Académique Documents
Professionnel Documents
Culture Documents
" "
" "
" "
" "
s cs s cs
cs c cs c
s cs s cs
cs c cs c
L
E A
K
where the order of the degrees of freedom is { }
" " ) )
, , , v u v u . The A, E, and / are the
cross sectional area, 1oungQs ,elastic- modulus and a+ial length respecti$ely. The c and s in the
matri+ stand for =os ,- and Sin ,- respecti$ely. The orientation of the bar and the angle are
shown below.
O
U1
V1
U2
V2
Y
X
26
This element does not ha$e any stiffness associated with rotational degrees of freedom.
Therefore, bending and torsion effects are not included in this model nor is it possible to load the
structure with moments. Also, the element, in the manner it is used in this analysis, does not
ha$e the ability to model large deformations and will not warn the user in case of buc;ling type
failures ,i.e. geometric nonlinearities-. Similarly, this type of analysis does not ha$e the ability to
correctly model stresses which are not in the elastic range of the material ,i.e. material
nonlinearities-.
V. STEP BY STEP INSTRUCTIONS FOR BUILDING THE TRUSS MODEL USING PATRAN
!reliminaries for using *S=>!AT:A( include.
2- /og on to the computer
2- =hange to the directory that you wish to contain your analysis results
5- /eft clic; STA:T ,lower left corner of the (T des;top-, go to !:CA:A*S, then top *S=
,common-, then to *S= !atran 0. This will bring up the *S=>!atran !rogram.
In the instructions below, the following abbre$iations and terms will be used.
TM = Top Men. This refers to the horiBontal menu options residing at the top of the screen after
!AT:A( has been initiated.
RM = R!"#t Men. This refers to the menus that pop up after an option has been chosen from
the top menu. These menus reside on the far right side of the !AT:A( des;top.
SM = S$o%&!n'te Men. This referees to the menus that pop up from options selected in the
right menu.
C(!)* = @nless otherwise stated, this indicates a clic; with the left mouse button.
Bo(&+')e will indicate te+t that occurs in the !AT:A( menus.
Italics te+t will indicate te+t that you must enter into te+t bo+es in the !AT:A( menus or te+t that
you choose in a menu scroll bo+.
2. Cur first step is to create a new database.
'rom the T* choose F!(e
In the resulting pull down menu choose Ne, D't'$'-e
A S* called Ne, D't'$'-e pops up
Turn off ,button up- Mo&!+. P%e+e%en)e-
@nder Ne, D't'$'-e N'/e enter truss.db
=lic; O0
A menu called Ne, Mo&e( P%e+e%en)e- will appear
Select To(e%'n)e to be $'-e& on t#e /o&e(
Set Mo&e( D!/en-!on = 2.0
An'(.-!- )o&e = MSC/N'-t%'n
An'(.-!- T.pe = -t%)t%'(
=lic; O0
2. The geometry of the truss will be determined ne+t.
'rom the T* choose Geo/et%.
28
A :* called Geo/et%. will result
Set A)t!on = Create
O$1e)t = Curve
Met#o& = XYZ
Set the C%3e ID list to 1
Set Re+e%en)e Coo%&!n'te F%'/e to Coord 0
Turn off the Ato E2e)te button ,unchec;-
Enter the following into the Ve)to% Coo%&!n'te- list.
-1,0,0.
Enter the following into the O%!"!n Coo%&!n'te- list.
-0,0,0.
,note that !AT:A( will accept either commas or blan;s as separators
between coordinates-
=lic; App(.
#uild the rest of the truss using the following table
/ine (umber Rector =oordinates Crigin =oordinates
2 S2,0,0T S0,0,0T
2 S0,2,0T S0,0,0T
5 S0,2,0T S2,0,0T
6 S2,0,0T S0,2,0T
8 S2,2,0T S0,0,0T
9 S2,0,0T S2,0,0T
" S2,)2,0T S2,2,0T
H S2,0,0T S2,2,0T
S0,2,0T S2,0,0T
(ote that the commands =onstruct, /ine, P1O do (CT wor; based on the coordinates of the 2
end points of the truss member. These commands generate lines based on the origin and the
$ector for that particular truss member.
(ote that if you ma;e a mista;e you can erase by clic;ing on the undo button on the top of the
!AT:A( des;top. This will erase the /AST =C(ST:@=TIC( =C**A(N C(/1. In other
words, it will ta;e the process bac; to before you hit the App(. button the last time.
5. The boundary conditions are specified ne+t.
'rom the T* choose Lo'&/BC5-
A :* called Lo'&/Bon&'%. Con&!t!on- will appear
Set A)t!on = Create
O$1e)t = (is)lace#ent
T.pe = *odal
Set C%%ent Lo'& C'-e = (e"ault
Enter Ne, Set N'/e as le"t"i0
, This is for the clamping of the left most bottom nodes-
=lic; Inpt D't'...
a S* appears
Set Lo'&/BC S)'(e F')to% = 1.
Set T%'n-('t!on- to -0,0,0.
/ea$e the Rot't!on- blan;
#e sure An'(.-!- Coo%&!n'te F%'/e is Coord0
=lic; CG
,bac; in the Lo'&/Bon&'%. Con&!t!on- :*-
=lic; Se(e)t App(!)'t!on Re"!on
29
a S* called Se(e)t App(!)'t!on Re"!on appears with a Select menu on
its left edge.
In the Se(e)t App(!)'t!on Re"!on S*
Turn on the Geo/et%. ,button down-
=lic; in bo+ under Se(e)t Geo/et%!) Ent!t!e-
In the Select *enu ,which is Fust to the left of the S*-
=lic; on the picture with a point
In the main $iew port, clic; on point 2 ,left most point on the bottom
edge-
A Selection =hoices menu will appear. =hoose !oint 2.
, This will cause the words &!oint 2& to appear in the Se(e)t Geo/et%!)
Ent!t!e- bo+ in the :*-
=lic; on A&& Fust below this bo+
, This will remo$e the words &!oint 2& from the Se(e)t Geo/et%!)
Ent!t!e- bo+ and add them to the App(!)'t!on Re"!on bo+-
=lic; CG
,The Lo'& / Bon&'%. Con&!t!on :* appears again-
=lic; App(.
,5 displacement constraint arrows and the numbers 2,2,5 should now appear in
the main $iewport window on the e+treme right point on the bottom of the truss-
#ac; in the :* called Lo'&/Bon&'%. Con&!t!on-
Set A)t!on = Create
O$1e)t = (is)lace#ent
T.pe = *odal
Set C%%ent Lo'& C'-e = (e"ault
Enter Ne, Set N'/e as ri:ht"i0
, This is for the clamping of the right most bottom nodes-
=lic; Inpt D't'...
a S* appears
Set Lo'&/BC S)'(e F')to% = 1.
Set T%'n-('t!on- to - ,0,0.
(ote the space left in before the first comma in the T%'n-('t!on-
$ector. This ensures that the P direction is (CT constrained
/ea$e the Rot't!on- blan;
#e sure An'(.-!- Coo%&!n'te F%'/e is Coord0
=lic; CG
,bac; in the Lo'&/Bon&'%. Con&!t!on- :*-
=lic; Se(e)t App(!)'t!on Re"!on
a S* called Se(e)t App(!)'t!on Re"!on appears with a Select menu on
its left edge.
