modal computation results in output file

Numerical methods and mathematical models of Elmer
Post Reply
Kankreu
Posts: 23
Joined: 04 Dec 2012, 08:44
Antispam: Yes

modal computation results in output file

Post by Kankreu »

Hello All,

I'm doing some simple modal computation (vs. Analytical results) of a cantilever beam with circular cross-section.
I get pretty good results (if mesh is fine enough).
Strangely the eigenvalues (or freq) don't appear in the result file (.ep), even at header of the modeshape table:
#time 1 1 1.00000000E+000
I see only the eigenvalue in the console:
EigenSolve: --------------------------------
EigenSolve: 1 ( 152756.31486517368 , 0.0000000000000000 )
EigenSolve: 2 ( 152770.98895009619 , 0.0000000000000000 )
EigenSolve: 3 ( 5900144.4374490837 , 0.0000000000000000 )
EigenSolve: 4 ( 5900626.1985544795 , 0.0000000000000000 )
EigenSolve: 5 ( 45067578.488053322 , 0.0000000000000000 )
EigenSolve: 6 ( 45071819.275828302 , 0.0000000000000000 )
EigenSolve: --------------------------------

It's ok as If I run in batch, I can do elmersolver case.sif > console.log but I think it might be useful to dump that in the .ep file.

Or do I miss something in the .sif?

Thanks,
K
Attachments
case.7z
(1023 Bytes) Downloaded 270 times
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: modal computation results in output file

Post by raback »

Hi K,

The .ep file is for distributed data only. For scalars data use SaveScalars solver and treat the values in some other software (excel, Matlab,..). There is a a keyword for the SaveScalars solver "Save Eigenvalues", or something, that lets you save the eigenvalues.

-Peter
Kankreu
Posts: 23
Joined: 04 Dec 2012, 08:44
Antispam: Yes

Re: modal computation results in output file

Post by Kankreu »

Hi Peter,

thanks for you reply, I'll try that.

Greetings,
K
Kankreu
Posts: 23
Joined: 04 Dec 2012, 08:44
Antispam: Yes

Re: modal computation results in output file

Post by Kankreu »

Hi Peter,

thanks for you reply, I tried and it works.

The syntax is the following:
"
Solver 2
Exec Solver = After Saving
Equation = Linear elasticity ! reference to my solver 1
Procedure = "SaveData" "SaveScalars"
Filename = "Eigenvalues.dat"
Variable 1 = Displacement
!Save Eigenvalues = True
Save Eigenfrequencies = Logical True
End
"

just a few questions:
- Why should I refer Displacement as Variable 1?
- I got lot of problem to get the frequencies.
I have to use "Save Eigenfrequencies = Logical True" which is not working with just True, why?

Greetings,
K

PS:
But won't it make sense to put just the frequency value in the header of the "frequencies datablock"?
#time 1 1 1.00000000E+000
into
#time 1 1 1.234574E+002
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: modal computation results in output file

Post by raback »

Answers:
1) SaveScalars is a generic auxiliary solver that could be used to work on many fields. Here you just have one to work with, the Displacement.
2) The keyword is not just in the SOLVER.KEYWORDS database and hence the type must be specified. There are ~1200 keywords in the database with the list is never exhaustive.
-Peter
Post Reply