10.

0 TRANSIENT DYNAMIC ANALYSIS

Introduction

Displacement, u

Transient dynanics analysis, sometimes called Time-History Analysis, is a technique used to

determine the dynamic response of a structure under the action of any general time dependant
loads. This type of analysis is used to determine the time-varying displacements, stresses,
strains, stresses and forces as it responds to any combination of static, transient and harmonic
loads. The time-scale of loading is such that inertial or damping effects are considered to be
important. The following graph of displacement versus the ratio of Input Frequency to
Output Frequency shows clearly when a transient dynamic analysis should be performed. If
the input frequency is low, then the response can be represented by an equivalent static
analysis. If the input frequency is too high, then the mass does not have time ot respond to the
loading. The critical region is shown between the two dotted lines where the response of the
structure is controlled by the damping of the structure and therefore, since the maximum
amplitude results in this region, a transient analysis taking into account the inertial effects must
be performed.

DO WE ANALYSE ??
YES
NO

NO

Input f is high
so mass cannot
respond t o load
Pseudo
St at ic

Input f=low

1
response cont rolled
by damping

Input f
Out put f

Maximum amplit ude occurs when Input f = Out put f

76

10.1 DAMPING
Damping is present in some form in most structures. The general equation of motion shows
damping associated with the velocity terms. Indeed, the damping present in the structure can
be thought of as an energy absorber. The
C = 0.1
funct ion
U
following diagrams show the response of a
of damping simple simple one-mass structure to variations in
the amount of damping present in the structure.
t
Generally, the amount of damping present is
small, of the order of 5 to 15 per cent in most
structures. In these cases, the displacement
C
=
1
U
response as a function of time is shown. Indeed,
this form of function is often termed
underdamping and is the most naturally present
form of damping.
t

If the level of damping is high, then the structure

is overdamped and the controlling amplitude is
more steep.

C= 0

t
If there is no damping present (as in the case of
modal analysis) then the energy remains in the structure, i.e. kinetic energy - potential energy kinetic energy - potential energy, and so on.
It is important to understand the significance of damping. For most transient analyses, the
value of damping is not particularly significant, that being in the order of 5-15%. variations
within this band will not have too much of a change in the structural response, especially if the
nature of the response is long term. However, if the response is short term, as in the pinging of
a glass or if the response is close to the resonant frequency, then the significance of the level of
damping becomes very important.
Note: If the design response in the stready state condition is near resonance, then mass
should be added to the structure to alter the natural frequency of the system.

77

There are various methods of including damping in most modern finite element codes.
Damping can be added via any or all of the following methods:

Mass Damping (ALPHAD), Material Damping (BETAD), & Elemental Damping

The complete expression for the damping matrix, [C] can be written
C = M + K +

NMAT

NEL

j =1

k =1

j K j + cc K +

Ck

where M represents mass damping, etc represent structural damping, and the final term
represents elemental damping. This form of damping equation is known as Rayleigh
Damping and is often represented by the expression

C = M + K
This implies that the parameters and are required in the input to the analysis. In Ansys, the
ALPHAD and BETAD commands apply. This model of damping is incorporated in most codes
since it is the only simple method of introducing damping. The values of and are decimal
numbers are not normally known, but are calculated from the modal damping ratios . The
modal damping ratio is the ratio of the actual damping to the critical damping for a particular
mode of vibration. If i is the natural circular frequency of a particular mode of interest, mode
I(normally 2f ), then and satisfy the relation
i

i =
+
2 i
2

Damping Ratio,

With -damping, the damping ratio is inverseely proportional to frequency. That is, lower
frequencies will be damped more and higher frequencies will be damped less. Therefore the
most dominant frequency should be used to calculate . With -damping, the damping is
directly proportional to frequency therefore lower frequencies will be damped less and higher
frequencies damped more. Graphing the function for damping ratio gives,

t ot al
damping
damping
f1

f2

Frequency (Hert z)
78

Note that the sum of the two damping functions is nearly constant over the frequency range
where they intersect. Therefore, given a damping ratio () and a frequency range in hertz (f 1
and f 2) two equations can be solved simultaneously for and .

+ f1 =
4f1

+ f 2 =
4f 2
10.2 LOAD HISTORY CURVE

stepped 3

Load

Load

A transient analysis, by definition, involves loads which are functions of time. In order to
specifiy such loads, the load versus time curve must be divided into suitable load steps. From
the following diagrams, each corneron the load-time curve may be one step.

1
2

stepped

4
3

5
Time

Time

For each load step, the load values and the time values must be specified in order to define
each corner of the load history. Options are available to allow the user to specify whether the
load step is applied as a ramp function over the time step or whether the load is applied as a
step function. In Ansys, the default is the ramp-type load step. The step-type function is
activated within a load step by using the KBC command (KBC,0 - default ramp-type load and
KBC,1 - step-type load).
Most load types are valid in transient analyses. Application of constraints (D commands),
forces (F commands), surface loads (pressure via SFL, SFA commands) and body loads (TEMP
command etc.) are all possible. The time at the end of the load step is entered via the TIME
command, i.e. TIME,0.02 etc.
79

Since the load history curve is of a transient nature, the equations of motion must be solved in
a manner which takes account of the time-varying load. These are solved by employing time
integration of the equations of motion and therefore the integration time step (ITS) is the time
increment used in this process. The selection of the integration time step is particularly
important.

T - t oo la r g e

The time step size determines the accuracy of the solution: the smaller its value, the higher the
accuracy. There are several factors which are worth considering in the selection of a good
integration time step. A time step which is too large will introduce errors which affect the
response of higher modes and hence the overall response. A time step which is too small will
waste computing (and your own !) resources. There are two main guidelines which can help
choose the optimum time step.
1) Attempt to resolve the response frequency. This implies that the time step should be small
enough to resolve the motion of the structure. Since the dynamic response can be thought of
as a combination of modes, the time step should be able to resolve the highest mode which
contributes to the response. For the Newmark integration scheme (programmed in Ansys), it
has been found that using approximately 20 points per cycle of the highest frequency of
interest results in a reasonable accurate solution. Therefore, if the highest frequency is f in
Hertz, then the recommended ITS can be found from ITS=1/20f. The following figure shows
the effect of ITS on a single DOF spring mass system. Note that 20 or more points results in
less that 1% error.

Effect of Integration Time Step on Period Elongation

80

2) Resolve the applied load-versus-time curve(s). The time step should be small enough to
follow the loading function. The response tends to lag the applied loads, especially for
stepped loads. Stepped loads require a small ITS at the time of the step change so that the step
can be closely followed.

Transient Input versus Transient Response

After calculating the time step using the above guidelines, the minimum values should be used
in the analysis. Most programs use a technique known as automatic time stepping
(AUTOTS,ON) which lets the program decide when to increase or decrease the time step
during the solution. Auto-time stepping is also known as time step optimisation and this
routine attempts to adjust the ITS during the solution based on the response frequency.
The main benefit of this is that the total number of substeps may be reduced resulting in
computing resource savings. Indeed, if auto-time stepping were not available, then the user
would require to re-run the analysis again and again adjusting the time steps manually.
Generally, it is a good idea to activate auto-time stepping in all transient analysis, althogh in
some specialised cases, it may not be beneficial (see program manuals for details).

81

10.3 TRANSIENT DYNAMIC ANALYSIS IN ANSYS

Transient dynamic analysis in Ansys, is not too difficult provided the above mentioned rules
are considered in the selection of the damping values and integration time steps. The geometry
and finite element model is created in the usual manner in PREP7 with loads and boundary
conditions being applied in the SOLUTION phase. There as various types of analysis options
such as FULL, REDUCED, MODAL SUPERPOSITION, however only the FULL method is
being considered here. This is selected using the transient option command, TRNOPT,FULL.
On entering the solution phase, by typing /SOLUTION, the ANTYPE,TRANS option is selected.
If a restart analysis is being undertaken, then ANTYPE,TRANS,RESTART is specified.
Thereafter, the boundary conditions may be applied and the load-history curve may be defined.
This definition forms a regular pattern and is dependent on the form of curve. Assuming, for
example, the load at node 44 varies from 0 to 200N acting down in the vertical direction for 3
seconds, remains constant for a further 5 seconds, then is step changed to half its value,
remains at this for 2 seconds and ramps to zero in a final 3 seconds. The Ansys command
sequence would look like this:
C*** Initialise force
F,44,FY,0
TIME,0.01
LSWRITE
F,44,FY,-200
TIME,3
LSWRITE
F,44,FY,-200
TIME,8
LSWRITE
F,44,FY,-100
KBC,1
TIME,10
LSWRITE
F,44,FY,0
KBC,0
TIME,13
LSWRITE
LSSOLVE,1,5
Each load step is defined using the LSWRITE command. The solution is initiated using the
LSSOLVE command for the number of load steps under consideration in the run. Normally
82