In the Se(e)t App(!)'t!on Re"!on S*
Turn on the Geo/et%. ,button down-
=lic; in bo+ under Se(e)t Geo/et%!) Ent!t!e-
In the Select *enu ,which is Fust to the left of the S*-
=lic; on the picture with a point
In the main $iew port, clic; on point 8 ,right most point on the bottom
edge-
A Selection =hoices menu will appear. =hoose !oint 8.
, This will cause the words &!oint 8& to appear in the Se(e)t Geo/et%!)
Ent!t!e- bo+ in the :*-
=lic; on A&& Fust below this bo+
2"
, This will remo$e the words &!oint 8& from the Se(e)t Geo/et%!)
Ent!t!e- bo+ and add them to the App(!)'t!on Re"!on bo+-
=lic; CG
,The Lo'& / Bon&'%. Con&!t!on :* appears again-
=lic; App(.
,2 displacement constraint arrows and the numbers 2,5 should now appear in the main $iewport
window on the e+treme right point on the bottom of the truss-
6. The loads are specified ne+t.
,=ontinuing on in the Lo'&/BC5- :*-
change A)t!on = Create
O$1e)t = /orce
T.pe = *odal
=hange the Ne, Set N'/e to to)load
=lic; Inpt D't'...
a S* appears
Enter the force $ector -0 , ;100 , 0.
lea$e the moments - . ,i.e. blan;-
=lic; O0
,=ontinuing on in the Lo'&/BC5- :*-
=lic; Se(e)t App(!)'t!on Re"!on
A S* called Se(e)t App(!)'t!on Re"!on appears with a select menu Fust to its
left
In the Se(e)t App(!)'t!on Re"!on menu
Select the Geo/et%. F!(te% = <eo#etr5
=lic; in the Se(e)t Geo/et%. Ent!t!e- $o2
In the select menu to the left of the S*
=lic; on the point icon
In the main $iewport, clic; on the point 6 ,top center point-
,point 6 will be added to the Se(e)t Geo/et%!) Ent!t!e- list-
In the Se(e)t App(!)'t!on Re"!on menu
=lic; A&&
,!oint 6 will be added to the App(!)'t!on Re"!on list-
=lic; O0
,Lo'&/BC5- menu now reappears-
=lic; App(.
,A $ector with the load of magnitude 200 in the U1 direction will appear on point 6
in the main $iewport-
8. The finite element mesh is specified ne+t.
'rom the T* choose E(e/ent-
A :* appears called F!n!te E(e/ent-
Set A)t!on = Create
O$1e)t = Mesh eed
T.pe = !ni"or#
Select N/$e% o+ E(e/ent- ,button down-
N/$e% = 2
Turn off the Ato E2e)te ,button up-
=lic; in C%3e- L!-t bo+
=lic; on cur$e 2 in the main $iewport
,cur$e 2 is the line between point 2 and point 2. This is the bottom left part of the truss-
,The words &=ur$e 2& will be added to the C%3e L!-t-
2H
=lic; App(.
,circles which represent finite element nodes will appear on points 2 and 2-
No the same for cur$es 2).
,The nodes created abo$e must now be tied together with elements-
,up at the top of the :*-
Set A)t!on = Create
O$1e)t = Mesh
T.pe = Curve
=lic; on #ar2 under E(e/ent Topo(o".
=lic; C%3e L!-t #o+
=lic; cur$e 2 in the main $iewport
=lic; App(.
No the same for cur$es 2)
To see the element numbers on the truss, clic; the K/abel =ontrolL button ,/oo;s li;e an
K/L- on the top row menu. This adds a label control tool bar which allows you to turn
on>off labels for different geometric and>or finite element entities.
,up at the top of the :*-
Set A)t!on = $%uivalence
O$1e)t = &ll
T.pe = 'olerance Cube
,The purpose here is to tie the nodes together that lie on top of one another-
/ea$e the No&e- to $e E2)(&e& list blan;
Set the E4!3'(en)!n" To(e%'n)e to .001
=lic; App(.
,The command window at the bottom of the !AT:A( des;top will tell you that 22 nodes
were deleted-
9. The materials are specified ne+t.
Cn the T* select M'te%!'(-
a :* will appear called M'te%!'(-
Set A)t!on = Create
O$1e)t = Isotro)ic
Met#o& = Manual In)ut
=lic; M'te%!'( N'/e bo+
Input the name to be &lu#inu#
=lic; Inpt P%ope%t!e- bo+
S* called Inpt Opt!on- appears
Input E('-t!) Mo&(- = 4.0$10
Input Po!--on = 0.1
O0
#ac; in the *aterials :*
=lic; App(.
,The E2!-t!n" M'te%!'(- bo+ should ha$e &lu#inu# in it-
". The properties for each element are assigned ne+t.
Cn the T* select E(e/ent P%ope%t!e-
2
a :* will appear called E(e/ent P%ope%t!e-
Set A)t!on = Create
D!/en-!on = 1d
T.pe = +od
=lic; P%ope%t. Set N'/e bo+
Enter truss1
=lic; Inpt P%ope%t!e-
a S* appears called Inpt P%ope%t!e-
=lic; in the M'te%!'( N'/e bo+
=lic; on the word &Aluminum& in the M'te%!'(- P%ope%t. Set bo+
, the words m.Aluminum will appear in the M'te%!'( N'/e bo+-
=lic; in the A%e' bo+
Enter .000= ,recall that the member0s cross section was 5cm + 5cm
s%uare-
=lic; O0
,#ac; in the E(e/ent P%ope%t!e- :*-
=lic; Se(e)t Me/$e%- bo+
In the select menu Fust to the left of the S*
=lic; in the bo+ which contains finite element with 2 end nodes
,This allows you to pic; finite element entities as opposed to
the geometric entities in the other bo+-
*o$e the cursor arrow to a point to the left and abo$e the highest, left)
most point on the truss. =lic; and hold down the left mouse button. Nrag
the cursor ,while holding down the mouse button- to a point to
the right of and below the right)most bottom node. A &selection bo+&
is formed while you drag. :elease the button.
,The words Elm 2. will appear in the Se(e)t Me/$e%- bo+-
=lic; A&&
,The words Element 2. appears in the App(!)'t!on Re"!on bo+-
=lic; App(. in the E(e/ent P%ope%t!e- menu
,truss2 will be added to the E2!-t!n" P%ope%t. Set- bo+-
H. The analysis is to be done is specified ne+t.
Cn the T* select An'(.-!-
a :* will appear called An'(.-!-
Set A)t!on = &nal56e
O$1e)t = $ntire Model
Met#o& = /ull +un
=lic; on So(t!on T.pe
a S* will appear
=lic; on T%'n-('t!on P'%'/ete%-
A S* called T%'n-('t!on P'%'/ete%- will appear
Set D't' Otpt to 7P2 and Print
=lic; O0
#ac; in the An'(.-!- :*
Set So(t!on T.pe = St't!) ,button down-
=lic; O0
,bac; in the :* An'(.-!--
=lic; App(.
