Spectrum Definition within CAE

[aidangoggin]

Hi,

Can anyone help me with the definition of a Spectrum within CAE, its in relation to a response spectrum analysis to recreate the base excitation of a component.

No one in my University has ever tackled this and they won't pay for support.

I'm trying to recreate a 1G test at 3.5mm from 20-200Hz.

From the Manual it says a spectrum should be defined with in the keyword editor as follows, does anyone have an example of any input so I could get an idea from it:

"Date lines to define a spectrum:
First line:
1. Magnitude of the spectrum.
2. Frequency, in cycles per time, at which this magnitude is used.
3. Associated damping, given as ratio of critical damping.
Repeat this data line as often as necessary to define the spectrum at all frequencies at each damping value."

I really need help with this please and any assistance would greatly be appreciated, thanks

 

[xerf]

If you use CAE and it is necessary to use keywords not supported by CAE (e.g. SPECTRUM) you have to use the keyword editor:
CAE->Menu->Model->Edit Keywords->"your_model"

Then you have to find the correct spot to fill in the additional model description. The correct place (i.e. "level") to insert a keyword is given in the keywords manual.

You have to be careful when using keywords editor because CAE generates automatically most of its content and some time it can end up in a mess if edited incorrectly. In the worst case, you might have to start from scratch your model, especially if you've just started learning ABAQUS and cannot identify the conflicting parts.

 

[aidangoggin]

hi xref,

I have found the location for the Definition to go in within the keywords editor,
what i'm having trouble with is the actual definition lines, for the spectrum.
The date lines as they call them, do i need to define lines for every frequency interval from 20-200Hz etc.

I have zero assistance on this one in Uni and i'm very new to ABAQUS so please excuse me if i'm a bit sloooow.

Thanks for your help

 

[xerf]

I am not familiar with Response spectrum analysis, but from
"Repeat this data line as often as necessary to define the spectrum at all frequencies at each damping value." I can infer that you face a "might need to" situation .

I work at a univ. with no ABAQUS support, also.

In this situation, the documentation is your best support and thanks God there's plenty of it. See:

ABAQUS Analysis User's Manual ->6.3.10 Response spectrum analysis

Good luck!

 

[aidangoggin]

Thanks,

I have been thru the documentation a number of times but i can't get my head around this section, it is not as well explained as some other areas, I'll battle on for another while,

I will try the definition again, fingers crossed

Thanks for your help xerf. [sorry i got your name wrong on the last post too]

 

[bkal]

I am not sure if I am missing something and not being an ABAQUS expert. I would imagine that you need provide one line of data per every frequency at which you want to define your spectrum, with each line containing three values: magnitude (acceleration?) frequency (Hz) and damping. Basically, you could end up with somethine like:

magnitude frequency damping ratio
0.0 0.0 0.05
0.1 1.0 0.05
0.2 2.0 0.05
.
1.0 5.0 0.05
.
0.2 40.0 0.05
You need to define your spectrum for a frequency range where you expect your strucutre to respond.

 

[aidangoggin]

This really driving me nuts, i have spent the last few days trying to get a Response Spectrum to work to no avail.
I have even built a very simple model to evaluate it but nothing.
As far as i can tell i have every thing in place correctly, the Spectrum is defined after the boundary conditions and the response spectrum references this named spectrum.
I have tried many variations of different parameters but to no avail, it just does not want to work for me, once processed a section appears in the output file but no equations are calculated.

Anyone got any ideas what I'm doing wrong, thanks

 

[xerf]

Maybe you have to modify the output request in order to include additional results (the ones you want) in the output files.

 

[aidangoggin]

thanks again for the responses, There must be some thing I'm missing, and hopefully some thing rather simple.

i have attached some of my Input file to see if anyone can stop a mistake,
Within the Spectrum i have the damping ratio set to zero as its a fraction of the critical damping ratio.
The following is from a small model that i'm trying to get to work first as the processing time is much lower.

If anyone can see what i am doing wrong i would really appreciate some assistance.

"** BOUNDARY CONDITIONS
**
** Name: BOUND Type: Displacement/Rotation
*Boundary
_PickedSet4, 1, 1
_PickedSet4, 2, 2
_PickedSet4, 3, 3
*Spectrum, NAME=SPEC, TYPE=VELOCITY, G=9810
0, 20., 0.
0.1, 21., 0.
........................... i have not included the full spectrum

0.3, 197., 0.
0.2, 198., 0.
0.1, 199., 0.
0, 200., 0.
** ----------------------------------------------------------------
**
** STEP: Step-1
**
*Step, name=Step-1, perturbation
freq extraction
*Frequency, eigensolver=Lanczos, acoustic coupling=off, normalization=mass, number interval=1, bias=1.
6, , , , ,
**
** OUTPUT REQUESTS
**
*Restart, write, frequency=0
**
** FIELD OUTPUT: F-Output-1
**
*Output, field, variable=PRESELECT
*End Step
** ----------------------------------------------------------------
**
** STEP: Step-2
**
*Step, name=Step-2, perturbation
resp spec
*Response spectrum, sum=SRSS
SPEC,1., 0., 0., 1.
*Modal Damping, definition=FREQUENCY RANGE
50., 0.4
100., 0.5
200., 0.6
1000., 0.7
5000., 0.6
**
** OUTPUT REQUESTS
**
**
** FIELD OUTPUT: F-Output-3
**
*Output, field
*Node Output, nset=Set-1
RF, RM, RT, U, UR, UT
**
** FIELD OUTPUT: F-Output-2
**
*Output, field, variable=PRESELECT
**
** HISTORY OUTPUT: H-Output-1
**
*Output, history
*Modal Output
GA, GU, GV
**
** HISTORY OUTPUT: H-Output-2
**
*Modal Output
GA, GU, GV
*End Step

 

[Soniau]

I think you should use:

*Spectrum, NAME=SPEC, TYPE=G, G=9810

I hope this would help you.

Be carefull whit the units