Flange Plate - Amb] - Conv Thermal resistance [Convection] between flange mounted
plate and ambient. Labelled Plate-Amb in Schematic.
Rt [Flange Plate - Amb] Thermal resistance [Total] between flange mounted plate
and ambient. Labelled Plate-Amb in Schematic.
Rt [Base Plate - Amb] - Radn Thermal resistance [Radiation] between foot mounted base
plate and ambient. Labelled Feet-Amb in Schematic.
Rt [Base Plate - Amb] - Conv Thermal resistance [Convection] between foot mounted
base plate and ambient. Labelled Feet-Amb in Schematic.
Rt [Base Plate - Amb] Thermal resistance [Total] between foot mounted base
Rt [EncCase - Amb] - Radn Thermal resistance [Radiation] between encoder case and
ambient. Labelled EncCase-Amb in Schematic.
Rt [Ins (Slot Base)] Thermal resistance across extra insulation (on top of Rt
[Liner (Slot Bottom)]) at the bottom of slot. Set using Ins [Slot Base] Thickness
Rt [Ins (Tooth Side)] Thermal resistance across extra insulation (on top of Rt
[Liner (Tooth Side)]) at the tooth sides. Set using Ins [Tooth Side] Thickness
Set of 3 Grids showing numerical Output Data for the thermal capacitance parameters.
Used in Transient Graph calculation.
Cap [Housing - Overhang (F)] Thermal capacity of housing overhang [front] node.
Cap [Flange Mounted Plate] Thermal capacity of flange mounted plate node.
Cap [Housing - Overhang (R)] Thermal capacity of housing overhang [rear] node.
Cap [Stator Lamination (back iron)] Thermal capacity of stator lamination (back iron)
schematic node.
Cap [Stator Lamination (tooth)] Thermal capacity of stator lamination (tooth) schematic
node.
Cap [Magnet Inner Surface] Thermal capacity of magnet inner surface schematic node.
Cap [Shaft Ohang - Front] Thermal capacity of shaft overhang (front) schematic node.
Cap [Shaft Ohang - Rear] Thermal capacity of shaft overhang (rear) schematic node.
Cap [End Space (F)] Thermal capacity of internal end-space (front) schematic
node.
Cap [End Space (R)] Thermal capacity of internal end-space (rear) schematic
node.
Cap [Foot Mounted Base] Thermal capacity of foot mounted base schematic node.
Cap [Winding Outer Surface] Thermal capacity of winding outer layer schematic node.
Set of 2 Grids showing numerical Output Data for the End Space Convection Correlation cooling.
The output data includes Heat Transfer Coefficients, Thermal Resistance, Dissipation and surface
areas for the internal surfaces of the motor in the end space region.
Heat transfer coefficients for all surfaces on the inside of the machine to the end space.
Thermal resistance values from all the internal surfaces to the end space node.
Dissipation value for all the internal surfaces to the end space.
Surface Areas of all the internal surfaces to the end space.
End Space Fluid Heat Transfer Coefficient scaling factor.
See Units.
Set of 2 Grids showing numerical Output Data for the winding parameters.
Wire Slot Fill (Wdg Area) Actual model slot fill (wire area/winding area).
Copper Slot Fill (Wdg Area) Actual model slot fill (copper area/winding area).
Slot Fill (Slot Area) Slot fill calculated from copper area / slot area (including
liner and wedge).
Slot Area Slot area - including liner and wedge but not slot opening.
Wedge Area Slot wedge area/slot (includes slot area used to force
conductors to base of slot).
Winding Area (+Liner) Slot area available for winding before liner insertion.
Winding Area Slot area available for winding after liner insertion.
Covered Wire Area Wire area in each slot (includes wire insulation) -
calculation based on winding layer model.
Copper Area Copper area in each slot (not including wire insulation).
Liner-Lam Gap Area The area of the gap between the Slot Liner and lamination
in each slot.
Liner-Lam Imp Area The area of the impregnation between the Slot Liner and
lamination in each slot.
Liner Area Liner area/slot - not including gap between liner and
lamination.
Coil-Divider Area Coil divider area (used to separate coils in same slot of
non-overlapping windings. See also Coil Divider Width.
Covered Wire Area (Cond) Round conductor wire area/slot (includes wire insulation) -
calculation based on actual conductors rather than layers.
Copper Area [Layer n] Copper area of winding layer = n of winding layer model.
Impreg Area [Layer n] Impreg area of winding layer = n of winding layer model.
Ins Area [Layer n] Wire insulation area of winding layer = n of winding layer
model.
Wire Copper Factor Proportion of wire area that is copper rather than
copper+insulation.
The proportional increase in winding thermal conductivity between layers due to the end winding is
calculated using:
EWdg Layer Cond Mult = Weight Copper Total / Weight Copper Active
Where the end-winding copper weight is calculated using the EWdg MLT or EWdg Fill parameters.
Winding Periphery Winding outer periphery (total for all slots - not
include winding push-back).
Periphery [Liner-Lam Gap] Average Periphery of Gap between liner and stator
lamination [one whole slot].
Periphery [Liner-Lam Gap (Slot Bottom)] Average Periphery gap between liner and stator
lamination slot bottom [one whole slot].
Periphery [Liner-Lam Gap (Tooth Side)] Average Periphery gap between liner and stator
lamination at tooth sides [one whole slot].
Periphery [Liner (Slot Bottom)] Average Periphery slot liner at bottom of slot [one
slot].
Periphery [Liner (Tooth Side)] Average Periphery slot liner covering tooth sides
[one slot].
Periphery [Ins Outer Layer (Slot Bottom)] Average Periphery of copper insulation for outer
layer at slot bottom (one slot)
Periphery [Ins Outer Layer (Tooth Side)] Average Periphery of copper insulation for outer
layer at tooth side (one slot)
Periphery [Impreg Outer Layer (Slot Bottom)] Average Periphery of impregnation for outer layer
at slot bottom (one slot)
Periphery [Impreg Outer Layer (Tooth Side)] Average Periphery of impregnation outer layer at
tooth side (one slot)
The drive sheet gives the calculated values for the following machine parameters for the
electromagnetic model:
Wedge Area
Slot Opening Area
Volume Copper Active
Volume Copper EWdg Front
Volume Copper EWdg Rear
These pages show numerical Output Data for different miscellaneous parameters.
See:
Miscellaneous Output Parameters [Housing],
Miscellaneous Output Parameters [Fluids],
Miscellaneous Output Parameters [Airgap],
Miscellaneous Output Parameters [Convergence]
Miscellaneous Output Parameters [Other].
Housing Length [Active]: Axial length of active section of housing. See Housing Length and
Axial Dimensions
Housing Length [Front]: Axial length of front overhang section of housing. See Housing
Length and Axial Dimensions
Housing Length [Rear]: Axial length of rear overhang section of housing. See Housing
Length and Axial Dimensions
Housing Bore [Active]: Bore of active section of housing.
Housing Bore [Front]: Bore of front overhang section of housing.
Housing Bore [Rear]: Bore of rear overhang section of housing.
Fin Number: The number of fins on the outside of the housing. See Housing
Type.
Fin Spacing [Average]: Average fin spacing. Used in Convection Heat Transfer calculation
of finned sections of the motor housing.
Fin Height [Average]: Average fin height. Used in Convection Heat Transfer calculation of
finned sections of the motor housing.
Corner Cutout Spacing: Average housing corner cutout spacing. Used in Convection Heat
Transfer calculation of corner cutout sections of the motor housing (see Corner Cutout [%]).
Corner Cutout Depth: Average housing corner cutout depth. Used in Convection Heat
Transfer calculation of corner cutout sections of the motor housing (see Corner Cutout [%]).
Area [Housing - Active Total]: Surface area of main housing [total] available for cooling to ambient
by convection/radiation. See Radiation Heat Transfer and Convection Heat Transfer.
