forrtl: severe (157...
 
Notifications
Clear all

forrtl: severe (157) error when trying to output ncforc

 

syed2468
(@syed2468)
Rookie
Joined: 4 months ago
Posts: 7
Topic starter  

Hello everyone,

I'm trying to model the three-point bending of a hollow aluminium tube. During this process there are a few different issues that I'm encountering. Firstly, when I output the interface force between the tube (Power Law Plasticity) and the indentor (rigid body) using ASCII_RCFORC I get intervals of time where the output force is zero. I've tried setting SOFT to 2, DEPTH to 5, and IGAP to 2 but that just results in a constant zero force. I've also tried outputting the contact force on the outer nodes of the tube using ASCII_NCFORC and defining the node set through *DATABASE_HISTORY_NODE_SET but that produces a forrtl: severe (157) error. I've tried the solutions that have been suggested in similar threads in the past like increasing memory and ncpu, enabling double precision, redefining contacts with and with SOFT, DEPTH, IGAP, SRP, and MRP settings, and making use of force transducers. If anyone can help me with these problem I would greatly appreciate it.


Quote
IceBreakerSours
(@icebreakersours)
Junior Member
Joined: 3 months ago
Posts: 22
 

Two comments -

a. Try a few different binaries (in Linux vs Windows).

b. See if you can reproduce this error in a simplified model.

 


ReplyQuote
syed2468
(@syed2468)
Rookie
Joined: 4 months ago
Posts: 7
Topic starter  

Thanks for your reply @icebreakersours. Can you please explain a bit more what you mean about trying different binaries? I'll try outputting NCFORC in a simple model and see how I go with that.


ReplyQuote
IceBreakerSours
(@icebreakersours)
Junior Member
Joined: 3 months ago
Posts: 22
 

By binaries, I meant versions of the code you are using. This is a quick/easy test to see whether there is something wrong with the model or the version of the code. If multiple binaries show you a similar end-result, odds are something is wrong with the model. It is possible that there is a bug that has gone unnoticed in multiple versions so, to be sure, this is not a fool-proof test. But given how convenient and efficient it is to run such a test, I recommend it.


ReplyQuote
Share: