Motor Protection

May 31, 2017

Tom Ernst
GE Grid Solutions
Motor Relay Zone of Protection

-Electrical Faults
-Abnormal Conditions
-Thermal Overloads
-Mechanical Failure

2
• Setting of the motor protection relay is based
on the motor datasheets information and
system configuration

• Datasheets are normally provided by motor

manufacturer

• System configuration data can be obtained

from single line diagram

3
CT Selection

869

4
Phase CTs
• The CT should be nominally sized at or
greater than motor FLA
• The CT must have an accuracy class high
enough so that the current waveform
presented to the relay will allow the
overcurrent to operate
• Higher CT ratio is generally better from a
saturation point of view than a lower CT ratio

5
Phase CTs
• Our Motor has a FLA of 413 Amps
• Our maximum fault current is 22KA

6
 Phase CTs
C100 400/5
INPUT PARAMETERS: ENTER: Saturation Curve CALCULATED:
Inverse of sat. curve slope = S= 16 --- Vs Rt = Total burden resistance = Rw + Rb = 0.295
RMS voltage at 10A exc. current = Vs = 125 volts rms slope pf = Total burden pow er factor = 1.000
Turns ratio = n2/1= N= 80 --- = 1/S Zb = Total burden impedance = 0.295
mf gr's
Winding resistance = Rw = 0.195 ohms Ve data Tau1 = System time constant = 0.027
Burden resistance = Rb = 0.100 ohms Lamsat = Peak flux-linkages corresponding to Vs 0.469
Burden reactance = Xb = 0.004 ohms volts log-log plot, w = Radian freq = 376.99
System X/R ratio = XoverR = 10.0 --- rms equal RP = Rms-to-peak ratio = 0.37410
Per unit offset in primary current = Off = 1.00 -1<Off<1 decade A = Coefficient in instantaneous ie
spacing
Per unit remanence (based on Vs) = lrem 0.00 --- versus lambda curve: ie = A * l^S : 4.89E+06
Symmetrical primary fault current = Ip = 22,000 amps rms dt = Time step = 0.000083
Ie amps rms 10 Lb = Burden inductance = 0.00001

Thick lines: Ideal (blue) and actual (black) secondary current in amps vs time in seconds.
Thin lines: Ideal (blue) and actual (black) secondary current extracted fundamental rms value, using a simple DFT with a one-cycle window.

800
600
400
200
0
-200
-400
-600
-0.017 0.000 0.017 0.033 0.050 0.067 0.083 0.100 0.117 0.133 0.150

Saturated Magnitude
Trace 7
 Phase CTs
C100 600/5
INPUT PARAMETERS: ENTER: Saturation Curve CALCULATED:
Inverse of sat. curve slope = S= 16 --- Vs Rt = Total burden resistance = Rw + Rb = 0.300
RMS voltage at 10A exc. current = Vs = 125 volts rms slope pf = Total burden pow er factor = 1.000
Turns ratio = n2/1= N= 120 --- = 1/S Zb = Total burden impedance = 0.300
mf gr's
Winding resistance = Rw = 0.200 ohms Ve data Tau1 = System time constant = 0.027
Burden resistance = Rb = 0.100 ohms Lamsat = Peak flux-linkages corresponding to Vs 0.469
Burden reactance = Xb = 0.004 ohms volts log-log plot, w = Radian freq = 376.99
System X/R ratio = XoverR = 10.0 --- rms equal RP = Rms-to-peak ratio = 0.37410
Per unit offset in primary current = Off = 1.00 -1<Off<1 decade A = Coefficient in instantaneous ie
spacing
Per unit remanence (based on Vs) = lrem 0.00 --- versus lambda curve: ie = A * l^S : 4.89E+06
Symmetrical primary fault current = Ip = 22,000 amps rms dt = Time step = 0.000083
Ie amps rms 10 Lb = Burden inductance = 0.00001

Thick lines: Ideal (blue) and actual (black) secondary current in amps vs time in seconds.
Thin lines: Ideal (blue) and actual (black) secondary current extracted fundamental rms value, using a simple DFT with a one-cycle window.