Area [Housing - Front Total]: Surface area of front overhanging housing section [total] available
for cooling to ambient by convection/radiation. See Radiation Heat Transfer and Convection Heat
Transfer.
Area [Housing - Rear Total]: Surface area of rear overhanging housing section [total] available for
cooling to ambient by convection/radiation. See Radiation Heat Transfer and Convection Heat
Transfer.
Housing Convection Inlet Area: Area between housing outer surface and the Cowling or fin outer
cover (Housing = Axial Fins (Cv)) inner surface. This area is used in the calculation of Forced
Convection Default Velocity from Forced Convection Default Flow Rate. See Blown Over [Input Data
Editor].
This output sheet shows the air properties at ambient and the temperature supplied by the user in
the Air Property Data sheet.
Air Specific Heat [Tsupplied] Specific heat of air at Air Property Temperature.
Air Dynamic Viscosity [Tsupplied] Dynamic viscosity of air at Air Property Temperature.
Air Kinematic Viscosity [Tsupplied] Kinematic viscosity of air at Air Property Temperature.
Air Prandtl Number [Tsupplied] Prandtl number of air at Air Property Temperature.
Used in combination with Re[Airgap] Critical - Vortex and Re[Airgap] Critical - Turbulent to determine
the flow regime and thus calculate Nu[Airgap] and h[Airgap] using the theory of convective heat
transfer between two concentric cylinders rotating relative to each other [1].
[1] Taylor, G.I.: 'Distribution of Velocity and Temperature between Concentric Cylinders', Proc Roy
Soc, 1935, 159, PtA, pp 546-578
Transient Maximum Function The maximum number of function evaluations calculated for the
Evaluations transient calculation
Transient Maximum Function The cycle in the duty cycle where the maximum number of function
Evaluations [cycle] evaluations occured.
Transient Maximum Function The period in the duty cycle where the maximum number of function
Evaluations [period] evaluations occured.
Transient Maximum Function The point in the duty cycle where the maximum number of function
Evaluations [point] evaluations occured.
Transient maximum error The maximum error in the transient calculation between successive
final evaluations.
Transient maximum error The cycle in the duty cycle where the maximum error occured.
[cycle]
Transient maximum error The period in the duty cycle where the maximum error occured.
[period]
Transient maximum error The point in the duty cycle where the maximum error occured.
[point]
Different parameters are displayed in this datasheet depending on the machine type.
dt [Winding - Housing] Temperature difference between the winding hot spot and housing.
The output datasheet provides a wide range of calculated values for the Housing Water Jacket
model.
There are the following output data subsheets for the Housing Water Jacket:
Temperature
Flow
Dissipation
Area
Thermal Resistance
WJ Fluid Inlet Temp [Active] Water jacket fluid inlet temperature [Active Section]. Equal to
outlet temperature of previous stage or WJ Fluid Inlet Temperature
if 1st stage.
WJ Fluid Inlet Temp [Front] Water jacket fluid inlet temperature [Front Section]. Equal to outlet
temperature of previous stage or WJ Fluid Inlet Temperature if 1st
stage.
WJ Fluid Inlet Temp [Rear] Water jacket fluid inlet temperature [Rear Section]. Equal to outlet
temperature of previous stage or WJ Fluid Inlet Temperature if 1st
stage.
WJ Outlet Temp [Active] Calculated value of the water jacket fluid outlet temperature (Active
Section). Calculated iteratively within the thermal calculation using
the formula:
(Power Dissipated) = (Volume Flow Rate) * (Delta Temperature) *
(Fluid Density) * (Fluid Specific Heat)
WJ Outlet Temp [Front] Calculated value of the water jacket fluid outlet temperature (Front
Section). Calculated iteratively within the thermal calculation using
the formula:
(Power Dissipated) = (Volume Flow Rate) * (Delta Temperature) *
WJ Outlet Temp [Rear] Calculated value of the water jacket fluid outlet temperature (Rear
Section). Calculated iteratively within the thermal calculation using
the formula:
WJ Fluid Average Temp [Active] Calculated value of the average water jacket fluid temperature
(Active Section). Calculated iteratively within the thermal
calculation using the formula:
WJ Fluid Average Temp [Active] = [ (WJ Fluid Inlet Temp [Active])
+ (WJ Outlet Temp [Active]) ] / 2
WJ Fluid Average Temp [Front] Calculated value of the average water jacket fluid temperature
(Front Section). Calculated iteratively within the thermal calculation
using the formula:
WJ Fluid Average Temp [Front] = [ (WJ Fluid Inlet Temp [Front]) +
(WJ Outlet Temp [Front]) ] / 2
WJ Fluid Average Temp [Rear] Calculated value of the average water jacket fluid temperature (Rear
Section). Calculated iteratively within the thermal calculation using
the formula:
WJ Fluid Average Temp [Rear] = [ (WJ Fluid Inlet Temp [Rear]) +
(WJ Outlet Temp [Rear]) ] / 2
WJ T[Fluid - A] Network This is the temperature of the node in the model whose
temperature should be close to the average fluid temperature of the
Active Section of the Housing Water Jacket Fluid. Motor-CAD
calculates the average fluid temperature and tries to adjust the
model so that this temperature agrees.
WJ T[Fluid - F] Network This is the temperature of the node in the model whose
temperature should be close to the average fluid temperature of the
Front Section of the Housing Water Jacket Fluid. Motor-CAD
calculates the average fluid temperature and tries to adjust the
model so that this temperature agrees.
WJ T[Fluid - R] Network This is the temperature of the node in the model whose
temperature should be close to the average fluid temperature of the
Rear Section of the Housing Water Jacket Fluid. Motor-CAD
calculates the average fluid temperature and tries to adjust the
model so that this temperature agrees.
This page gives the calculated flow values for the housing water jacket.
WJ Channel Number [Calc] Average number of housing water spiral groove cooling channels.
WJ Channel Fluid Velocity Calculated value of the water jacket channel fluid velocity (Active
[Active] Section). This is calculated from the WJ Fluid Volume Flow Rate
and the cross-sectional area of the channel.
WJ Channel Fluid Velocity [Front]Calculated value of the water jacket channel fluid velocity (Front
Section). This is calculated from the WJ Fluid Volume Flow Rate
and the cross-sectional area of the channel.
WJ Channel Fluid Velocity [Rear] Calculated value of the water jacket channel fluid velocity (Rear
Section). This is calculated from the WJ Fluid Volume Flow Rate
and the cross-sectional area of the channel.
WJ Parallel Paths
Housing WJ Flow [Active]
Housing WJ Total Flow [Active]
WJ Channel Pressure [Active]
WJ Channel Pressure [Front]
WJ Channel Pressure [Rear]
WJ Channel Flow Resistance
[Active]
WJ Channel Dissipation Calculated value of loss dissipated by water jacket fluid (Active
[Active] Section).
WJ Channel Dissipation [Front] Calculated value of loss dissipated by water jacket fluid (Front
Section).
WJ Channel Dissipation [Rear] Calculated value of loss dissipated by water jacket fluid (Rear
Section).
h-WJ Channel [Active] Calculated value of convection heat transfer coefficient of water
jacket channel (Active Section).
h-WJ Channel [Front] Calculated value of convection heat transfer coefficient of water
h-WJ Channel [Rear] Calculated value of convection heat transfer coefficient of water
jacket channel (Rear Section).
Fluid Temperature rise at 40C This is the temperature rise of the fluid per unit loss with the fluid
properties at 40C. This is used for the fluid paths calculations (see
fluid temperature calculation method).
This sheet gives different calculated values for the Housing Water Jacket including:
WJ Channel Area [Active] Calculated value of housing water channel area available for cooling
(Active Section).
WJ Channel Area [Front] Calculated value of housing water channel area available for cooling
(Front Section).
WJ Channel Area [Rear] Calculated value of housing water channel area available for cooling
(Rear Section).
Rt Housing-WJ Fluid [Active] Calculated value of thermal resistance between the housing and water
jacket fluid (Active Section).
