Problem with input reading parfeap

[JReinoso]

Dear Feap members,

Currently I am generating a quite large micromechanical model for firber reinforced composites (around 2 million elements). the point is that prior to testing my user elements I am trying to check the input file that I generate. Sorry for bothering you but I try to follow the guide of parfeap to do so but  I am not able that the prescribed boundary conditions are read by FEAP beaucse the residual is always zero. It should be a stupid problem.
 
I attach the input file and the auxiliary files

[FEAP_Admin]

Please post the ISaitoJR_xxxx files and the _mesh.txt file.  If these are too large to post to the forum, please email them to feap@berkeley.edu .

[JReinoso]

Thanks for your prompot reply. They are quite large. I will share with the given email a folder in google drive (if it does not work, I would appreciate that you can indicate me an alternative way to transfer the files)

[FEAP_Admin]

There are a couple of issues at play.  (1) You have commented out the EDISP command; thus the boundary values FEAP sees are zero.  (2) If you uncomment the EDISP command you still get zero boundary values because FEAP is finding no nodes on the plane with x1 = 0.0974504.   The issue is the automatic tolerances for the search.  For your input FEAP is computing a gap of 3.9993286199996315e-08.  This is quite strict and is thus finding zero nodes.

You can make the gap larger using the GAP option to EDISP and EBOUN; however, this can be a tricky to get correct (getting just the correct nodes and no more).

Code: [Select]

EBOUN
GAP 0.0001
1 0.0974504 0.002 0 0

EDISP
GAP 0.0001
1 0.0974504 0.002 0 0

Another and perhaps better option would be to have whatever program is creating the _mesh.txt file to include the boundary conditions in the file.