500
400
300
200
100
0
-100
-200
-300
-0.017 0.000 0.017 0.033 0.050 0.067 0.083 0.100 0.117 0.133 0.150

Increase the ratio - less

saturation 8
 Phase CTs
C200 600/5
INPUT PARAMETERS: ENTER: Saturation Curve CALCULATED:
Inverse of sat. curve slope = S= 16 --- Vs Rt = Total burden resistance = Rw + Rb = 0.300
RMS voltage at 10A exc. current = Vs = 300 volts rms slope pf = Total burden pow er factor = 1.000
Turns ratio = n2/1= N= 120 --- = 1/S Zb = Total burden impedance = 0.300
mf gr's
Winding resistance = Rw = 0.200 ohms Ve data Tau1 = System time constant = 0.027
Burden resistance = Rb = 0.100 ohms Lamsat = Peak flux-linkages corresponding to Vs 1.125
Burden reactance = Xb = 0.004 ohms volts log-log plot, w = Radian freq = 376.99
System X/R ratio = XoverR = 10.0 --- rms equal RP = Rms-to-peak ratio = 0.37410
Per unit offset in primary current = Off = 1.00 -1<Off<1 decade A = Coefficient in instantaneous ie
spacing
Per unit remanence (based on Vs) = lrem 0.00 --- versus lambda curve: ie = A * l^S : 4.04E+00
Symmetrical primary fault current = Ip = 22,000 amps rms dt = Time step = 0.000083
Ie amps rms 10 Lb = Burden inductance = 0.00001

Thick lines: Ideal (blue) and actual (black) secondary current in amps vs time in seconds.
Thin lines: Ideal (blue) and actual (black) secondary current extracted fundamental rms value, using a simple DFT with a one-cycle window.

500
400
300
200
100
0
-100
-200
-300
-0.017 0.000 0.017 0.033 0.050 0.067 0.083 0.100 0.117 0.133 0.150

Increase ratio & C-rating – almost no

Saturation
9
Motor Data Sheets

Motor Performance Data Thermal Limit Curves

10
 Motor Thermal Limit Curves

 Thermal Limit Curves:

A
A. Cold Running Overload
C
B. Hot Running Overload
B
C. Cold Locked Rotor Curve
D D. Hot Locked Rotor Curve
E. Acceleration curve @ 80%
E F
rated voltage
F. Acceleration curve @100%
voltage

11
 Motor Thermal Parameters
G

 Motor Data Sheet Parameters K

G. Temperature Rise, Insulation Class
H. Full Load Current
I. Locked Rotor Current
J. Locked Rotor Time; Cold/Hot
K. Number of Starts per hour; 12
Motor Specifications

Information required to set Thermal

Model:
• Motor FLA
• Locked rotor current
• Locked rotor time hot & cold
• Stopped & running cool time constants
• Service factor
• Motor thermal damage curves

13
Settings Example
Select CT Rating, Voltage Sensing
Voltage Sensing : Enter
the connection type,
secondary volts and ratio.
VTratio = 14400/120 =
120:1
Vsec = Vnom/VTratio =
13800/120 = 115 V
Phase CT: The phase CT should be
chosen such that the FLA is 75% to 150%
of CT primary. Since the FLA is 297 a
300:5 CT may be chosen.
Ground CT: Zero sequence core balance
CT is used for high impedance grounded
systems. The primary rating should be
large enough to assure that the CT can
handle all potential fault ground levels
without saturating.
• 50 A >> systems with less than 50
amps of ground fault current.
• 200 A or 300 A >> systems with up to
300 amps of ground fault current.
• No ground CT required on low
impedance or solidly grounded
systems (Use neutral functions (3I0 is
calculated from the phase CTs).
• Secondary rating can be same as
phase CTs (1A/5A) or special 50:0.025
A.
14
Settings Example Motor FLA: Set as specified by the
Select FLA, Ground CT data sheets.
Overload Factor: This is the pick-
up of the OL curve. Set 10-15%
above data sheet service factor.
NP Voltage, HP & Poles: Set as
specified in the data sheets.
Load Average Calc. Period: Set
this longer than the oscillatory
duration of oscillating loads like
reciprocal compressors. Set at 0
for non-oscillatory loads.
Max Acceleration Time: Set this to
the longest acceleration time
expected plus a margin (the
acceleration time trip function is
enabled separately - see
Protection > Group X >> Motor).

15
Settings Example
Select Overload Curve for Thermal Model

Overload Curve
Set the overload curve below cold thermal limit and above hot thermal limit. If only hot curve
is provided by manufacturer, then must set at or below hot thermal limit
The best fitting curve is time dial multiplier 9 in this example.
Note that this is a 3 dimensional curve: f(A ,T, TCU), TCU = thermal capacity used. Curve
values given are for TCU = 0 (40 °C stator temp). The curve represents TCU = 100%.