,The analysis will ta;e a few seconds to run-
20
(ow we0ll read the results into the graphics database
,bac; in the :* An'(.-!--
Set A)t!on = +ead 7ut)ut2
O$1e)t = +esult $ntities
Met#o& = 'ranslate
=lic; on Se(e)t Re-(t- F!(e
=hoose truss.o)2 ,you may need to go to the root or home directory to find this.
If this file does not e+ist, then there was an error in your model. Ao to the file
truss.log or truss.f09 to attempt to find out what error occurred.-
#ac; in the An'(.-!- :*
=lic; App(.
. RisualiBe the results
'rom the T* choose Re-(t-
A :* called Re-(t- appears
Set Action = =reate
CbFect = ?uic; !lot
@nder Se(e)t F%!n"e Re-(t =hoose (is)lace#ents, 'ranslational
et 8'nt!t. > Y Co#)onent
@nder Se(e)t De+o%/'t!on Re-(t-, choose (is)lace#ents, 'ranslational
=lic; App(.
, A deformed plot appears with colors indicating the le$el of deformation. (ote that the
$isual deformation of the truss is magnified so that you can see the deformation KmodeL.
The actual truss deformations are $ery small< as can be seen by the numerical $alues,
which are (CT scaled-
(ote that you can also $iew the stress results in this manner. Simply choose St%e--<
Ten-o% from the Se(e)t F%!n"e Re-(t options. :ecall that there are a number of ways
to compute and e+trapolate the stresses for a bar and these will ma;e significant
differences in the $alues which are plotted.
20. =hec; the written report of the truss results.
The file containing the written results from the analysis is scaled truss.f09. Cpen the file ,by
simply double clic;ing on it-. The file might be in the root or home directory or in the directory
from which you ran the analysis.
In this file find the displacement $ectors and record the numerical $alues. These will help you
answer some of the %uestion below. Also, find the $ectors for the stresses and constraint forces
and record these $alues.
(e+t you will end your *S= !AT:A( session by sa$ing your database and e+iting
Cn the T* select F!(e
'rom the pull down menu select S'3e
Cn the T* select F!(e
'rom the pull down menu select 8!t
VII. 8UESTIONS FROM THE TUTORIAL: MODELING A TRUSS
The %uestions below refer to the truss model described at the beginning of this tutorial. Also,
information from the output file truss.f09 will be needed in order to answer many of these
22
%uestions. As used below, the term &/e/$e%& refers to the portion of a truss structure between
two Foints. 'or e+ample, the top of this structure has two horiBontal members which are
connected by the Foint at which the load is applied.
1'. 3hat is the ma+imum displacement for the structure 4
1$. Is this displacement consistent in location, magnitude and direction with your physical
intuition 4
6'. 3hat is the ma+imum stress in the structure 4
6$. Is this stress consistent in location, magnitude and direction with your physical intuition 4
=. Are there any members with $ery low stresses4 Noes this ma;e physical sense4
>. 7ow many e%uations are sol$ed in order to determine the displacements for this structure 4
?. 3hat assumptions are in$ol$ed in using this specific element as opposed to using a 2 node
beam element with 9 degrees of freedom ,5 displacements and 5 rotations- per node 4
@. The present model uses a single 2)node bar element for each truss member. 3ould the
accuracy of the model increase if two bar elements were used to model each truss member 4
Vustify your answer.
A'. The resultant forces ,sometimes called constraint, restoring or reaction forces-, are located
at the nodes where the boundary conditions are applied. State how these resultant forces can be
used as a &necessary but not sufficient& test of the accuracy of your analysis.
A$. Noes your analysis pass this test 4
B. If two nodes in your final truss structure ha$e the e+act same physical location but different
node numbers, what part of the !AT:A( analysis procedure has been left out 4
C'. 7ow could the element properties be changed to model this truss if the members in the
structure were circular hollow aluminum bars. Assume that the outside diameter is 5 cm and the
inside diameter is 2 cm. :emember that this structure only models the membrane ,a+ial-
deformation not the bending deformation of each member.
C$. If you wanted to account for bending deformation in your model, could you use this same
adFustment to the physical properties to model the truss with hollow members 4
1;. Assume that the cross sectional area of the truss members is incorrectly input in s%uare cm
as opposed to s%uare meters. If the other data for the problem is input using meters, what would
the ma+imum deflection of the truss be 4
11. Assuming that the rotations of the cross sections of the bars are small, what will be the
difference between the results of your !AT:A( analysis and the e+act analysis 4 ,&e+act& here
refers to the analytic analysis using standard structural analysis methods-
16'. Some truss structures may be designed so that, if certain members of the truss are
damaged to the e+tent that they no longer ha$e significant stiffness, the structure will still be able
to handle reasonable loading. This type of truss assembly is said to ha$e redundant members.
3ithout changing the number of elements in the structure, suggest a method of using *S=
!AT:A( to determine if there are redundant members in this truss structure.
16$. @se the method de$eloped in 25a- to determine if one of the diagonal members is
redundant.
16). @se the method de$eloped in 25a- to determine if one of the $ertical members is
redundant.
1='. !redict the deflection if the direction of the load is changed from the negati$e 1 direction, to
the O direction ,note from your nodal location information that this truss is located in the P ) 1
plane-.
1=$. :un the analysis and e+plain the displacement results.
1>'. !redict the effect of remo$ing the displacement boundary condition on the lower right node
of the truss structure 4
1>$. :un the analysis and e+plain the displacement results.
22
MSC/PATRAN TUTORIAL # =
MODELING A CANTILEVERED BEAM DITH END LOAD
USING > NODE SHELL ELEMENTS
I. THE PHYSICAL PROBLEM
The beam below is cantile$ered or &built in& on the left edge. This means that both the
translations and the rotations are held to Bero along this edge. A point or concentrated load of
magnitude 2000 ( ,appro+imately 228 lb- in the negati$e 1 direction is found at the tip of the
beam. This problem is part of a standard set of test cases for finite elements published in a paper
by *ac(eal and 7arder ,*ac(eal founded the company that ma;es the 'EA code
*S=>(AST:A( and *S=>!AT:A(-. The set of problems is called &The *ac(eal ) 7arder Test
=ases&. The material properties for the beam are E= 200 + 20
.
'or the displacement at the tip of the beam ,ma+imum displacement-.
EI
PL
L x
Y
,
% (
,
Some basic %uestions to consider before creating the computational model are.
a- 3here will the stresses be tensile and where will they be compressi$e4
b- 3hat will be the magnitude and direction of the reaction forces>moments4
c- 3here will the stresses be Bero4
d- 7ow do the displacements $ary along the length ,linear, %uadratic etc.-4
e- 3hat will the local effect of the concentrated load be on the stresses4
f- Is the model fully constrained from rigid body rotations and displacements4
Answering these %uestions %ualitati$ely, along with the %uantitati$e analytical solutions for the
ma+ stress and displacement will pro$ide reinforcement that your computational model is
correctly constructed.
III. GEOMETRIC AND FINITE ELEMENT MODEL
As is the standard procedure for building *S=>!atran models, we will build the geometry first and
then construct a finite element mesh on that geometry. The geometry will proceed from creation
of points to lines to surfaces for this simple model. (e+t, we will use 6 node shell elements
deforming in their membrane mode to model the beam. In this e+ercise, we will $ary the e+act
number and configuration of these elements. This is discussed in detail in the ne+t paragraph.