Rt Housing-WJ Fluid [Front] Calculated value of thermal resistance between the housing and water
jacket fluid (Front Section).
Rt Housing-WJ Fluid [Rear] Calculated value of thermal resistance between the housing and water
jacket fluid (Rear Section).
Rt Housing-WJ Wall [Active] Calculated thermal resistance value for duct wall (Active section)
Rt Housing-WJ Wall [Front] Calculated thermal resistance value for duct wall (Front section)
Rt Housing-WJ Wall [Rear] Calculated thermal resistance value for duct wall (Rear section)
The output data sheet provides a wide range of calculated values for the Shaft Spiral Groove model.
There are the following output data sub-sheets for the Shaft Spiral Groove:
Temperature
Flow
Dissipation
Area
Thermal Resistance
SG Fluid Inlet Temp [Rear] Shaft spiral groove fluid inlet temperature [Rear Section]. Equal to outlet
temperature of previous stage or Shaft Groove Fluid Inlet Temperature if
1st stage.
SG Outlet Temp [Rear] Calculated value of the shaft spiral groove fluid outlet temperature (Rear
Section). Calculated iteratively within the thermal calculation using the
formula:
SG Fluid Average Temp [Rear] = [ (SG Fluid Inlet Temp [Rear]) + (SG
Outlet Temp [Rear]) ] / 2
SG Fluid Inlet Temp [Active]Shaft spiral groove fluid inlet temperature [Active Section]. Equal to
outlet temperature of previous stage or Shaft Groove Fluid Inlet
Temperature if 1st stage.
SG Outlet Temp [Active] Calculated value of the shaft spiral groove fluid outlet temperature (Active
Section). Calculated iteratively within the thermal calculation using the
formula:
SG Fluid Average Temp Calculated value of the average shaft spiral groove fluid temperature
[Active] (Active Section). Calculated iteratively within the thermal calculation
using the formula:
SG Fluid Average Temp [Active] = [ (SG Fluid Inlet Temp [Active]) + (SG
Outlet Temp [Active]) ] / 2
SG Fluid Inlet Temp [Front] Shaft spiral groove fluid inlet temperature [Front Section]. Equal to outlet
temperature of previous stage or Shaft Groove Fluid Inlet Temperature if
1st stage.
SG Outlet Temp [Front] Calculated value of the shaft spiral groove fluid outlet temperature (Front
Section). Calculated iteratively within the thermal calculation using the
formula:
SG Fluid Average Temp [Front] = [ (SG Fluid Inlet Temp [Front]) + (SG
Outlet Temp [Front]) ] / 2
See also Shaft Spiral Groove [Output Data], Shaft Spiral Groove [Input Data Editor]
SGroove Fluid Velocity Calculated value of the shaft spiral groove channel fluid velocity (Active
[Active] Section). This is calculated from the Shaft Groove Fluid Volume Flow
Rate and the cross-sectional area of the channel.
SGroove Fluid Velocity Calculated value of the shaft spiral groove channel fluid velocity (Front
[Front] Section). This is calculated from the Shaft Groove Fluid Volume Flow
Rate and the cross-sectional area of the channel.
SGroove Fluid Velocity Calculated value of the shaft spiral groove channel fluid velocity (Rear
[Rear] Section). This is calculated from the Shaft Groove Fluid Volume Flow
Rate and the cross-sectional area of the channel.
See also Shaft Spiral Groove [Output Data], Shaft Spiral Groove [Input Data Editor]
SGroove Dissipation Calculated value of loss dissipated by shaft spiral groove fluid (Active
[Active] Section).
SGroove Dissipation [Front] Calculated value of loss dissipated by shaft spiral groove fluid (Front
Section).
Calculated value of loss dissipated by shaft spiral groove fluid (Rear
SGroove Dissipation [Rear] Section).
h-SGroove [Active] Calculated value of convection heat transfer coefficient of shaft spiral
groove channel (Active Section).
h-SGroove [Front] Calculated value of convection heat transfer coefficient of shaft spiral
groove channel (Front Section).
h-SGroove [Rear] Calculated value of convection heat transfer coefficient of shaft spiral
groove channel (Rear Section).
See also Shaft Spiral Groove [Output Data], Shaft Spiral Groove [Input Data Editor]
See also Shaft Spiral Groove [Output Data], Shaft Spiral Groove [Input Data Editor]
Rt Shaft-SGrooveFluid Calculated value of thermal resistance between the shaft and shaft spiral
[Active] groove fluid (Active Section).
Rt Shaft-SGroove[Front] Calculated value of thermal resistance between the shaft and shaft spiral
groove fluid (Front Section).
Rt Shaft-SGrooveFluid Calculated value of thermal resistance between the shaft and shaft spiral
[Rear] groove fluid (Rear Section).
See also Shaft Spiral Groove [Output Data], Shaft Spiral Groove [Input Data Editor]
The output data sheet provides a wide range of calculated values for the Slot Water Jacket model.
There are the following output data sub-sheets for the Slot Water Jacket:
Temperature
Flow
Dissipation
Area
Thermal Resistance
See also Slot Water Jacket [Output Data], Slot Water Jacket [Input Data Editor]
See also Slot Water Jacket [Output Data], Slot Water Jacket [Input Data Editor]
See also Slot Water Jacket [Output Data], Slot Water Jacket [Input Data Editor]
See also Slot Water Jacket [Output Data], Slot Water Jacket [Input Data Editor]
See also Slot Water Jacket [Output Data], Slot Water Jacket [Input Data Editor]
When the Spray Cooling method is selected the predicted thermal resistance for all surfaces in the
flow path are given in the table shown above and in Thermal Resistance [Output Data]
Note: that at present these are not shown in the Schematic but can be viewed in the circuit editor if
required.
The calculated values of the Rotor Water Jacket are given in the output data sheet.
These are a set of tables that give the output data calculated for the wet rotor functionality.
There are the following output data subsheets for the Wet Rotor:
Temperature
Flow
Dissipation
Area
Thermal Resistance
See also Wet Rotor, Wet Rotor Convection Correlation and Wet Rotor Data [Input Data Editor].
Thermal Resistance:
Rt [Front EWdg Bore]: Calculated value of thermal resistance between the front
end-winding bore and end-space node when the Wet Rotor calculation option is selected.
Rt [Shaft Front]: Calculated value of thermal resistance between the front
shaft overhang and end-space node when the Wet Rotor calculation option is selected.
Rt [Wet Rotor [F] - Ambient]:
Rt [Stator Bore]:
Rt [Rotor Surface]:
Rt [Mid Gap - Ambient]:
Rt [Coil Divider]:
Rt [Wet Rotor Coil Divider - Ambient]:
Rt [Slot Wedge]:
Rt [Wet Rotor Slot Wedge - Ambient]:
Rt [Rear EWdg Bore]: Calculated value of thermal resistance between the rear
end-winding bore and end-space node when the Wet Rotor calculation option is selected.
Rt [Shaft Rear]: Calculated value of thermal resistance between the rear
shaft overhang and end-space node when the Wet Rotor calculation option is selected.
Rt [Wet Rotor [R] - Ambient]:
See Wet Rotor, Wet Rotor Convection Correlation and Wet Rotor Data [Input Data Editor].
Fluid Inlet Temp [Wet Rotor F] Wet rotor fluid inlet temperature [end space - front]. Equal to outlet
temperature of previous stage or Wet Rotor Inlet Temperature if 1st
stage.
Fluid Outlet Temp [Wet Rotor Calculated value of the wet rotor fluid outlet temperature (end
F] space - front). Calculated iteratively within the thermal calculation
using the formula below.
Fluid Average Temp [Wet Calculated value of the average wet rotor fluid temperature (end
Rotor F] space - front). Calculated iteratively within the thermal calculation
using the formula below.
Fluid Inlet Temp [Airgap] Wet rotor fluid inlet temperature [Airgap]. Equal to outlet
temperature of previous stage or Wet Rotor Inlet Temperature if 1st
stage.