16
Settings Example
Select Overload Curve TD Multiplier for Thermal Model

Overload Curve
Set the overload curve TD multiplier below cold thermal limit and
above hot thermal limit. If only hot curve is provided by mfgr,
then must set at or below hot thermal limit. The best fitting curve
TD multiplier is 9 in this example.
This can be verified with Hot Stall Time of 30s at 540% FLA by
using the standard overload curve equation above.

17
Settings Example
Select Overload Curve for Thermal Model
Select overload curve using Hot Stall Time and Locked Rotor Current when
Overload Curves are not available:
Example: For Hot Stall Time = 30s and LRA = 540% FLA

Substitute in the above equation:

30s = TD MULTIPLIER x 2.2116623

(0.02530337 x 4.42 + 0.05054758 x 4.4)

TD MULTIPLER = 30 x (0.02530337 x 4.42 + 0.05054758 x 4.4)

2.2116623
= 9.66
SELECT TDM 9 (which is below this intersection
point) 18
Settings Example
Determine Unbalance Bias K Factor for Thermal Model
Unbalance Bias Of Thermal Capacity
Enable the Unbalance Bias of
Thermal Capacity so that the heating
effect of unbalance currents is added
to the Thermal Capacity Used.

K=175/LRA 2 = 175/ 5.4 2 = K=230/LRA 2 = 230/ 5.4 2 =

6 8
(Typical) (Conservative)

19
Settings Example
Stopped & Running Cool Time Constants

Stopped and Running Cool Time Constants

This information is usually supplied by the motor
manufacturer but is not part of the data that was given
with this motor. If RTD’s are present and will be wired to
the relay biasing of the thermal model will be used so it is
not critical to have these cooling times from the
manufacturer: the default values of 15 and 30 minutes can
be used for the running and stopped cool times
respectively. 20
Settings Example
Determine Hot/Cold Safe Stall Ratio for Thermal Model (method 1)

LRTHOT
HCR 
LRTCOLD

Hot/Cold Ratio =
30/35 = 0.86

Hot/Cold Curve Ratio

The hot/cold curve ratio is calculated by simply dividing the
hot safe stall time by the cold safe stall time or use the motor
thermal limits curve. For this example, both are available.
Using the data sheets the Hot/Cold Curve Ratio equals 30 / 35
= 0.86

21
Settings Example
Determine Hot/Cold Safe Stall Ratio for Thermal Model (method 2)

Overload Curve Method

Hot/Cold Curve Ratio
If the thermal limits curves are being used to determine
the
HOT/COLD ratio proceed as follows:

• From the thermal limits curves run a line perpendicular

to the current axis that intersects the hot and cold
curves at the stall point
LRTcold = 8sec
• Draw lines from each points of intersection to the time
axis.
• Record the corresponding times. In this case, 6 and 8
seconds respectively.
LRThot = 6sec • The Hot/cold ratio can now be calculated as follows:
= 6s/8s = 0.75

LRC = 5.4FLA NOTE:

• If hot and cold times are not provided and only one
curve is given verify with the manufacturer that it is the
hot curve ( which is the worst case), then the Hot/ Cold
ratio should be set to 1.0

22
Settings Example
Determine RTD Bias Setpoints for Thermal Model

Enable RTD Biasing

This will enable the temperature from the Stator RTD sensors, to be
included in the calculations of Thermal Capacity. RTD bias model
determines the Thermal Capacity Used based on the temperature
of the Stator and is separate from the overload model for
calculating Thermal Capacity Used. RTD biasing is a back up
protection element which accounts for such things as loss of
cooling or unusually high ambient temperature. This measured
temperature is used to bias or modify the thermal capacity value
stored in the motor relay.

23
Settings Example
Determine RTD Bias Setpoints for Thermal Model
• Motor relay will use the calculated
MAX POINT thermal capacity unless the RTD thermal
TEMP: 155°C capacity is higher.
TCU: 100%
• This feature will not trip the motor at the
max point temp unless the average
current is greater than the overload
pickup setting
RTD Bias Function
MID POINT
TEMP: 130°C Set to Enabled/YES
TCU: 25% RTD Bias Minimum
Set to 40 ° C which is the ambient temperature obtained from
MIN POINT the data sheets.
TEMP: 40°C RTD Bias Center Point
TCU: 0% The center point temperature is set to the motor’s hot running
temperature and is calculated as follows:
Temperature Rise of Stator + Ambient Temperature.
The temperature rise of the stator is 80 ° C + 10% hot spot
allowance, obtained from the data sheets.
Therefore, the RTD Center point temperature is set to 900C +
400C or 130 ° C.
RTD Bias Maximum
This setpoint is set to the rating of the insulation or slightly
less. A class F insulation is used in this motor which is rated at
155 ° C, so setting should be 155 ° C.
24
Settings Example
Determine RTD Bias Setpoints for Thermal Model