(e+t, the material and element properties will be entered. 3e will constrain the 5 displacement
and 5 rotational degrees of freedom on the left edge ,for both nodes-. This creates the
cantile$ered or built)in, end condition. Then we will, place a point load of magnitude 2000 on the
26
top right node of the tip ,or right)most- element. This load will be in the negati$e 1 direction.
'inally, the nodes must be e%ui$alenced before the analysis is ready to run.
#elow, we show 8 mesh configurations for the beam ,labeled KaL through KeL-. =omparison of
results between mesh KaL and mesh KbL will indicate of how the number of elements affects the
model0s ability to correctly model a beam problem. Increasing the number of elements in a mesh
in order to increase the accuracy of the results is called KhL con$ergence. *eshes KbL U KeL all
ha$e 9 elements< but the elements ha$e different orientations. Elements that ha$e non)regular
shapes are said to be distorted. Nistorted elements can cause errors in the 'EA results. This
can be a significant problem in comple+ meshes as e$en the best automatic mesh generators
often produce some distorted elements. The elements in *S=>(astran ha$e been specifically
designed to minimiBe this unfortunate effect, but some sensiti$ity to element distortion may still
remain. Nifferent types of element distortion result in different le$els of error. E$aluating results
from the meshes KbL ) KeL will pro$ide you with some feel for how these elements perform when
they are distorted.
Me-#e- +o% t#e E#F Con3e%"en)e G D!-to%t!on An'(.-!-
I. Re)t'n"('% 6 E(e/ent Me-#:
1000
6.0
N
m
20 cm
$7 Re)t'n"('% @ E(e/ent Me-#:
1000
6.0
N
m
20 cm
28
)7 1; De"%ee P'%'((e(o"%'/ Me-#:
1000
6.0
N
m
20 cm
10
Degrees
Typical
&7 >? De"%ee P'%'((e(o"%'/ Me-#:
1000
6.0
N
m
20 cm
45 Deg.
Typ.
e7 >? De"%ee T%'peHo!& Me-#:
1000
6.0
N
m
20 cm
45 Deg.
Typ.
29
IV. FINITE ELEMENT THEORY
The exact details of the formulation of the 4 node shell elements in MSC/Nastran is rather complicated.
However, the basic formulation of an isoparametric 4 node membrane element is not extremely difficult
and will provide us with sufficient background information to begin to understand the h convergence and
distortion sensitivity studies. This basic form is constructed as follows:
Isoparametric Formulation of a 2-D Membrane Element [K] Matrix
Assume the element has the configuration shown below:
4.0
2.0
Y
X
1 2
3
4
The physical and natural coordinate locations of the 6 nodes are.
Cur goal is to find the element stiffness matri+
ASS@*E.
V
T
dV B E B K - . - . - . - .
ASS@*E. 2 displacement degrees of freedom ,dof- per node
3ith . I#J = the strain ) displacement matri+ such that [ ]{ } { } B u
where. WuX is the dof $ector and W
'
where I(J is the shape function matri+
1
]
1
1 , " )
1 , " )
0 0 0 0
0 0 0 0
- .
N N N N
N N N N
N
and the rules for the shape functions are . 2-
N
i
must be =2 at node &i&
2-
N
i
must be =0 at any node not = &i&
This leads to the shape functions. N
)
)
1
) ) ( %( % < N
2
1
4
1 1 + ( )( ) <
N
3
1
4
1 1 + + ( )( ) < N
4
1
4
1 1 + ( )( )
Step 6: 'ind the I#J matri+.
:ele$ant strains are
'
'
1
1
1
1
1
]
1
'
v
u
D
v
u
x y
y
x
xy
yy
xx
- . 0
0
/ 0
'
So
0 / . -. -0 / . -0 / D N u B u
with [ ] [ ][ ] B D N
Therefore,
1
1
1
]
1
x y x y x y x y
y y y y
x x x x
N N N N N N N N
N N N N
N N N N
B
, 1 , 1 , , , , , " , " , ) , )
, 1 , , , " , )
, 1 , , , " , )
0 0 0 0
0 0 0 0
- . where the commas denote
partial differentiation.
Step =: @se the Vacobian to find deri$ati$es.
Isoparametric Assumption.
T
y x y x y x y x N
y
x
/ , , , , -0 .
1 1 , , , , " " , ) )
'
i.e. the isoparametric assumption is that geometry can be interpolated using the same
interpolation functions as the displacements.
The Vacobian matri+
1
1
1
1
]
1
1
]
1
1
1
1
]
1
1 1
, ,
" "
) )
, 1 , , , " , )
, 1 , , , " , )
- .
y x
y x
y x
y x
N N N N
N N N N
y
x
y
x
J
2H
and from chain rule
'
'
1
]
1
'
,
,
)
,
,
, ,
, ,
,
,
- .
i
i
i
i
y y
x x
y i
x i
N
N
J
N
N
N
N
So in this particular case.
1
1
1
1
]
1
1
]
1
+ +
+ +
" 0
" 1
0 1
0 0
) ) ) )
) ) ) )
1
)
- .
J =
1
]
1
1
]
1
) 0
0 "
1 0
0 2
1
)
which implies that
1
]
1
) 0
0
- .
"
)
)
J
This allows us to find the entries in I#J
Step >: !erform the numerical integration.
Assume that the element has constant thic;ness = t implies
A
T
dy dx B E B t K - . - . - . - .
3hich, according to the rules of calculus can be written.
d d J B E B t K
T
- . - . - . - .
where J is the determinant of the Vacobian matri+.
Aaussian numerical integration is then used to find the final numbers for the element stiffness.
This ta;es the form.
. - . - . - . -
( , %
K h B E B J w w
j
nj
i
ni
T
i j
n
i j
) )
3here ngF and ngi are the number of gaussian integration points in the KFL and KiL directions
respecti$ely and wF and wi are the associated gaussian weighting factors.
Un&e%-t'n&!n" t#e E#F Con3e%"en)e E2pe%!/ent:
'rom step 2 abo$e we gain insight into the KhL con$ergence study. :emember that the that the
analytic formula for the displacements as a function P ,distance from built)in edge- is.
EI
x L Px
x
y
3
% , (
% (
"
where ! is the load, / is the length, E is the Elastic *odulus and I is the
bending moment in inertia. This e%uation shows that the displacement is a cubic function of the
distance from the cantile$er. As the bi)linear ,linear in both
and
V
T
dV B E B K - . - . - . - .
ASS@*E. 2 displacement degrees of freedom ,dof- per node
3ith . I#J = the strain ) displacement matri+ such that [ ]{ } { } B u
where. WuX is the dof $ector and W
'
where I(J is the shape function matri+
1
]
1
1 , " )
1 , " )
0 0 0 0
0 0 0 0
- .
N N N N
N N N N
N
and the rules for the shape functions are . 2-
N
i
must be =2 at node &i&
2-
N
i
must be =0 at any node not = &i&
(CNE ,+,y-
( , )
2 ,0,0- ,)2,)2-
2 ,6,0- ,2,)2-
5 ,6,2- ,2,2-
6 ,0,2- ,)2,2-
8
This leads to the shape functions. N
)
)
1
) ) ( %( % < N
2
1
4
1 1 + ( )( ) <
N
3
1
4
1 1 + + ( )( ) < N
4
1
4
1 1 + ( )( )
Step 6: 'ind the I#J matri+.