additional information, such as the time step (DELTIM,value) and damping values (ALPHAD
and BETAD) would have been specified prior to executing the solution. Automatic timestepping is recommended and is initialised using the AUTOTS,ON command. Solutions should
be viewed by checking the OUTPUT window. This allows more information to be viewed
which pertains to the development of the integration routine as it progresses along the time
history curve. Once the solution is activated, the information going to the screen looks like the
following:
Load step file number 2. Begin solution ...
*****

ANSYS SOLVE

COMMAND
L O A D

*****
S T E P

LOAD STEP NUMBER. . . . . . . . .

TIME AT END OF THE LOAD STEP. . .
AUTOMATIC TIME STEPPING . . . . .
STARTING TIME STEP SIZE. . . .
MINIMUM TIME STEP SIZE . . . .
MAXIMUM TIME STEP SIZE . . . .
STEP CHANGE BOUNDARY CONDITIONS .
TRANSIENT (INERTIA) EFFECTS
STRUCTURAL DOFS. . . . . . . .
TRANSIENT INTEGRATION PARAMETERS
ALPHA. . . . . . . . . . . . .
DELTA. . . . . . . . . . . . .
RAYLEIGH DAMPING MULTIPLIERS
ALPHA (MASS) . . . . . . . . .
BETA (STIFFNESS) . . . . . . .
PRINT OUTPUT CONTROLS
ITEM
FREQUENCY
COMPONENT
ALL
NONE
DATABASE OUTPUT CONTROLS
ITEM
FREQUENCY
COMPONENT
ALL
ALL

.
.
.
.
.
.
.

.
.
.
.
.
.
.

O P T I O N S
.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
2
. 1.5000
.
ON
. 0.62083E-01
. 0.62083E-01
. 1.4900
.
NO

. . . . . . .

ON

. . . . . . . 0.25251
. . . . . . . 0.50500
. . . . . . . 3.0000
. . . . . . . 0.20000E-02

Equation Solution Element= 52 Cum. Iter.= 2 CP= 108.810

Time= 0.72083E-01 Load Step= 2 Substep= 1 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
1 COMPLETED.
CUM ITER =
2
*** TIME = 0.720833E-01
TIME INC = 0.620833E-01 NEW TRIANG MATRIX
*** RESPONSE FREQ = 0.9020
PERIOD=
1.109
PTS/CYC = 18.
*** AUTO STEP TIME: NEXT TIME INC = 0.62083E-01 UNCHANGED
Solution Preparation Element= 18 Cum. Iter.= 3 CP= 119.790
Time= 0.13417 Load Step= 2 Substep= 2 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
2 COMPLETED.
CUM ITER =
3
*** TIME = 0.134167
TIME INC = 0.620833E-01 OLD TRIANG MATRIX
*** RESPONSE FREQ = 0.5092
PERIOD=
1.964
PTS/CYC = 32.
*** AUTO STEP TIME: NEXT TIME INC = 0.62083E-01 UNCHANGED
Element Formation Element= 53 Cum. Iter.= 4 CP= 130.780
Time= 0.19625 Load Step= 2 Substep= 3 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
3 COMPLETED.
CUM ITER =
4
*** TIME = 0.196250
TIME INC = 0.620833E-01 OLD TRIANG MATRIX
*** RESPONSE FREQ = 0.3969
PERIOD=
2.519
PTS/CYC = 41.
*** AUTO TIME STEP:
NEXT TIME INC = 0.12597
INCREASED (FACTOR =
2.0291)
83

Element Formation Element= 48 Cum. Iter.= 5 CP= 141.760

Time= 0.32222 Load Step= 2 Substep= 4 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
4 COMPLETED.
CUM ITER =
5
*** TIME = 0.322225
TIME INC = 0.125975
NEW TRIANG MATRIX
*** RESPONSE FREQ = 0.3374
PERIOD=
2.964
PTS/CYC = 24.
*** AUTO STEP TIME: NEXT TIME INC = 0.12597
UNCHANGED
Element Formation Element= 28 Cum. Iter.= 6 CP= 152.860
Time= 0.44820 Load Step= 2 Substep= 5 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
5 COMPLETED.
CUM ITER =
6
*** TIME = 0.448199
TIME INC = 0.125975
OLD TRIANG MATRIX
*** RESPONSE FREQ = 0.2945
PERIOD=
3.395
PTS/CYC = 27.
*** AUTO STEP TIME: NEXT TIME INC = 0.12597
UNCHANGED
Element Formation Element= 24 Cum. Iter.= 7 CP= 163.790
Time= 0.57417 Load Step= 2 Substep= 6 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
6 COMPLETED.
CUM ITER =
7
*** TIME = 0.574174
TIME INC = 0.125975
OLD TRIANG MATRIX
*** RESPONSE FREQ = 0.2612
PERIOD=
3.829
PTS/CYC = 30.
*** AUTO STEP TIME: NEXT TIME INC = 0.12597
UNCHANGED
Element Formation Element= 20 Cum. Iter.= 8 CP= 174.880
Time= 0.70015 Load Step= 2 Substep= 7 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
7 COMPLETED.
CUM ITER =
8
*** TIME = 0.700148
TIME INC = 0.125975
OLD TRIANG MATRIX
*** RESPONSE FREQ = 0.2299
PERIOD=
4.350
PTS/CYC = 35.
*** AUTO STEP TIME: NEXT TIME INC = 0.12597
UNCHANGED
Element Formation Element= 13 Cum. Iter.= 9 CP= 185.760
Time= 0.82612 Load Step= 2 Substep= 8 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
8 COMPLETED.
CUM ITER =
9
*** TIME = 0.826123
TIME INC = 0.125975
OLD TRIANG MATRIX
*** RESPONSE FREQ = 0.2000
PERIOD=
4.999
PTS/CYC = 40.
*** AUTO STEP TIME: NEXT TIME INC = 0.12597
UNCHANGED
Element Formation Element= 10 Cum. Iter.= 10 CP= 196.850
Time= 0.95210 Load Step= 2 Substep= 9 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
9 COMPLETED.
CUM ITER =
10
*** TIME = 0.952097
TIME INC = 0.125975
OLD TRIANG MATRIX
*** RESPONSE FREQ = 0.1738
PERIOD=
5.753
PTS/CYC = 46.
*** AUTO TIME STEP:
NEXT TIME INC = 0.28764
INCREASED (FACTOR =
2.2833)
Element Formation Element= 5 Cum. Iter.= 11 CP= 207.780
Time= 1.2397 Load Step= 2 Substep= 10 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
10 COMPLETED.
CUM ITER =
11
*** TIME =
1.23973
TIME INC = 0.287637
NEW TRIANG MATRIX
*** RESPONSE FREQ = 0.1430
PERIOD=
6.993
PTS/CYC = 24.
*** AUTO TIME STEP:
NEXT TIME INC = 0.26027
DECREASED (FACTOR =
0.9048)
Element Formation Element= 1 Cum. Iter.= 12 CP= 219.870
Time= 1.5000 Load Step= 2 Substep= 11 Equilibrium Iteration= 1.
Element Output Element= 51 Cum. Iter.= 12 CP= 230.800
Time= 1.5000 Load Step= 2 Substep= 11 Equilibrium Iteration= 1.
*** LOAD STEP
2
SUBSTEP
11 COMPLETED.
CUM ITER =
12
*** TIME =
1.50000
TIME INC = 0.260266
NEW TRIANG MATRIX
*** RESPONSE FREQ = 0.1246
PERIOD=
8.025
PTS/CYC = 31.
FINISH SOLUTION PROCESSING

