different results wrt. ENHA option

[sicheng6699]

Dear FEAPers,

The cantilever using brick element is rigid supported on one side, and a minimum body force in x (longitudinal direction) is applied on it.
The linear elastic cases are carried out, with and without ENHA option is calculated, which give total different results. (I think with the ENHA option is wrong.) The screen shots show the deformation.

Maybe there is something wrong within the input file. Could you please help me to find where and why?
Thanks a lot!

Ps.: FEAP version 84, linux 64

Best regards

[sicheng6699]

The density is used for the computation of eigen frequencies.
It seams that, if the density is changed, the deformation with ENHA option is changed correspondingly.
Why are they related? 

[Prof. R.L. Taylor]

There is a bug in the residual file (resid3d.f in ./elements/solid3d)  -- around line 98 there is a statement that should read:

r(i,j) = r(i,j) - (cfac*ac(i) + lfac*al(i,j))*aj0

Currently there is a ** between the lrfac  and al(i,j)

I hope this fixes your bug report -- let us know

[sicheng6699]

It works correctly! Thank you so much!
Now the eigen frequencies with ENHA option are also correct.

May I ask another question about eigen form.
Is it also possible to export the eigen vectors with something like "OUTP EIGV n1 n2 ", in a file like Itest.eigv? 

 

[Prof. R.L. Taylor]

Currently, we do not output eigenvectors, however, you can probably add it to the pouta.f routine if you wish.  The eigenvectors are stored without the fixed dof values (in compressed form) same as for the residual vector.