:ele$ant strains are
'
'
1
1
1
1
1
]
1
'
v
u
D
v
u
x y
y
x
xy
yy
xx
- . 0
0
/ 0
'
So
0 / . -. -0 / . -0 / D N u B u
with [ ] [ ][ ] B D N
Therefore,
1
1
1
]
1
x y x y x y x y
y y y y
x x x x
N N N N N N N N
N N N N
N N N N
B
, 1 , 1 , , , , , " , " , ) , )
, 1 , , , " , )
, 1 , , , " , )
0 0 0 0
0 0 0 0
- . where the commas denote
partial differentiation.
Step =: @se the Vacobian to find deri$ati$es.
Isoparametric Assumption.
T
y x y x y x y x N
y
x
/ , , , , -0 .
1 1 , , , , " " , ) )
'
i.e. the isoparametric assumption is that geometry can be interpolated using the same
interpolation functions as the displacements.
The Vacobian matri+
1
1
1
1
]
1
1
]
1
1
1
1
]
1
1 1
, ,
" "
) )
, 1 , , , " , )
, 1 , , , " , )
- .
y x
y x
y x
y x
N N N N
N N N N
y
x
y
x
J
and from chain rule
'
'
1
]
1
'
,
,
)
,
,
, ,
, ,
,
,
- .
i
i
i
i
y y
x x
y i
x i
N
N
J
N
N
N
N
90
So in this particular case.
1
1
1
1
]
1
1
]
1
+ +
+ +
" 0
" 1
0 1
0 0
) ) ) )
) ) ) )
1
)
- .
J =
1
]
1
1
]
1
) 0
0 "
1 0
0 2
1
)
which implies that
1
]
1
) 0
0
- .
"
)
)
J
This allows us to find the entries in I#J
Step >: !erform the numerical integration.
Assume that the element has constant thic;ness = t implies
A
T
dy dx B E B t K - . - . - . - .
3hich, according to the rules of calculus can be written.
d d J B E B t K
T
- . - . - . - .
where J is the determinant of the Vacobian matri+.
Aaussian numerical integration is then used to find the final numbers for the element stiffness.
This ta;es the form.
. - . - . - . -
( , %
K h B E B J w w
j
nj
i
ni
T
i j
n
i j
) )
3here ngF and ngi are the number of gaussian integration points in the KFL and KiL directions
respecti$ely and wF and wi are the associated gaussian weighting factors.
Un&e%-t'n&!n" t#e Co/pt't!on'( V!$%'t!on An'(.-!- :
The elements as formed above must be assembled into a global stiffness matrix. In the same manner,
element mass matrices are formed using the equation
d d J N N M
T
- . - . - .
. A similar form e+ists for the :ayleigh damping matri+ I=J. The
stiffness, mass and damping matrices are then used in the dynamics e%uilibrium relationship
/ 0 / 0 - . / 0 - . / 0 - . ! d K d " d M + +
where the o$er)dots indicated deri$ati$es with respect to
time and WfX is the forcing function. This set of e%uations can be sol$ed for the time history of the
motion ,transient dynamics- or for the eigen$alues and eigen$ectors. 'or the $ibration analysis,
the damping and the forcing function are assumed to be Bero. The resulting eigen$alue problem
of the second ;ind is .
/ 0 0 / 0 - . / 0 - . + d K M
where eigen$alues are the natural fre%uencies
and the
eigen$ectors WdX gi$e the node shapes.
R. STE! #1 STE! I(ST:@=TIC(S 'C: *CNE/I(A T7E RI#:ATIC( C'
T7E =A(TI/ERE:EN #EA* @SI(A *S=>!AT:A(
!reliminaries for using !AT:A( include.
a- /og on to the computer
b- =lic; STA:T ,lower left corner of the 3indows Nes;top-, go to !rograms, Select *S=
,common-, Select *S= !atran.0.
92
In the instructions below, the following abbre$iations and terms will be used.
TM = Top Men. This refers to the horiBontal menu options residing at the top of the screen after
!AT:A( has been initiated.
RM = R!"#t Men. This refers to the menus that pop up after an option has been chosen from
the top menu. These menus reside on the far right side of the !AT:A( des;top.
SM = S$o%&!n'te Men. This referees to the menus that pop up from options selected in the
right menu.
C(!)* = @nless otherwise stated, this indicates a clic; with the left mouse button.
Bo(&+')e will indicate te+t that occurs in the !AT:A( menus.
Italics te+t will indicate te+t that you must enter into te+t bo+es in the !AT:A( menus or te+t that
you choose in a menu scroll bo+.
2. Cur first step is to create a new database.
'rom the T* choose F!(e
In the resulting pull down menu choose Ne,
A S* called Ne, D't'$'-e pops up
Turn on ,chec;ed- Mo&!+. P%e+e%en)e-
@nder F!(e N'/e enter bea#;vib.db
=lic; O0
2. (e+t set the analysis preference.
A Ne, Mo&e( P%e+e%en)e- window will appear as a :*
@nder To(e%'n)e choose ?ased on Model
Set Mo&e( D!/en-!on to10.0
@nder An'(.-!- Co&e choose MCA*&'+&*
=hoose An'(.-!- T.pe = tructural
clic; O0
5. The geometry of the beam will be determined ne+t.
'rom the T* choose Geo/et%.
A :* called Geo/et%. will result
Set A)t!on = Create
O$1e)t = Point
Met#o& = XYZ
Set the Po!nt ID list to 1
Set Re+e%en)e Coo%&!n'te F%'/e to Coord 0
Turn off the Ato E2e)te button
Enter the following into the Po!nt Coo%&!n'te- list.
[0,0,0]
,note that !AT:A( will accept either commas or blan;s as separators
between coordinates-
=lic; App(.
A point will appear in the main $iewport at coordinates I0,0,0J
@se this same procedure to create points at coordinates I2,0,0J, I2,0.2,0J and I0,0.2,0J
#ac; at the top of the :* called Geo/et%.
Set A)t!on = Create
O$1e)t = Curve
Met#o& = Point
Set the C%3e ID list to 1
92
Turn Atoe2e)te off
Set St'%t!n" Po!nt L!-t = Point 1
Set Ending !oint /ist = Point 2
=lic; App(.
#ac; at the top of the :* called Geo/et%.
Set A)t!on = Create
O$1e)t = Curve
Met#o& = Point
Set the C%3e ID list to 2
Turn Atoe2e)te off
Set St'%t!n" Po!nt L!-t = Point 1
Set Ending !oint /ist = Point 4
=lic; App(.
#ac; at the top of the :* called Geo/et%.
Set A)t!on = Create
O$1e)t = ur"ace
Met#o& = Curve
Set the S%+')e ID list to 1
Set P't%'n 6 Con3ent!on o""
Opt!on = 2 Curve
Set M'n!+o(& o"" ,not chec;ed-
Set St'%t!n" C%3e L!-t = Curve 1
Set Ending =ur$e /ist = Curve 2
=lic; App(.