Fluid Outlet Temp [Airgap] Calculated value of the wet rotor fluid outlet temperature (airgap).
Calculated iteratively within the thermal calculation using the
formula below.
Fluid Average Temp [Airgap] Calculated value of the average wet rotor fluid temperature (airgap).
Fluid Inlet Temp [Wet Rotor R] Wet rotor fluid inlet temperature [end space - rear]. Equal to outlet
temperature of previous stage or Wet Rotor Inlet Temperature if 1st
stage.
Fluid Outlet Temp [Wet Rotor Calculated value of the wet rotor fluid outlet temperature (end
R] space - rear). Calculated iteratively within the thermal calculation
using the formula below.
Fluid Average Temp [Wet Calculated value of the average wet rotor fluid temperature (end
Rotor R] space - rear). Calculated iteratively within the thermal calculation
using the formula below.
Wet Rotor Fluid Axial Velocity The wet rotor fluid axial airgap velocity. This is calculated from the
[Airgap] Wet Rotor Volume Flow Rate and the cross-sectional area of the
airgap.
Fluid Peripheral Vel [Airgap] The wet rotor fluid peripheral airgap velocity. This is calculated to
be 1/2 the rotor surface peripheral velocity - i.e. fluid at centre of
gap is average of stationary stator bore velocity and rotating rotor
surface velocity.
Fluid Effective Vel [Airgap] The wet rotor fluid effective airgap velocity. This is a combination of
the axial and peripheral airgap fluid velocities and is calculated
using the formula below.
Wet Rotor Fluid Axial Velocity The wet rotor fluid axial velocity (end space - front). This is
[Wet Rotor F] calculated from the Wet Rotor Volume Flow Rate and the cross-
sectional area of the end space.
Fluid Peripheral Vel [Wet The wet rotor fluid peripheral velocity (end space - front). This is
Rotor F] calculated to be 1/2 the rotor surface peripheral velocity - i.e. fluid
at centre of space is average of stationary end-winding bore
velocity and rotating shaft surface velocity.
Fluid Effective Vel [Wet Rotor The wet rotor fluid effective (end space - front) velocity. This is a
F] combination of the axial and peripheral (end space - front) fluid
velocities and is calculated using the formula below.
Wet Rotor Fluid Axial Velocity The wet rotor fluid axial velocity (end space - rear). This is
[Wet Rotor R] calculated from the Wet Rotor Volume Flow Rate and the cross-
sectional area of the end space.
Fluid Peripheral Vel [Wet The wet rotor fluid peripheral velocity (end space - rear). This is
Rotor R] calculated to be 1/2 the rotor surface peripheral velocity - i.e. fluid
at centre of space is average of stationary end-winding bore
velocity and rotating shaft surface velocity.
Fluid Effective Vel [Wet Rotor The wet rotor fluid effective (end space - rear) velocity. This is a
R] combination of the axial and peripheral (end space - front) fluid
velocities and is calculated using the formula below.
Wet Rotor Fluid Velocity [Slot The wet rotor fluid slot wedge axial velocity. This is calculated from
Wedge] the Wet Rotor Slot Wedge Volume Flow Rate and the cross-
sectional area of the slot wedge.
Wet Rotor Fluid Velocity [Coil The wet rotor fluid coil divider axial velocity. This is calculated from
Divider] the Wet Rotor Coil Divider Volume Flow Rate and the cross-
sectional area of the coil divider.
Effective Velocity Formula (Wet Rotor Effective Section Velocity) = sqrt[ sqr(Fluid Axial Vel
[Section]) + sqr(Fluid Peripheral Vel [Section]) ]
Dissipation [Front EWdg Calculated value of loss dissipated from front end-winding bore to
Bore] the end-space fluid.
Dissipation [Shaft Front] Calculated value of loss dissipated from front shaft overhang to the
end-space fluid.
Dissipation [Stator Bore] Calculated value of loss dissipated from stator bore to the wet rotor
fluid.
Dissipation [Rotor Surface] Calculated value of loss dissipated from rotor surface to the wet
rotor fluid.
Dissipation [Rear EWdg Bore] Calculated value of loss dissipated from rear end-winding bore to
the end-space fluid.
Dissipation [Shaft Rear] Calculated value of loss dissipated from rear shaft overhang to the
end-space fluid.
Fluid Dissipation[ESpaceF] Calculated value of loss dissipated from the front end-winding bore
and front shaft overhang to the wet rotor fluid.
Fluid Dissipation[Airgap] Calculated value of loss dissipated from stator bore and rotor
surface to the wet rotor fluid.
Fluid Dissipation[ESpaceR] Calculated value of loss dissipated from the rear end-winding bore
and rear shaft overhang to the wet rotor fluid.
h[Front EWdg Bore] Calculated value of convection heat transfer coefficient of front end-
winding bore.
h[Shaft Front] Calculated value of convection heat transfer coefficient of front shaft
overhang.
h[Rear EWdg Bore] Calculated value of convection heat transfer coefficient of rear end-
winding bore.
h[Shaft Rear] Calculated value of convection heat transfer coefficient of rear shaft
overhang.
Area [Stator Bore] Calculated value of the stator bore area cooled by the wet
rotor fluid.
Area [Rotor Surface] Calculated value of the rotor surface area cooled by the wet
rotor fluid.
Area [Front EWdg Bore] Calculated value of the front end-winding bore - area cooled
by the wet rotor fluid.
Area [Shaft Front] Calculated value of the front shaft overhang - area cooled by
the wet rotor fluid.
Wet Rotor Flow Area [Slot Wedge] Calculated cross sectional flow area of slot wedge (if slot
wedge flow is enabled).
Wet Rotor Flow Area [Coil Divider] Calculated cross sectional flow area of coil divider (if coil
divider flow is enabled).
Area [Coil Divider] Calculated value of the coil divider - area cooled by the wet
rotor fluid.
Area [Slot Wedge] Calculated value of the slot wedge - area cooled by the wet
rotor fluid.
Wet Rotor Flow Area [Airgap] Calculated cross sectional flow area of airgap.
Area [Rear EWdg Bore] Calculated value of the rear end-winding bore - area cooled
by the wet rotor fluid.
Area [Shaft Rear] Calculated value of the rear shaft overhang - area cooled by
the wet rotor fluid.
The output datasheet provides a wide range of calculated values for the through ventilation model.
There are the following output data subsheets for Through Ventilation:
Temperature
Flow
Dissipation
Area
Thermal Resistance
For each section of the flow path in a Through Ventilation machine the Inlet, Outlet and Average
temperatures are calculated.
Calculated values for fluid flow resistance, pressure drop and volume flow rate for all the main
sections of the Through Ventilation flow circuit (inlet, outlet, airgap, rotor and stator) are given in the
upper section of the editor shown above.
The calculated Through Ventilation flow path component dissipation is also displayed in Losses
[Output Data]
Calculated Through Ventilation flow area data is displayed in the table above.
Values of h - Tvent and Area - Tvent in a Through Ventilation machine are used to calculate the
thermal resistance values that are shown above and put in the Schematic
The blown over flow rates shown above are only displayed when a cowling is used.
In Output Data [Numerical Values] calculated value of the Blown Over fluid volume flow rate and fluid
temperature rise are available for various sections of the flow (Inlet, Active, etc). These values are
only available if a Cowling is fitted - the flow cross-sectional area is required. Also the effect of
Leakage from the open fin channels is taken into account.
Data relating to the Stall Copper Loss Distribution calculation. Data only relevant and presented
In the above there is 120W of copper loss and the volume of copper in the active and end windings is
equal.
