"Small equation pivot term" error causes analysis to fail

[numanchris]

I've created and applied constraints/loads to a fairly simple car chassis in APDL (created using keypoints/lines and defining a hollow rectangular cross section). The analysis works until I add the final 10ish members to the chassis and then the problem wont solve. Instead I receive the "There is at least 1 small equation solver pivot term. Please check for an insufficiently constrained model" error message.

I don't see how it can be unconstrained if adding in a few extra bars causes it to fail when it was working perfectly before?

Any help would be great, I'll attach my log file too,

thanks!

 

[numanchris]

An update...adding the command "PIVCHECK,OFF" at the end of the log file allows ANSYS to solve the simulation without the pivot error stopping it and highlights my problem...it seems certain lines aren't actually connected to one another...in certain places ive added in a keypoint using the "add keypoint w/ ratio" button to add a keypoint a certain distance along a line...it seems to be these new keypoints that are causing the issue but I dont understand why?

Thanks!

 

[petb]

Hi,

They cause problems because they are not attached to the underlying line geometry. You define new lines not attached to the original geometry. After meshing you have a number of unconstrained elements that causes the pivot error. Either redefine your lines using the extra KP:s or use the command "lovlap" to fix it. Do not turn off the small pivot check. If you are having small pivots you are doing something wrong. Turning it off and solving and solving won't make it right.

/petb

 

[numanchris]

That's great, thanks, sorted now! I'm fairly new to ansys and I didn't realise if you placed a keypoint in the middle of a line it didn't become part of the line, silly me! And I just cturned the pivcheck off to highlight where the error was, cheers