2. The finite element mesh is specified ne+t.
'rom the T* choose E(e/ent-
A :* appears called E(e/ent-
Set A)t!on = Create
O$1e)t = Mesh
T.pe = ur"ace
Set No&e I& > 1
Set E(e/ent I& L!-t > 1
Set G(o$'( E&"e Len"t# > 0.025
Set E(e/ent Topo(o". > 8uad3
Set Me-#e% > Iso#esh
=lic; in the S%+')e L!-t bo+
=lic; and drag to select the entire structure
The 3ords &Surface 2& should appear in the S%+')e L!-t
=lic; App(.
Set A)t!on = $%uivalence
O$1e)t = &ll
T.pe = 'olerance Cube
,The purpose here is to tie the nodes together that lie on top of one another-
Set the E4!3'(en)!n" To(e%'n)e to .001
=lic; App(.
,The command window at the bottom of the !AT:A( des;top will tell you that 0 nodes were
deleted. This step will become critical if, in more complicated models, you are attempting to Foin
portions of a model which ha$e been meshed separately.-
5. The boundary conditions are specified ne+t.
'rom the T* choose Lo'&/BC5-
95
A :* called Lo'&/Bon&'%. Con&!t!on- will appear
Set A)t!on = Create
O$1e)t = (is)lace#ent
T.pe = *odal
Set C%%ent Lo'& C'-e = (e"ault
Enter Ne, Set N'/e as lBcant
, The name can be whate$er name you wish. The name lBcant is chosen as this
is for the cantile$er of the left most nodes-
=lic; Inpt D't'...
a S* called Input Nata appears
Set Lo'&/BC S)'(e +')to% =1
Set T%'n-('t!on- to -0,0,0.
Set Rot't!on- to -0,0,0.
#e sure An'(.-!- Coo%&!n'te F%'/e is Coord0
=lic; CG
,bac; in the Lo'&/Bon&'%. Con&!t!on- :*-
=lic; Se(e)t App(!)'t!on Re"!on
A SM )'((e& Se(e)t App(!)'t!on Re"!on 'ppe'%-
Turn on the FEM ,button down-
=lic; in bo+ under Se(e)t No&e-
@se the cursor to highlight the set of nodes along the left $ertical edge of the
beam. There should be 8 nodes there.
=lic; CG
,The Lo'& / Bon&'%. Con&!t!on :* appears again-
=lic; App(.
,5 displacement constraint arrows and 5 rotation constraint arrows should now
appear on each node in the main $iewport window on the e+treme left edge of
the beam. (umbers 2,2,5,6,8,9 will appear with the arrows to show that all 9 of
the dof are constrained there-
6. The materials are specified ne+t.
Cn the T* select M'te%!'(-
a :* will appear called M'te%!'(-
Set A)t!on = Create
O$1e)t = Isotro)ic
Met#o& = Manual In)ut
=lic; M'te%!'( N'/e bo+
Input the name to be alu#inu#
=lic; Inpt P%ope%t!e- bo+
S* called Inpt Opt!on- appears
Input E('-t!) Mo&(- =40.0$=
Input Po!--on = 0.1
Input the Den-!t. to be 2400
=lic; O0
#ac; in the M'te%!'(- :*
=lic; App(.
8. The properties for each element are assigned ne+t.
Cn the T* select P%ope%t!e-
96
a :* will appear called E(e/ent P%ope%t!e-
Set A)t!on = Create
D!/en-!on = 2d
T.pe = hell
=lic; P%ope%t. Set N'/e bo+
Enter bea#B)ro)
=lic; Inpt P%ope%t!e-
a S* appears called Inpt P%ope%t!e-
=lic; in the M'te%!'( N'/e bo+
=lic; on the word &aluminum& in the M'te%!'( P%ope%t. Set- bo+ at the
bottom of the S*
, the words m.aluminum will appear in the M'te%!'( N'/e bo+ at the top
of the S*-
=lic; in the T#!)*ne-- bo+
Enter 0.01
=lic; O0
,#ac; in the E(e/ent P%ope%t!e- :*-
=lic; Se(e)t Me/$e%- bo+
a !atran Se(e)t /en will appear on the left edge of the :*
=lic; on the icon which contains the surface or face icon
*o$e the cursor arrow to a point to the left and abo$e the highest, left)
most point on the beam. =lic; and hold down the left mouse button. Nrag
the cursor ,while holding down the mouse button- to a point to the right of
and below the right)most bottom node. A &selection bo+& is formed while
you drag. :elease the button.
,The words Surface 2 will appear in the Se(e)t Me/$e%- bo+-
=lic; A&&
,The words Surface 2 appears in the App(!)'t!on Re"!on bo+-
=lic; App(. in the E(e/ent P%ope%t!e- menu
,beamEprop will be added to the E2!-t!n" P%ope%t. Set- bo+-
9. The analysis is to be done is specified ne+t.
Cn the T* select An'(.-!-
a :* will appear called An'(.-!-
Set A)t!on = &nal56e
O$1e)t = $ntire Model
Met#o& = /ull +un
=lic; Translation !arameters
In the S* that appears, set D't' Otpt = 7)2 and Print
=lic; O0
#ac; in the :* An'(.-!-
Set So(t!on T.pe = *or#al Modes ,button down-
=lic; O0
=lic; App(.
,The analysis will ta;e a few seconds to run. A S* indicating that *S=>(astran is
wor;ing may appear-
". A graphical representation of the mode shapes can be produced.
A graphical representation of the mode shapes pro$ides an easy way to begin to determine if you
ha$e constructed your model correctly.
Cn the T* select An'(.-!-
98
Set A)t!on = +ead 7ut)ut2
O$1e)t = +esults $ntities
Met#o& = 'ranslate
=lic; Se(e)t Re-(t- F!(e
A S* appears called Se(e)t F!(e
=lic; the file $e'/K3!$.op6
J1ou may need to loo; in your home or root directory to find the file. If
this file does not e+ist, then you ha$e made a mista;e in constructing
your model. Ao to E+plorer ,right)clic; on Start and choose E+plore- and
find the file beam)$ib.log and beam.f09. Cpen these files by double
clic;ing on them and search for the word KerrorL to determine what your
mista;e is-.
?ea#;vib.o)2 then appears in the F!(e N'/e bo+
=lic; O0
,bac; in the An'(.-!- menu-
=lic; App(.
Cn the T* select Re-(t-
A :* will appear called Re-(t-
Set A)t!on = Create
O$1e)t = 8uic9 Plot
In t#e Select :esult =ase $o2 )(!)* De+'(t< Mode 1
In t#e Select 'ringe :esult $o2 )(!)* Eigenvectors, translational
In t#e Apply 'ringe :esult $o2 )(!)* Eigenvectors, translational
Set ?uantity = Magnitude
Turn on the animation button ,so it displays a chec;-
=lic; App(.
(This will create the animation of the first mode)
Investigate other, higher order mode shapes. Be sure to record data and screen captures needed to
answer the questions below.
H. (e+t you will end your *S=>!AT:A( session by sa$ing your database and e+iting.
Cn the T* select F!(e
'rom the pull down menu select S'3e
Cn the T* select F!(e
'rom the pull down menu select 8!t
VI. E9ERCISES:
a- =ompare the 'EA results with the analytic results for the first 5 pairs of mode shapes and
fre%uencies which are associated with bending of the beam in the direction of minimum KIL.
