I am running a simulation analysis on an assembly with about 60 parts and when ever I change something and need to add a contact set, Solidworks will crash as soon as I mesh the part and it gets to around 28%. I am getting about 5-7 system crashes due to this each day. I am running this analysis over a Network, but I have run similiar assembly's locally on my computer with the same issue. Does anyone have any ideas on what to try?
I found that my results were being saved over the network, so I changed this path and it still didn't help. I also updated my graphics driver and that didn't help either. My VAR had me wipe out Solidworks completely including everything Solidworks related out of the registry and this still didn't help. Now I am being told that I should reload my operating system. Does this sound right? Back in SW2009, it seems like I would get the "Solidworks has terminated due to an unhandeled error" message. Now in SW2010, I am getting "An error has occurred and Solidworks needs to re-start" I have attached a copy of the message I am seeing.
VARs are required to attempt to reproduce the issue for their paying customers. The VAR should rule-out if your issue is not file-specific. You can issue a "Pack and Go" command and zip up the data set, send the whole kit 'n kaboodle to them, and ask them to reproduce the issue on a different computer with the same version of software (down to the SP number). If they can - then that dataset should be given to SW Corp for repair (if they can repair it). If that's not it, the VAR should recommend shutting off software drivers in sequence (such as OpenGL, mouse drivers, video drivers to include "safe mode"). These are just suggestions but please work with the VAR to diagnose the issue in the proper order. Last resort is to reformat the hard drive.
Yes, the error message is one with which many users have become fmiliar. It occurs whenever the program crashes.
Try deleting and redefining the study. I've found these problems are worse afte you change the model and upgrade the study or change the study parameters.
You are right-on. I forgot about that one! I have had success also copy/paste the original study to a new one - then delete the old one; save; close; close SW; restart SW and it works fine. Go figure. Also defined a new study and drag and drop all the items that you can from the old one, to the new one. There are many times I've done these "fixes" and cannot reproduce them. The only pattern that seems to emerge is that something gets corrupt in the databases of the original studies (SW has informed me of this on several occasions...hmmm, how does a database get corrupt? - unanswered). Thank you for the tip. - Tony
I forgot about theother ones you have mentioned, especially the creating a new study and copyng across the bits and pieces from another study.
Derek Bishop wrote:
Yes, the error message is one with which many users have become fmiliar. It occurs whenever the program crashes.
Try deleting and redefining the study. I've found these problems are worse afte you change the model and upgrade the study or change the study parameters.
I've currently have an assembly that is giving me fits. It was successfully meshing and executing for a time until I made a change to the model. Now I can no longer get it to mesh successfully. Now, whenever I try to mesh, I get the message window indicating "these part have failed to mesh" and then SW crashes. I agree that the problems seem to multiply after changing parts of the model or the study. Any workaround to this? I've tried Pack and Go several times to new locations, deleting old studies, duplicating studies, and creating now ones.
Usually, this method solves my failing meshes :
After a new geometry is updated, have you tested the "interference detection" in the assembly. There might be bodies in interference.
For each part, have you tested the Geometry analysis for short edges, etc...
Alternatively, in the study, supress some or many of the parts ("Exclude from analysis"), then remesh.
Add, step by step, more parts, then remesh; until the failing mesh.
Mesh this failing part alone.
Eventually add a controled mesh (smaller mesh size) on that part. There might be a geometry dimension that is smaller than the global mesh size.
Hope this helps.
Yes. Two things. 1). Please run locally (as you already are). We have encountered numerous issues running over network and the SW guys tell us there's propensity to crash due to many files being written and read across the line, esp. if transient and time step data needs to be written. 2) Second thing we've had good success with, is running the SolidWorks Rx. Our customers report very good results running this periodically, esp. with regard to Simulation. In Program Files group, find the SW Program group and there should be a subgroup labeled "SolidWorks Tools". Run program "SolidWorks Rx" and locate the "System Maintenance" tab in there. Blast everything away if you can. Next, reboot computer. Close all programs except essential antivirus if possible. The meshers and solvers need lots of read-write address space on the memory cards. We hope this helps you.