Hi all,
Recently, I am using ansys ls-dyna to simulate the rolling of aluminum plates with shrinkage cavities, and have successfully calculated several operations. However, after changing the shrinkage hole size, a current job is always unable to be calculated, and an early termination occurs. I used the high computing platform provided by the school to output the result prompt: "forrtl: severe (174): SIGSEGV, segmentation fault occurred", and the prompt output by my own computer is as shown in the title. I have tried redefining the contact, changing the file path, changing the encoder, etc., but they have not been able to solve it.
It is worth mentioning that the place where the error occurs each time is at the same physical movement time node, that is, the time when the rolled piece has just been rolled to half its length.
I sincerely hope that you experienced seniors can give me some suggestions, thank you!
Hi @tttthhhxx,
Have you checked the existing topics here related to addressing that same issue? Looks like the usual suspects to cause this are:
- not running Dyna with double precision
- running on a Windows machine
- SMP vs MPP
- Possible contacts
Here are some previous topics:
https://feassistant.com/forums/ls-dyna/error-severe-157-program-exception-access-violation/#post-643
https://feassistant.com/forums/ls-dyna/the-error-of-ls-dyna-cutting-simulation/#post-281
@negativevolume I will check the question you mentioned and try to recalculate tomorrow morning. Thank you very much for your answers.
@negativevolume Hi! thank you for your reply again. I thought about the problem you mentioned, but I don't understand why another k file with the same parameter settings can be calculated smoothly if such a problem occurs (the difference between the two files seems to be the only difference in the size of the shrinkage hole). I have already sent the screenshot of my error report and two k files to contact@feassistant.com. If possible, could you please check it for me? Thanks a lot!!
Friends, I seem to know the solution to my model's problem. Because there are many pairs of contacts in my model, I split their contact time, and then it can be solved successfully. Although I haven't figured out why it can be successfully calculated without splitting it before, but fortunately, this problem can be bypassed.
@tttthhhxx Sorry if this is a silly question, but what do you mean about splitting contact time? Like do you mean you are setting birth and death times for them manually?