1ou can use the analytic e%uations shown earlier to produce the analytic results.
b- Study the first 8 mode shapes produced by the (astran and comment on which modes are
not associated with bending about the minimum KIL direction.
99
c- :erun the analysis using only .00928 as the global edge length ,produces 6 times as many
elements-. Noes a refinement in the mesh appear to produce more closely con$erged
results4
d- =hange the !oisson0s ratio to 0.0. :erun the analysis using the original global edge length of
0.028. =ompare these errors with those found while using a !oisson0s ratio of 0.05, !ropose
an e+planation for the differences.
e- Identify the possible sources of that might ma;e our results a poor model of the actual
physical structure.
MSC/PATRAN TUTORIAL # A
THERMAL ANALYSIS OF A COOLING FIN USING SHELL ELEMENTS
I. THE PHYSICAL PROBLEM
The problem you will model is a fin of aluminum alloy, 0.2 m long, 0.002 m thic; and large width.
This is the type of fin that might pro$ide air)cooling on a motorcycle engine. 'or the finite element
model, we consider a representati$e strip of the fin 0.02 m in depth ,shown as the region between
the dotted lines in the drawing-. The 200)degree wall is representati$e of the hot temperature of
the engine. Cur goal is to find the temperature distribution down the fin. If the outside tip of the
engine is too hot, it can be a safety concern. 7eat is conducted down the fin ,away from the heat
source of the engine- and heat is also lost through con$ection from the top and bottom surfaces
to the air. The ambient temperature of the air is ;nown to be 28 =
o
and the con$ection coefficient
,film coefficient- is ;nown to be 50 ,3>m
2
- . The fin itself is made of aluminum which has a
conducti$ity of 2"" ,3>m
2
G-.
9"
0.20
3all
200 =
0.02
II. THINKING ABOUT THE MECHANICS
The analytic solution for the temperatures for this problem is readily a$ailable. Any 7eat Transfer
te+t will pro$ide e%uations for the temperature distribution of a fin considering conduction away
from the heat source and con$ection from the top and bottom surfaces. These results can be
used to gi$e basic analytic comparison solutions for certain sections of the structure. (ote that
we assume no radiation occurs and that only the top and bottom surfaces ha$e significant
con$ection heat transfer ,the con$ection from the edges of the fin is neglected-. These
assumptions are normal for a first le$el analysis where the temperatures are in the ranges used in
this problem.
III. GEOMETRIC AND FINITE ELEMENT MODEL
As is the standard procedure for building *S=>!atran models, we will build the geometry first and
then construct a finite element mesh on that geometry. The geometry will proceed from creation
of cur$es to a surface for this simple model. (e+t, we will use 6 node 2)dimensional elements to
model the fin. (e+t, the material and element properties will be entered. 3e will set the wall
temperature and the con$ection characteristics for the top and bottom of the fin. 'inally, the
nodes must be e%ui$alenced before the analysis is ready to run.
IV. FINITE ELEMENT THEORY
The e&act details of the formulation of the 1 node "5d elements in 67'+astran is rather complicated.
8owever, the basic formulation of the "5d thermal element is not e&tremely difficult and will provide us
with sufficient bac!ground information to begin to understand the general application areas and
convergence of these elements. This basic formulation for the "5d thermal, linear, quasistatic element can
be found in most any 9inite :lement Analysis te&t (see for e&le 9inite :lements for 7tress Analysis, by
R.;. 'oo!, <ohn =iley > 7ons, )**4.% .
R. I(ST:@=TIC(S 'C: *CNE/I(A T7E 'I( @SI(A *S=>!AT:A( \
*S=>(AST:A(
!reliminaries for using !AT:A( include.
a- /og on to the computer
b- =lic; STA:T ,lower left corner of the 3indows Nes;top-, go to !rograms, Select *S=
,common-, Select *S= !atran.0.
The instructions below give details for modeling the thermal fin problem discussed above. The instructions
are ?T as detailed as have been given in other problems as it is e&pected that you have begun to get a feel
for how to do certain tas!s in Patran.
In the instructions below, the following abbre$iations and terms will be used.
TM = Top Men. This refers to the horiBontal menu options residing at the top of the screen after
!AT:A( has been initiated.
9H
0.002
RM = R!"#t Men. This refers to the menus that pop up after an option has been chosen from
the top menu. These menus reside on the far right side of the !AT:A( des;top.
SM = S$o%&!n'te Men. This referees to the menus that pop up from options selected in the
right menu.
C(!)* = @nless otherwise stated, this indicates a clic; with the left mouse button.
Bo(&+')e will indicate te+t that occurs in the !AT:A( menus.
Italics te+t will indicate te+t that you must enter into te+t bo+es in the !AT:A( menus or te+t that
you choose in a menu scroll bo+.
2. Cur first step is to create a new database.
'rom the T* choose F!(e
In the resulting pull down menu choose Ne,
A S* called Ne, D't'$'-e pops up
Turn on ,chec;ed- Mo&!+. P%e+e%en)e-
@nder F!(e N'/e enter "in.db
=lic; O0
2. (e+t set the analysis preference.
A Ne, Mo&e( P%e+e%en)e- window will appear as a :*
@nder To(e%'n)e choose ?ased on Model
Set Mo&e( D!/en-!on to 0.2
@nder An'(.-!- Co&e choose MCA*&'+&*
=hoose An'(.-!- T.pe = 'her#al
clic; O0
5. The geometry of the beam will be determined ne+t.
Select Geo/et%. from T*.
Cn :*, select A)t!on= =reate, O$1e)t= =ur$e, Met#o&= P1O
(ote C%3e ID L!-t has a 2.
Re+e%. Coo%&!n'te F%'/e should be =oord 0
Set Ve)to% Coo%&!n'te- L!-t to 0.2 0 0 ,1ou will be drawing lines ,$ectors- with these
+yB components.-
O%!"!n Coo%&!n'te- L!-t = 0 0 0 =lic; APPLY. ,A line from origin to point 0.2,0,0
should appear on screen.-
*a;e second cur$e. 3ith same $ector, set O%!"!n Coo%&!n'te- L!-t to 0 0.02 0.
=lic; App(.. ,A second cur$e appears on the screen.-
(ow create a surface between the cur$es.
Cn the Geo/et%. :*, choose A)t!on= =reate< O$1e)t= Surface< Met#o&= =ur$e. Set
Opt!on to 2 =ur$e.
(ote there is a St'%t!n" C%3e L!-t and En&!n" C%3e L!-t.
=lic; in the St'%t!n" C%3e L!-t bo+. .Select the first cur$e by using the mouse. =lic; the
small bo+ on cur$e 2 on the screen. =lic; in the En&!n" C%3e L!-t bo+. Then clic; on
cur$e 2. (ote a surface is created.
6. =reate the finite elements.
Cn the T* select E(e/ent- and get a :*.
=hoose A)t!on= =reate< O$1e)t=*esh< T.pe= Surface
=hoose the siBe of the elements. Type in G(o$'( E&"e Len"t#. 0.02
Select I-o/e-#. =lic; in S%+')e L!-t bo+. Select the Surface 2 with the cursor.
=lic; App(.. (ote the model has 20 elements.