MAX POINT
TEMP: 155°C
TCU: 100%

MID POINT
TEMP: 130°C
TCU: 25%

MIN POINT
TEMP: 40°C
TCU: 0%

25
Settings Example
Enable Start Inhibit

Enable Start Inhibit

This function will limit starts when the
motor is already hot. The motor relay
learns the amount of thermal capacity
used at start. If the motor is hot, thus
having some thermal capacity used, the
relay will not allow a start if the available
thermal capacity is less than the required
thermal capacity for a start.

If Start Inhibit is not used, must wait until

Thermal Capacity Used (TCU) falls below
15% before the motor can be re-started.

Using Start Inhibit allows one to start a

motor sooner.

26
TCU / Start Inhibit Example

Thermal Capacity required to start

For example, if the THERMAL CAPACITY USED for the
last 5 starts is 24, 23, 27, 25, and 21% respectively, the
LEARNED STARTING CAPACITY is 27% × 1.25 =
33.75% used.
Thermal Capacity used due to
Overload
If the motor had been running in an overload condition
prior to stopping, the thermal capacity would be some
value; say 80%.

If Motor is Stopped:
When the motor has cooled and the level of thermal
capacity used has fallen to 66%, a start will be
permitted.

27
Settings Example
Starts/Hr, Time Between Starts

Starts/Hour
Starts/Hour can be set to the # of cold starts as per the data sheet.
For this example, it is 2

Time Between Starts

In some cases, the motor manufacturer will specify the time between
motor starts. In this example, this information is not given so this
feature can be disabled or set at a typical 20 min between starts.

28
Settings Example
VFD Support Functions

Bypass Switch
If the VFD has a bypass switch then set this for the contact input that
is ON when the switch is closed.
Starting Frequency
Traditionally, the frequency tracking function started at 50/60 Hz and
then looked at zero crossings of several cycles to determine the
correct actual frequency. This caused the first 5 – 10 cycles of current
measurement to be wrong when the motor was started from a VFD.
Starting frequency feature allows the tracking to start at a more
realistic frequency (6 Hz in this case).

29
Advanced Diagnostics
Broken Rotor Bar Detection
 The Broken Rotor Bar element uses two different algorithms to
detect broken or cracked rotor bars:
 Power Based Coherent Demodulation: This technique uses
multiplication of voltage and current samples thereby shifting
the fundamental to DC and fault frequency to lower closer to DC
value, to detect the broken rotor bar component. This method is
running when voltage is available and is meeting MOTOR
VOLTAGE SUPERVISION setting check.
 Conventional current based FFT method: In case voltage is not
available or the voltage magnitude is lower than the MOTOR
VOLTAGE SUPERVISION setting value, the algorithm switches
to analyzing the frequency spectrum from current samples only,
to detect the broken rotor bar component.
 Alarm settings are based on an increase in dB as each motor will
exhibit a different signature when healthy.

30
Advanced
Diagnostics - BRB
FFT of Stator Current of
Induction Machine with
Rotor Bar Fault – signature
is only about 12 Hz off of
fundamental

The FFT of the

resultant multiplied
signal – more
robust signature
than with the
current only FFT
method signal.

31
Advanced Diagnostics
BRB

Pickup is set based on

initial in-service dB
measurements taken
when the motor is
known to be healthy
plus a change margin
(~15% increase).

32
Advanced Diagnostics
Stator Inter-Turn Fault
 The Stator Inter-Turn Fault element uses sequence components to
detect stator turn failure of the induction machine.
 Local heating caused by shorted turns can rapidly cause
additional damage to adjacent windings and stator iron
 Alarm to avoid additional damage

Normalized cross-coupled impedance ratio:

Znp/Zpp = (V2 – Znn*I2)/V1 ~ 0 under balanced non-fault conditions
Zpp = positive sequence impedance
Znp = cross-coupled negative-to-positive sequence impedance
V1 = positive sequence voltage (motor terminals)
V2 = negative sequence voltage (motor terminals)
I2 = negative sequence current (motor terminals)
Znn = negative sequence impedance

33
Advanced Diagnostics
Stator Inter-Turn Fault

Operating quantity:
OP = Znp/Zpp –
ZUBbase

ZUBbase =
normalized
cross-coupled
impedance
ratio under
non-fault
conditions

34
Thank You

Questions?