84

Items shown above, such as iteration number, load step, substep, auto-time step values show
how the solution is progressing and are worth watch especially in regions where the load
history curve makes a change in direction (i.e. at the corner points).
Once the solution is complete, the results must be examined in POST1, the general postprocessor or POST26, the time-history post-processor (or both!).
Postprocessing
General post-processing for deflected shapes and stresses at any point in the analysis can be
done in POST1. By selecting the correct load step and iteration, the required point in the
analysis can be obtained. This is perform using the SET command (SET,load step,iteration). The
iteration number refers to the iteration within the load step and not the cumulative iteration
number. The last iteration in any load step can be selected by typing SET,1,LAST for load step
1 and so on.
If time-history results are required, the POST26 must be used. This is entered by typing
/POST26 and the required result and position must be chosen so that a plot versus time may be
generated. Ansys then examines that point in each load step iteration and generates a curve.
Nodal data is selected using the NSOL command (NSOL,var,node,type,direction), where var is a
data variable (say 2, 3 etc), node is the node under examination, type refers to the data type
(u-displacements, v-velocity etc.) and direction relates to the direction of the response chosen
(x, y, z). For example, say the displacement at Node 44 were required from time equals zero
to time equals 5, then the command sequence would be:
/POST26
TIME,0,5
NSOL,2,44,U,Y
PLVAR,2
c*** if three results were requested on the same graph
NSOL,2,44,U,X
NSOL,3,44,U,Y
NSOL,4,44,U,Z
PLVAR,2,4
C*** this would result in 3 separate traces vs time.

Various data items are available for both nodal and elemental results. Additional items such as
reaction force can also be extracted.
85

Case Study - Transient Analysis of a Diving Board

A diving board is proposed with dimensions, 4.5m long, 750mm wide and 60mm thick. The
material is GRP with a modulus of 30x106N/m2 and a density of 2000kg/m3. The load is
applied over a section of the board adjacent to the outermost edge. The board is fully fixed at
the inner edge and restraint in the vertical direction one third of the way along the length. The
load is ramped on for 1.5 seconds and off in a further 1.5 seconds. The analysis will progress
for a further 9 seconds to study the response of the board. Suggested time steps have been
made from a preliminary modal analysis (ITS=1/20f) from the third natural frequency. Auto
time-stepping is switched on. Damping is initially selected at =0.1 and =0.00002. A study of
the influence of damping is undertaken. A full listing of the input is found on the class web
pages as TRANS.INP.
The following sequence of slides shows the influence of varying both ALPHAD and BETAD on
the vertical displacement at the tip of the board (Node 24).
1

ANSYS 5.0 A
FEB 26 1995 1
19:55:51
PLOT NO.
1
POST26
ZV =1
DIST=0.75
XF =0.5
YF =0.5
ZF =0.5
CENTROID HIDDEN

=0.1
=0.00002

ANSYS 5.0 A
FEB 26 1995
19:56:38
PLOT NO. 2
DISPLACEMENT
STEP=4
SUB =4
TIME=3.763
RSYS=0
DMX =63.99
SEPC=25.277

Y
Z

DSCA=0.003191
XV =1
YV =1
ZV =1
DIST=2.042
XF =2.25
YF =-0.10209
ZF =0.375
CENTROID HIDDEN

UY

Transient Analysis of Diving Board

Y
Z

X
MN
MX

Transient Analysis of Diving Board

Transient Analysis of Diving Board

ANSYS 5.0 A
1
FEB 26 1995
19:57:27
PLOT NO. 3
NODAL SOLUTION
STEP=4
SUB =4
TIME=3.763
SX
(AVG)
TOP
RSYS=0
DMX =63.99
SMN =-0.230E+07
SMNB=-0.641E+07
SMX =0.166E+08
SMXB=0.205E+08
-0.230E+07
-203343
0.189E+07
0.399E+07
0.608E+07
0.818E+07
0.103E+08
0.124E+08
0.145E+08
0.166E+08

