I am currently modelling a joint of a geodesic dome. I previously ran an analysis only with preload of bolts ignoring all contact surfaces of the shank of the bolt and other geometry . This time I ran the analysis considering all contact surfaces(Normal Behaviour: hard contact, Tangential behavior:penalty) and provided loads on all faces of the members(face perpendicular to the face of the circular plate).
I have just started working on abaqus so please provide me some insight.
Difficult to say, but you should probably switch NLGEOM on, that's usually better in contact.
Just 2 ideas to make the model more robust:
The slave can be meshed coarser than the master.
The position tolerance for tie-tie (and other) constraints can be increased. However , it should also be checked from which value on the result starts to change.
Hi Sajith,
See if the attached Convergence guidelines helps you solve the issue.
Martin Baeker Sir, I did run it with NLGEOM on, but error kept on repeating. Later I reduced the increment size less than 20 and I got convergence. Right now my doubt is does reducing the step increment cause any inaccuracy in results.
Sorry, I don't understand and don't have enough information to say anything. Is it implicit/explicit? Which parameter did you change? Maximum time increment? Step time?
- Badges
- Science topic