The drive sheet gives the calculated values for the following material properties for the
electromagnetic model based on the component temperatures:
Magnet resistivity
Magnet Resistivity at 20C
Magnet temperature
17 DataFiles
17.1 *.mot datafile
Input data for the Motor-CAD program is stored in a file having the *.mot file extension. The name
and path of the currently loaded datafile is included in the program title-bar:
The default file location folder is set using the Default File Locations Pull Down Menu option and is
saved in Defaults.INI
Reading old datafiles using a newer version of the program & Reading new datafiles using
older versions of the program:
In order to ease file maintenance, *.mot file are stored using the windows INI (text) format. This
makes it possible for old datafiles to be used with newer versions of the program - the new
parameters being set to default values. Also, new datafiles can be used with older versions of the
program - any new parameters being ignored.
When an old datafile is read into a newer program version, a warning is displayed showing all new
parameters and their default values:
The datafile format used is the windows INI standard. The data is grouped into the following
categories:
Units
Calculation & Design Options
Fluid Data
Dimensions
Interface Gaps
Winding Design
Miscellaneous
Loss Variation with Temperature & Load
Loss Data
Radiation Data
Convection Data
Through Ventilation
End Space Convection Data
Material Data
Weight Data
Transient Data
Wet Rotor
Water Jacket
Rotor Water Jacket
Slot Water Jacket
Shaft Spiral Groove Spray Cooling
The Solids Database and fluids database information is stored in *.mdb files.
They contains material data for solid and fluid materials typically used in electric machines, i..e
thermal conductivity, specific heat, density and visosity. The information is stored in the database
file in the form shown below:
The default database files are given in the Default File Locations
When Motor-CAD is started if the Solids.mdb file is not in existence then it is created and a copy of
several standard materials are added to it.
Care should be taken if using the default values from the databases to ensure that these
values are accurate for the materials being used.
The default location for the file is a sub-folder called transientResults under the folder where the
current .mot datafile is located. The format of the datafile is shown below:
The Custom Graph Settings feature allows the user to save and load customised graph settings. The
settings for the Legend, Point Type, Line Colour and whether a node is plotted are can be saved and
loaded from .gsf files (graph settings file).
The default location for the file is a sub-folder called graphsettings under the folder where the current
.mot datafile is located.
The file Defaults.INI stores the users defaults, e.g. Default Units, Default Motor Type and Default File
Locations
The file Defaults.INI is saved in different folders depending on the version of Windows.
In XP:
C:\Documents and Settings\My_Username\Application Data\MotorDesign\Motor-CAD
In versions prior to v2 of Motor-CAD it was stored in the folder where Motor-CAD was installed.
History.INI is also stored in this folder.
17.6 History.INI
Links to the last 10 filenames opened are stored in the History.INI file.
The file History.INI is saved in different folders depending on the version of Windows.
In XP:
C:\Documents and Settings\My_Username\Application Data\MotorDesign\Motor-CAD
In Vista:
C:\Users\My_Username\AppData\roaming\MotorDesign\Motor-CAD
The Flux2D model export is selected from the menu as shown below:
Model Type:
Steady State Thermal This can be useful to check the Motor-CAD winding model.
See the Winding Temperature Model Tutorial.
Transient Thermal This sets the model for a transient thermal analysis.
Transient Magnetic This sets the model for a transient magnetic analysis. Notes:
this is not yet fully featured. It will provide a model template
which the user will have to adapt to meet their requirements.
(.e.g. winding, materials etc)
in the magnets.
Solver:
Automatic Flux2D will automatically select solver.
SuperLU This is the fastest solver, however it requires a lot of memory
so may sometimes fail due to lack of memory.
ICCG Slower solver which has a lower memory requirement than
SuperLU.
GMRes
BiCGStab
Analysis Options:
Automatic Solve The Flux2D model will be solved by the script.
Calculate ISOValues Flux2D postprocessing will be run to generate a colour
contour.
Calculate ISOLines Flux2D postprocessing will be run to generate contour lines.
Note: due to the open functionality of Flux2D it is not possible to test that every possible geometry
will be generated and solve correctly. Please contact support@motor-design.com if you have
problems exporting a geometry to Flux2D.
The Motor-CAD link to Portunus allows users to include a reduced order thermal network generated
by Motor-CAD in a larger system simulation in Portunus.
Create a reduced node circuit for the machine in Motor-CAD as shown below.
Open Portunus and then click on the Motor-CAD icon in the toolbar as shown below. The thermal
circuit is transferred automatically from Motor-CAD to Portunus using ActiveX.
The circuit can be checked by running the DC analysis in Portunus and selecting to view the final
temperatures.
It can be seen from the results above that they compare well with those in Motor-CAD.
The thermal circuit from Motor-CAD can now be coupled to electrical, mechanical, electromagnetic
and thermal system simulations in Portunus.
losses recalculated to give a more accurate simulation. This last step can be performed using an
automated iterative calculation that makes sure that the temperatures at which the losses are
calculated in PC-BDC are the same as those predicted in Motor-CAD - which itself used the PC-
BDC losses to predict the temperatures.
Start PC-BDC with a standard example - [File] - [New] - choose a surface radial motor and press
[OK].
You can calculate the performance using the [Analysis] - [Dynamic Design] option. It will first ask
you to choose some materials - select the defaults and press [OK].
You can view the performance graphs using [Results] - [Simulation Graphs]:
You can view the performance such as Torque, Losses, Efficiency, etc using [Results] - [Design
Sheet]:
The problem is that the losses are calculated at temperatures within PC-BDC that are based on a
very simple model that needs test data to set it up. The alternative is to use Motor-CAD to predict
the temperatures.
Save the PC-BDC file in a folder of your choice - I usually put it in the folder where I store Motor-CAD
data files:
Now open Motor-CAD. Select the [File] - [Open] pull down menu. Set the [Files of type] to (*.bd4)
and select the PC-BDC file you have just created. Press [Open].
Motor-CAD cannot directly read the file as the format is specific to the PC-BDC software. However,
the SPEED Import/Export dialog box shown below appears so the user can set the data which they
wish to import and export.
A .mot file with the same name as the .bd4 file will have been created and put in the same folder as
the .bd4 file. You can view the names and folder locations of the two files by making the dialog box
larger:
Leave the default option to import all Speed data and not export any Motor-CAD data:
The Radial Cross section will now change to show the effects of the import of the PC-BDC geometry
(12 slots, etc):
PC-BDC did not have any information on the housing so we may now select an housing type more
appropriate for out application:
The Axial Cross Section has imported all available data from PC-BDC - however we will need to
change a few parameters that were not available:
In this case we will reduce the [Motor Length] and [Bearing Dia(F)] to give a better looking motor:
The winding hotspot is only 41.9C (20C ambient). We can now go back to PC-BDC and set a larger
current in the [Template Editor] - ISP increased from 15A to 30A in this case:
The shaft torque (Tshaft) is now 1.86Nm (was 0.94Nm) and the copper loss (WCu) is now 77.1W
(was 19.7W).
We can now import the new losses to predict the new temperature. Press the button in Motor-CAD.
We do not need to import the geometry so de-select [SPEED geometry]. Press [OK].
The winding hotspot is now 94.4C. The new losses that were imported are shown below:
The main problem with the analysis we have carried out so far is that the temperatures at which the
losses are calculated will be different to the predicted temperatures. We can now carry out an
iterative calculation to make sure that the losses and temperatures in PC-BDC and Motor-CAD
agree. Press the button and select [Iterate to Converged Solution]. This will then automatically set
the [Speed Import] - [Speed losses] and the [Motor-CAD Export] - [Motor-CAD temperatures] and
you will not be able to de-select them as they are both required for an iterative solution.
Press [OK] and you will see that PC-BDC and Motor-CAD both carry out calculations. When
convergence has been achieved the following message will be displayed:
The losses calculated in PC-BDC are shown below (go to PC-BDC and open the Design Sheet):
Open the PC-BDC Template Editor (Thermal page) and see that Motor-CAD has correctly set the
magnet and winding temperatures:
In this example we showed a link between PC-BDC and Motor-CAD for the brushless permanent
magnet motor. The link is also available for:
1. Introduction:
It easy to automatically pass data between the SPEED (motor design software from the SPEED
Laboratory at the University of Glasgow) and Motor-CAD electric motor design software packages.