=0.1
=0.002

Transient Analysis of Diving Board

86

ANSYS 5.0 A
FEB 26 1995
20:07:56
PLOT NO.
4
POST26
ZV =1
DIST=0.75
XF =0.5
YF =0.5
ZF =0.5
CENTROID HIDDEN

UY

ANSYS 5.0 A 1
FEB 26 1995
20:23:07
PLOT NO.
5
POST26
ZV =1
DIST=0.75
XF =0.5
YF =0.5
ZF =0.5
CENTROID HIDDEN

=0.3
=0.002

=3.0
=0.002
UY

ANSYS 5.0 A
FEB 26 1995
20:42:26
PLOT NO.
7
POST26
ZV =1
DIST=0.75
XF =0.5
YF =0.5
ZF =0.5
CENTROID HIDDEN

UY

Transient Analysis of Diving Board

Transient Analysis of Diving Board

The influence of BETAD is small and the influence of ALPHAD is significant in this case.
However, close examination of the results shows that although the response when =0.3 is as
expected, the response time of 6-8 seconds to reach maximum amplitude is much more than
would be reasonable. Therefore, the choice of GRP alone as a suitable material must be
questioned (perhaps laminated wood and GRP ? - examine at your leisure !).
10.4 FINAL COMMENTS
Some comments are worth making at this stage on both dynamic analysis and Ansys usage.
When performing dynamic analyses, make sure the units are consistent! If you use newtons,
millimetres, kilograms and seconds, then the frequency will be in millihertz - use metres to
avoid this. Check the units of Youngs Modulus too!
Be careful on the choice of time step. For example, a linear static analysis of the diving board
takes 25 seconds to solve on a 486-66PC. For the transient analysis, 41 iterations were
required taking 700 seconds to solve and this is a simple model ! Dont use excessive model
sizes. Use shell and beams where possible. Make a 3D-solid model only as a final resort.
Making multiple windows: Ansys, by default plots to Window 1, so when you type
/VIEW,1,1,1,1 the first 1 refers to the window number and the rest to X,Y,Z respectively.
Therefore if multiple windows are required, then the /WIND command is used. Ansys sizes the
screen in +1/-1 coordinates in X and Y, with the centre of the screen being (0,0). The
following sequence gives three windows, one in the full left hand section (window 1) and two
equal sized in the right hand section (window 2 top and window 3 bottom). If different
displays are required in each window, issue plot commands after /WIND command.
87

/WIND,1,-1,0,-1,1
EPLOT
/WIND,2,0,1,0,1
PLDISP
/WIND,3,0,1,-1,0
NPLOT

If superimposed pictures are required, issue /NOERASE before plotted (/ERASE will cancel).
Output is best obtained by creating picture files (/SHOW,picfile) and using the ANSYS Display
program from within Windows. Thereafter, use the clipboard to put the required plots into
Microsoft Word.
File TRANS.INP (c) D H Nash 25th Feb 1996
/PREP7
/TITLE,Transient Analysis of A Diving Board
C*** Define parameters
length=4.5
width=0.75
thick=0.06
C*** Create geometry
K,1
K,2,0,0,width
KGEN,2,1,2,1,length
A,1,2,4,3
C*** Select Element Type and Size
ET,1,43
ESIZE,width/3
ESHAPE,2
C*** Select Element Thickness
R,1,thick
C*** Mesh areas
AMESH,1
C*** Mat props for Isotropic GRP
MP,EX,1,30E6
MP,NUXY,1,0.3
MP,DENS,1,2000
C*** Select Fixed Boundary at edge
NSEL,R,LOC,X,0
D,ALL,ALL
NSEL,ALL
C*** Select Board support at 25% along length +/- WIDTH/3 scan
NSEL,R,LOC,X,(0.25*length)-(width/3),(0.25*length)+(width/3)
D,ALL,UY,0
NSEL,ALL
SAVE
FINI
C*** Remove the following comment lines to perform a Modal Analysis
c*** /SOLUTION
C*** /TITLE,Modal Analysis for 1st Five Modes using Full Subspace
C*** ANTYPE,MODAL
C*** MODOPT,SUBSP,5
C*** SOLVE
C*** FINI
c*** /SOLUTION
C*** EXPASS,ON
C*** MXPANS,5
88

C*** SOLVE
C*** FINI
/SOLUTION
/TITLE,Transient Analysis of Diving Board
ANTYPE,TRANS
TRNOPT,FULL
C*** Apply load as ramped function (KBC,0) Note: stepped is KBC,1
KBC,0
C*** Define F3 = third natural frequency
F3=0.8
ITS=1/(20*F3)
DELTIM,ITS
AUTOTS,ON
ALPHAD,3
BETAD,0.002
OUTRES,,ALL
OUTPR,ALL
C*** Initialise force
C*** Select last part of board for load application
NSEL,R,LOC,X,length-width,length
F,ALL,FY,0
NSEL,ALL
TIME,0.01
LSWRITE
C*** Select last part of board for load application
NSEL,R,LOC,X,length-width,length
F,ALL,FY,-200
NSEL,ALL
C*** Define time at end of step
TIME,1.5
LSWRITE
C*** Remove forces
NSEL,R,LOC,X,length-width,length
FDEL,ALL,ALL
F,ALL,FY,0
NSEL,ALL
TIME,3
LSWRITE
TIME,12
LSWRITE
LSSOLVE,1,4
FINI

89