9
8. =reate #oundary conditions
At T*, select Lo'&/BCN-. Aet :*.
(ow create the con$ection characteristics for the bottom of the fin
=hose A)t!on=create, O$1e)t=con$ection, t.pe=element uniform
(ame the con$ection #=. In Ne, Set N'/e, type top)convec
Select T'%"et E(e/ent T.pe = 2N
=lic; Inpt D't'. Aet submenu. Type 10 for con$ection coefficient ,w>m
2
c- for top
surface con$ection. Type 25 for ambient temp. =lic; O0.
#ac; in Lo'&/ BC menu, clic; Se(e)t App(!)'t!on Re"!on bo+
Select FEM as the Aeometry 'ilter.
=lic; in Se(e)t 6D E(e/ent- o% E&"e- bo+
@sing mouse, clic; on all the elements. ,7old shift down for
multiple selections.- =lic; A&&. The application region
bo+ should list the elements 2.20. =lic; O0
#ac; in Lo'&/BC menu clic; App(..
(ow create the con$ection characteristics for the bottom of the fin
=hose A)t!on=create, O$1e)t=con$ection, t.pe=element uniform
(ame the con$ection #=. In Ne, Set N'/e, type bot)convec
Select T'%"et E(e/ent T.pe = 2N
=lic; Inpt D't'. Aet submenu. Type 10 for con$ection coefficient ,w>m
2
c- for bottom
surface con$ection. Type 25 for ambient temp. =lic; O0.
#ac; in Lo'&/ BC menu, clic; Se(e)t App(!)'t!on Re"!on bo+
Select FEM as the Aeometry 'ilter.
=lic; in Se(e)t 6D E(e/ent- o% E&"e- bo+
@sing mouse, clic; on all the elements. ,7old shift down for
multiple selections.- =lic; A&&. The application region
bo+ should list the elements 2.20. =lic; O0
#ac; in Lo'&/BC menu clic; App(..
(ow create the base temperature #=. In Lo'&/BC :*
A)t!on=create, O$1e)t=Temp, t.pe=nodal
In Ne, Set N'/e type ?asete#). =lic; on Inpt D't'.
In submenu Inpt D't', type 200 in Te/pe%'t%e bo+. =lic; O0.
#ac; in Lo'&/BC S*, clic; on Se(e)t App(!)'t!on Re"!on.
In submenu, select FEM as Geo/et%. F!(te%.
=lic; on Se(e)t No&e-.
@sing mouse, select the nodes 2 and 22 at the e+treme left of the model. =lic; A&&.
=lic; O0.
#ac; in Lo'&/BC menu, clic; App(.. ,The screen should show 200 at nodes 2 and 22.-
9. =reate and select material
Cn T* select M'te%!'(-.
In submenu, A)t!on=create, O$1e)t=isotropic, /et#o&=manual input.
In M'te%!'( N'/e bo+, type alu#inu#. =lic; Inpt P%ope%t!e-.
In submenu, Inpt Opt!on-, enter thermal conducti$ity as 144. ,w>m
2
;-.
=lic; O0. If S* does not disappear, =lic; C'n)e(.
#ac; in the :* , =lic; App(.
In T*, select P%ope%t!e-
In submenu A)t!on=create, O$1e)t=2N, T.pe=shell.
In P%ope%t. Set N'/e type shellB)ro).
=lic; on Inpt P%ope%t!e- In submenu, clic; on Aluminum in the *aterial !roperty Sets
bo+.
*. Aluminum appears in the *aterial (ame bo+ at the top of the form.
"0
Set T#!)*ne--= 0.002 =lic; O0.
#ac; in E(e/ent P%ope%t!e-, clic; Se(e)t /e/$e%-.
@se mouse to select the entire model. ,1ou can clic; and draw a bo+ around the entire
model to select it.- =lic; A&&. =lic; App(..
". /oad #oundary =onditions
In order to ha$e both the con$ection on top and on the bottom as well as the and base
temperature
#= on the model, all 5 boundary conditions must be combined into a single load
case.
In T*, select Lo'& C'-e-.
In S*, A)t!on==reate, Lo'& )'-e n'/e, type "inBcase.
In De-)%!pt!on, type. h>10 on "in Dith base> 200C.
and a#bient>25C.
=lic; on A--!"n/P%!o%!t!He BC under the Se(e)t In&!3!&'( Lo'&-/BC
=lic; on )on3eR$otK)on3e) and then on )on3eRtopK)on3e) and then on
te/pR$'-ete/p
As you clic; on these each of the 5 is added to the A--!"ne& Lo'&/BC
At the bottom of the menu, clic; O0
#ac; in the /oad =ase :*, clic; App(.
H. AnalyBe ,sol$e- for temperature.
In T* select An'(.-!-. In S*, A)t!on= Analysis O$1e)t = entire model,
Met#o& = 'ull :un, Io$ n'/e = "in
=lic; t%'n-('t!on P'%'/ete%- and set output to Cp2 \ !rint
=lic;, CG
#ac; in the Analysis :*, clic; So(t!on T.pe
=hoose Ste'&. St'te An'(.-!-
=lic; O0
#ac; in the Analysis :*, choose S$)'-e C%e'te
@nder A3'!('$(e S$)'-e-, select finEcase
@nder A3'!('$(e Lo'&)'-e-, select finEcase
=lic; App(.
=lic; C'n)e(
#ac; in the Analysis :*, clic; S$)'-e Se(e)t
@nder S$)'-e- +o% So(t!on Se4en)e 1?=, select finEcase
@nder S$)'-e- Se(e)te&, clic; on Nefault ,this remo$es default from the list-
=lic; O0
#ac; in Analysis S* =lic; App(.
. To read in the results for post)processing
.In the :*=Analysis
Set A)t!on = +ead 7ut)ut2E O$1e)t = +esults $ntitiesE Met#o& = 'ranslate
=lic; Se(e)t Re-(t- F!(e
A S* appears called Se(e)t F!(e
=lic; the file +!n.op6
J1ou may need to loo; in your home or root directory to find the file. If
this file does not e+ist, then you ha$e made a mista;e in constructing
your model. Ao to E+plorer ,right)clic; on Start and choose E+plore- and
find the file fin.log and fin .f09. Cpen these files by double clic;ing on
them and search for the word KerrorL or KfatalL to determine what your
mista;e is-.
fin.o)2 then appears in the F!(e N'/e bo+
=lic; O0
,bac; in the An'(.-!- menu-
"2
=lic; App(.
20. Select the T* Re-(t-
A :* will appear called Re-(t-
Set A)t!on = Create
O$1e)t = 8uic9 Plot
1ou can display different results, but the main focus will be on the temperatures.
22. (e+t you will end your *S=>!AT:A( session by sa$ing your database and e+iting.
Cn the T* select F!(e
'rom the pull down menu select S'3e
Cn the T* select F!(e
'rom the pull down menu select 8!t
VI. E9ERCISES:
I. =ompare the 'EA results with the analytic results for the problem found from a 7eat
Transfer te+t. 7ow do the results compare. Niscuss any discrepancies.
II. =reate the temperature plot. Noes the distribution ma;e physical sense4 3hy or why
not4
III. 3hat assumptions are we ma;ing that might significantly affect the reliability of the
results4
"2