Losses and geometry data can be passed from SPEED to Motor-CAD and temperatures and
geometry back. A routine also exists within Motor-CAD to iterate until the losses and temperatures
within the two packages converge, i.e. the losses depend upon the temperature and vice versa. The
data transfer is carried out using ActiveX technology. ActiveX is a Windows standard for passing of
data and for calling one software from another. Both software packages have full ActiveX capabilities
such that both can then be run as a black box calculation tool from other ActiveX enabled packages
such as Excel VBA and Matlab. This makes is possible to develop very powerful integrated design
tools to carry out automated optimisation (i.e. using such methods as genetic algorithms) and/or
sensitivity analysis. It can also be used to carry out advanced no linear calculations such as
transient modelling of dynamic loads.
The dialog box that controls the data that is to be imported from SPEED and Exported from Motor-
CAD is activated using the [File] - [SPEED] pull down menu. If a Motor-CAD datafile has not yet
been created then the user is presented with the [Save As] dialog box such that the current data can
be saved. The [SPEED Import/Export] dialog box shown in Figure 1 should then be activated.
The names of both the SPEED and Motor-CAD files that will be linked are displayed at the top of the
dialog. It is assumed that the two files reside in the same folder and have the same name (other then
the part after the dot, i.e. .bd4 and .mot in this case).
If the SPEED file does not yet exist then the [SPEED Export] section of the dialog box will not be
enabled (as shown in Figure 1). In this case the SPEED file name that will be created is displayed at
the top of the dialog - this file is created once an export is started using the [OK] button. Once both
files exit then the [SPEED Export] functions become enables as shown in Figure 2.
Figure 1: SPEED Import/Export dialog box (*.mot files exists & .db4 file does not exist)
Figure 2: SPEED Import/Export dialog box (*.bd4 & *.mot files exist)
The user simply ticks the data that is to be imported and exported and then presses the [OK] button
to start the transfer. For some time limited versions of the SPEED software a dialog box similar to
that shown in Figure 3 will be shown before the SPEED software is activated - if this is the case then
the user needs to press the [OK] button to continue with the transfer. Note that in some cases this
dialog can be hidden under other open windows on the desktop such that the user thinks the
computer is locked up - this will be fixed in future versions of the SPEED software).
The user can select losses, geometry and thermal duty cycle to be imported from SPEED to Motor-
CAD.
Note that the import function is carried out before the export - thus if the user has both the import
SPEED geometry and export Motor-CAD geometry check-boxes ticked the available dimensions
from SPEED overwrite the Motor-CAD equivalent values
The losses are calculated using the SPEED Static or Dynamic Simulation algorithm in PC-BDC,
Steady State Simulation in PC-IMD, etc. The losses that can be transferred directly to the equivalent
Motor-CAD loss parameter are transferred directly (i.e. stator copper loss and tooth and yoke iron
losses). The exception is the friction and windage loss which must first be split into its individual
components (windage and front and rear bearing friction losses). To achieve this division use is made
of the parameters shown in Figure 4. For instance if [Windage/Friction Ration] = 0.7 then 70% is
taken as being and windage and 30% as friction. If [Friction Front/Rear Ratio] = 0.6 then 60% is in
the front (drive end) bearing and 40% in the rear bearing.
Figure 4: Parameters to split friction and windage loss into individual components
In the case of the induction machine then the stray load loss is transferred. This is divided in Motor-
CAD between the stator tooth and rotor tooth nodes according to the
[Calculation Options].
Motor-CAD has a model that can be used to calculate the variation in losses with temperature and
load. This option is especially useful when dealing with duty cycle transient loads as the user need
only specify the losses at one operating point (torque, speed and temperature) - the losses at other
operating point are then calculated from this operating point. This particular option is only available
for permanent magnet motors as the torque is proportional to current and the back-emf is
proportional to speed. This makes it relatively easily to define its full electromagnetic performance
characteristics using just a few parameters. The other motor types are in comparison very non-linear
making this process impossible.
The parameters that are required to perform the loss calculation with temperature and load are
shown in Figure 5. These are automatically filled in from the PC-BDC calculation.
Basically in the calculation the loss in flux in the machine due to the magnet heating up is taken into
account in the calculation of the current for a given torque. This leads to an increase in the copper
loss on top of that due to the resistance getting larger with increasing winding temperatures. The
small variation in iron loss is also modelled. In the future it will also be possible to have a direct
linkage with the SPEED software from the transient thermal analysis such that the losses are
updated with time during the transient. It is hoped to release this new feature in the near future.
Figure 5: Loss Variation with Temperature & Load Data (BPM Motors)
The user can select the amount of data that is to be imported using the check boxed shown in
Figure 6. A brief description of what data is transferred is given below:
Radial Cross Section - The basic dimensions as listed in the SPEED Outline Editors are imported
into Motor-CAD - with appropriate transformations of radii as used in SPEED to diameters as
used in Motor-CAD. The only axial dimensions that are imported are the active component
lengths (stator, magnet and rotor). In some circumstances it is not possible to import the exact
geometry - in such cases the closest match will be made (e.g. for some of the more complex
rotor bar shapes in PC-IMD). In some cases it is not possible at all to make a reasonable
import, e.g. in the case of outside rotor BPM motors or for IPM motors - in such cases Motor-
CAD will eventually be updated to accommodate such configurations - please contact Motor
Design Ltd if you have any particular requirements.
Axial Cross Section - The amount of axial information that can be imported into Motor-CAD is
dependent upon the motor type. The BPM motor has the least capabilities. PC-IMD and PC-
SRD do have a basic model of the housing and endcaps. If available these are imported together
with the active component lengths.
End Winding Overhang - The PC-BDC and PC-IMD have a calculation for the axial length between
extremities of the end windings (LaxPack parameter in SPEED). This is based on compressed
end winding (using the EndFill parameter in SPEED). LaxPack is used to set the amount by
which the end winding extend beyond the stator lamination in Motor-CAD. The value of Ext
(winding extension) is also imported. In the case of PC-SRD the parameter LgthEnds (length
over ends) is used in a similar fashion.
Winding Design - Basically Motor-CAD imports the number of individual wires in a slot. The wire
bare diameter is also imported (also covered diameter if available). If WireSpec = AWG_Table is
used in the SPEED software then the actual AWG wire gauge is imported rather than the wire
size. From the winding design in SPEED an appropriate choice is made for if the winding is of
the Overlapping or Non-Overlapping type. The mean length per turn is also imported together
with the liner thickness.
Intelligent geometry scaling - when selected then the dimensions that are not available in SPEED
are automatically set to sensible values in Motor-CAD. If this option is not selected then the
dimensions in Motor-CAD are unchanged.
In almost all import cases the user will need to set further data to describe such features as fin
dimensions and gaps around the winding after the transfer. This is because the thermal problem is
much more 3-dimensional than the electromagnetic problem.
The user can select temperatures and geometry to be exported from Motor-CAD to SPEED.
The average winding and ambient temperatures are exported for all machine types. In the case of the
BPM motor the magnet temperature is also exported. The cage temperature is exported in induction
motors.
This is basically the reverse of importing the geometry from SPEED. Again appropriate choices are
made to give the best fit of data such as induction motor rotor bar types (where an exact fit between
the constructional features does not exist between the two software packages). As with the import
feature the user has choices as to what data is exported - Figure 7.
Note that the import function is carried out before the export - thus is the user has both the import
SPEED geometry and export Motor-CAD geometry check-boxes ticked the available dimensions
from SPEED overwrite the Motor-CAD equivalent values.
4. Calculation Options:
The user can select the following options using the selector shown in Figure 8:
Single Shot Import/Export
Iterate to Converged Solution
In this case the user simply ticks the data that is to be imported and exported and then presses the
[OK] button to start the transfer. Only the current values are imported/exported.
If this option is selected then the Import SPEED Losses ( ) and Export Motor-CAD
21 3D Geometry Export
Motor-CAD has options for exporting 3D geometry. The export dialog is launched from the
menu as shown below:
or may be launched directly from the 3D Viewer by clicking the 3D Geometry Export button
shown below
When launched this way, the default export option of STL format is automatically selected.
STL format
Stereo Lithography Format (*.stl) files describe the surfaces of 3D shapes. The format is
available as an import option in many CAD packages. There are 2 formats available, Binary
(default) and ASCII. It is recommended that Binary is used, since ASCII files can be become
very large.
The geometry that is exported is based on the selection of visible components in the 3D
viewer component tree. When the export button is clicked, each selected component
generates a stl file, the name of each is based on the component name and group to which it
belongs.
The generated stl files shown below may be selected for import into your selected CAD
software.
22 Star-CCM+ export
Motor-CAD can export 3D geometry to Star-CCM+ in the form of java macros. These macros are
generated automatically by Motor-CAD, which can then be run from Star-CCM+.
The geometry selected from the 3D viewer is used to prepare a set of macros for import. The
components displayed in the 3D viewer will be exported. Those components that are not displayed
will not be exported.
The primary macro is generated and by default is named the same as the active mot file. Additional
macros are generated for each selected component which are used by the primary macro to
construct the geometry in Star-CCM+ as shown below:
In Star-CCM+ select the Play Macro option and select the primary Java macro which was generated
by Motor-CAD.
23 ActiveX Links
23.1 ActiveX
ActiveX links are useful for automation of data transfer between programs and to run program
functions from an external program.
For instance the SPEED and Motor-CAD complement each others capabilities (electromagnetic &
thermal) and it is useful to pass data between the two programs:
for accurate calculations imperative that carry out both calculations simultaneously
losses depend on temperature and temperature on losses
The good thing is that both SPEED and Motor-CAD have ActiveX capabilities. In fact a specific
SPEED Import/Export function has been built directly into Motor-CAD - this uses ActiveX technology
to pass data and run Speed from Motor-CAD. It is also possible to run both SPEED and Motor-CAD
from any other ActiveX enabled software such as:
Excel & Word using VBA Macro
Matlab
Motor-CAD scripting interface using VBS
this can be done by using the SPEED Import/Export capabilities in Motor-CAD or by writing
specific scripts to link to the designers own software.
optimisation
able to run automated optimisation studies using such methods as genetic algorithms
run automated sensitivity analysis studies and/or check tolerance build up and so produce
robust designs
allow users to introduce their own formulations into Motor-CAD
keep secret their own efforts
add new circuit components such as integrated power electronics
ActiveX must be enabled in Motor-CAD by selecting the Register for ActiveX option as shown below.
When enabled Motor-CAD will write to the registry each time that it is run.
For Motor-CAD to be able to register for ActiveX the computer user must have administrator rights on
the machine and the "Run as Administrator" option should also be used when registering Motor-
CAD for the first time. If for any reason the ActiveX registration fails then it can cause a Motor-CAD
error and Motor-CAD will no longer run. If this occurs then the Register for ActiveX setting can be
disabled by editing the defaults.ini file.
An example of a Motor-CAD VBA ActiveX Script can be viewed by following the following link -
ActiveX Excel Example
An example of a Motor-CAD Matlab ActiveX Script can be viewed by following the following link -
ActiveX Matlab Example
See also Motor-CAD ActiveX methods and ActiveX Input and Output Parameters.
Typical commands to set input variables and read input or output variables are shown below:
Res = mcad.GetVariable("Schematic_Node_Winding_Outer_Layer",MyOuter_Winding_Node)
If Res = -1 Then
MsgBox "Error getting Outer_Wining_Node"
End If
If Res = -2 Then
MsgBox "Variable not found"
End If
Res = mcad.SetVariable("Wet_Rotor_Fluid_Volume_Flow_Rate",My_Flow_Rate)
If Res = -1 Then
MsgBox "Error setting Flow Rate"
End If
If Res = -2 Then
MsgBox "Variable not found"
End If
The parameter name can be copy and pasted from an Excel file that gives all available Input and
Output Parameters. All input parameters should be input in SI units (except dimensions in mm and
areas in mm2).
mcad.DoSteadyStateAnalysis
mcad.DoTransientAnalysis
mcad.Quit
The easiest way to write a new script is to edit and old one. The Motor-CAD installation CD gives a
few examples are given on the Motor-CAD installation CD. Copies can also be found in the ActiveX
folder placed in the folder where Motor-CAD is installed using the setup.exe command. This is
typically:
C:\Program Files\Motor-CAD\ActiveX
1. Motor-CAD_VBA_Macro_Script_Example_1.xls
Variation in winding temperature with Fin Spacing in a Natural Convection cooled motor with
Radial Fins
2. Motor-CAD_VBA_Macro_Script_Example_2.xls
Run a thermal transient as set up in the loaded .mot datafile and plots the transient in Excel.
Uses the Append_Next_Transient_To_Existing_Data call to change parameters part way
through a transient calculation.
Example (1):
mcad.Visible = True
mcad.LoadFromFile "c:\motor-cad\radial_test1.mot"
'---------- MAKE SURE HOUSING TYPE IS SET TO RADIAL FINS [Type 3] --------------
Res = mcad.SetVariable("Housing_Type", 3)
If Res = -1 Then
MsgBox "Error setting Housing_Type"
End If
If Res = -2 Then
MsgBox "Variable not found"
End If
'+++ GET INNER WINDING NODE (depends on the number of layers in the winding) +++
Res = mcad.GetVariable("Schematic_Node_Winding_Outer_Layer", MyOuter_Winding_Node)
If Res = -1 Then
MsgBox "Error getting Outer_Wining_Node"
End If
If Res = -2 Then
MsgBox "Variable not found"
End If
'set up a string so we can get the temperature of the inner most winding layer
'string details are given in the Output Parameters worksheet
Winding_Temperature_String = "Node_Temp[" + CStr(MyInner_Winding_Node) + "]"
'Initial Spacing
My_Fin_Spacing = 1
My_Fin_Thickness
My_Fin_Spacing = My_Fin_Spacing + 2
Next
mcad.Quit
End Sub
Example (2):
mcad.Visible = True
'create a Motor-CAD data file based on the default motor (we could load an
existing .mot if required instead)
mcad.SaveToFile "c:\motor-cad\Test_Trans.mot"
Motor-CAD
'we could however vary them independently
If Res = -2 Then
MsgBox "Stator_Copper_Loss_@Ref_Speed not found"
End If
'+++++++++++GET INNER WINDING NODE (this can change if winding changes) +++++++++
++
'details of Schematic_Node_Winding_Outer_Layer parameter given in Output
Parameters worksheet in Excel
Res = mcad.GetVariable("Schematic_Node_Winding_Outer_Layer", MyOuter_Winding_Node)
If Res = -1 Then
MsgBox "Error getting Outer_Wining_Node"
End If
If Res = -2 Then
MsgBox "Schematic_Node_Winding_Outer_Layer not found"
End If
'set up a string so we can get the temperature of the inner most winding layer
'string details are given in the Output Parameters worksheet
Winding_Temperature_String = "Node_Temp[" + CStr(MyInner_Winding_Node) + "]"
Housing_Temperature_String = "Node_Temp[" + CStr(MyHousing_Node) + "]"
Next
End Sub
Information on the full range of Motor-CAD input and output variables can be viewed by following the
following link - ActiveX Input and Output Variables
For Matlab script examples follow the link - ActiveX Matlab Example
Matlab Code:
mcad = actxserver('MotorCAD.AppAutomation');
Typical commands to set input variables and read input or output variables are shown below:
[funcresult,Weight.BackIron] = invoke(mcad,'GetVariable','Weight_Stator_Lam_Back_Iron');
invoke(mcad,'SetVariable','Stator_Copper_Loss_@Ref_Speed', Copper_Loss);
The parameter name can be copy and pasted from the Input or Output Parameter worksheet. All
input parameters should be input in SI units (except dimensions in mm and areas in mm2).
invoke(mcad,'DoSteadyStateAnalysis');
invoke(mcad,'DoTransientAnalysis');
FilePath = 'C:\Matlab\';
FileName = strcat(FilePath,'model_name.mot');
res = invoke(mcad,'LoadFromFile',FileName);
if res ~= 0
disp('Could not load MOT file.');
invoke(mcad,'quit');
return;
end;
res = invoke(mcad,'SaveToFile',FileName);
invoke(mcad,'Quit');
The easiest way to write a new script is to edit and old one. The Motor-CAD installation CD gives a
few examples are given on the Motor-CAD installation CD. Copies can also be found in the ActiveX
folder placed in the folder where Motor-CAD is installed using the setup.exe command. This is
typically:
C:\Program Files\Motor-CAD\ActiveX
1. Run a design with 4 different load points (speed/torque) and create 4 new .mot files (one for
each). The motor has a water jacket that we can input the flow rate in l/min:
The main routine is called MotorCAD_Matlab_Script_Example_1.m
The main routine calls the routine called Thermal_Calc.m
MotorCAD_Matlab_Script_Example_1.m
MotorCAD_File_Name = 'Matlab_Test';
% Run a design with 4 different load points (speed/torque) and create 4 new .mot
files (one for each)
% The motor has a water jacket that we can input the flow rate in l/min
spacer = '=====================================================================';
% Speed Data
Speed_ = [500 1000 3000 5000];
Copper_Loss_ = [300 500 400 300];
WJ_Flow_Rate_ = [3 5 5 7]; %l/min
n=1;
WJ_Inlet_Temp = 40;
Thermal_Calc %call a separate .m file to calculate the performance
spacer_ = spacer
n=2;
WJ_Inlet_Temp = 40;
Thermal_Calc %call a separate .m file to calculate the performance
spacer_ = spacer
n=3;
WJ_Inlet_Temp = 40;
Thermal_Calc %call a separate .m file to calculate the performance
spacer_ = spacer
n=4;
WJ_Inlet_Temp = 40;
Thermal_Calc %call a separate .m file to calculate the performance
spacer_ = spacer
Thermal_Calc.m
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Open Motor-CAD
mcad = actxserver('MotorCAD.AppAutomation');
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
'Number_Winding_Layers');
[funcresult,Node.Magnet] = invoke(mcad,'GetVariable',
'Schematic_Node_Magnet');
[funcresult,Node.Back_Iron] = invoke(mcad,'GetVariable',
'Schematic_Node_Stator_Lam_Back_Iron');
[funcresult,Node.Tooth] = invoke(mcad,'GetVariable',
'Schematic_Node_Stator_Lam_Tooth_Middle');
[funcresult,Node.Tooth_Tip] = invoke(mcad,'GetVariable',
'Schematic_Node_Stator_Lam_Tooth_Tip');
% set up a string so we can get the temperature of the inner most winding layer
% string details are given in the Excel Output Parameters worksheet
String.Winding_Inner_Temperature = sprintf('Node_Temp[%g]',Node.Inner_Winding);
String.Winding_Outer_Temperature = sprintf('Node_Temp[%g]',Node.Outer_Winding);
String.Magnet_Temperature = sprintf('Node_Temp[%g]',Node.Magnet);
String.EWdgF_Temperature = sprintf('Node_Temp[%g]',Node.EWdgF);
String.EWdgR_Temperature = sprintf('Node_Temp[%g]',Node.EWdgR);
String.Housing_Temperature = sprintf('Node_Temp[%g]',Node.Housing);
String.Back_Iron_Temperature = sprintf('Node_Temp[%g]',Node.Back_Iron);
String.Tooth_Temperature = sprintf('Node_Temp[%g]',Node.Tooth);
String.Tooth_Tip_Temperature = sprintf('Node_Temp[%g]',Node.Tooth_Tip);
[funcresult,Temp.Magnet] = invoke(mcad,'GetVariable',String.
Magnet_Temperature);
[funcresult,Temp.Winding_Average] = invoke(mcad,'GetVariable','T_
[Winding_Average]');
[funcresult,Temp.WJ_Inlet_Temp] = invoke(mcad,'GetVariable',
'WJ_Fluid_Inlet_Temperature');
[funcresult,Temp.WJ_Outlet_Temp] = invoke(mcad,'GetVariable',
'WJ_Fluid_Outlet_Temp_[Active]');
invoke(mcad,'Quit');
Information on the full range of Motor-CAD input and output variables can be viewed by following the
following link - ActiveX Input and Output Variables
For Excel VBA examples follow the link - ActiveX Excel Example
All the input and output parameters that can be used in an ActiveX script can be obtained directly
from Motor-CAD.
The window below is displayed. The list can be searched and saved to a separate text file if required.
Note: Model parameters can also be changed in this window by entering the new values and
pressing return.
24 Command Line
Motor-CAD can be called from the command line using the syntax shown below.
The example below will open Motor-CAD without showing any user interface, load the model file and
then run a transient analysis. The transient results will be placed in the results.txt file.
25 Units
25.1 Units of Length
The units of length (mm or inches) are indicated on the status line.
Both mm & inches are available. The selection is made using the Main Pull Down Menu command O
ptions - Units which initiates the Units Selection Form.
Please contact Motor Design Ltd if you require units that have not been implemented within Motor-
CAD.
25.2 Units
A selection of units are available (SI & British). The selection is made using the Main Pull Down
Menu command Options - Units which initiates the Units Selection Form shown below.
The Default Units to be used in Motor-CAD are stored in the file Defaults.INI (same directory as
Motor-CAD) - these can be loaded and saved using the two buttons provided:
[Load Default Units]
[Save As Default Units]
Please contact Motor Design Ltd if you require units that have not been implemented within Motor-
CAD.
The file Defaults.INI stores the users defaults, e.g. Default Units, Default Motor Type and Default File
Locations
The file Defaults.INI is saved in different folders depending on the version of Windows.
In XP:
C:\Documents and Settings\My_Username\Application Data\MotorDesign\Motor-CAD
In versions prior to v2 of Motor-CAD it was stored in the folder where Motor-CAD was installed.
History.INI is also stored in this folder.
26 Defaults
26.1 Default Folder
The file Defaults.INI stores the users defaults, e.g. Default Units,Display Size, Font Size, Default
Motor Type and Default File Locations
In versions prior to v2 it was stored in the folder where Motor-CAD was installed.
The file Defaults.INI stores the users defaults, e.g. Default Units, Default Motor Type and Default File
Locations
The file Defaults.INI is saved in different folders depending on the version of Windows.
In XP:
In versions prior to v2 of Motor-CAD it was stored in the folder where Motor-CAD was installed.
History.INI is also stored in this folder.
Used to set the default window size used to display Motor-CAD. The information is stored in the
Defaults.INI file.
The Default Display Size function is set using the Pull Down Menu. When initiated it brings up the
form shown below:
Used to set the default Motor Type that is displayed when Motor-CAD is first executed. The
information is stored in Defaults.INI. Before a selection has been made the default motor type is set
to the first valid motor type licence that is found on the Security Key
When the Default Motor Type Pull Down Menu command is initiated it brings up the form shown
below:
When the Default File Locations Pull Down Menu command is initiated it brings up the form shown
below:
are stored in the file whenever the transient calculation is run. Caution: this file can get very large.
This data can be viewed using the Transient Results File Analysis
Duty Cycle File Location - the location where the duty cycles are loaded from by default.
Transient Test Data Location - the location where the transient test data is loaded from by default
FE Results Data Location - the location of result files created by the Finite element module
Matrix Export Location - the location to export the matrix files when these are exported. See pull
down menu for details.
Reduced Node Model Location - the location to export the reduced model file that specifies the
reduced node circuit.
Index
-I-
Impingement Jet Correlation 575
Iron Loss Coefficients 820