I am currently assessing an assembly for fatigue with Simulation 2010 but I'm having some bolt problems. The assembly consists of two flanged spools with three circular plates sandwiched between the two flanges. The flanges are bolted together with preloaded bolts, being two bolt groups via one of the threaded sandwich plates. The loading applied are a shear force and bending moment applied to the bottom spool, with the top spool restrained.
The bolts are modelled as bolt elements with preload, being a standard screw. I've set up frictional no penetration node to surface contact between plates. There are numerous loadcases, so each study is a duplication of the first, with a different magnitude of load applied for each.
Initially, the bolt elements had a tendency to randomly rearrange their attachment points to other holes. I have installed SP3.1 and so far this has not reoccurred. I have ran the static analysis and extracted the resultant bolt loads using "List Pin/Bolt/Bearing Force" function to compare against theoretical for verification purposes. The results of some loadcases provide predicted results (sinusoidal distribution along the bolt group) but some provide only preload. The locations of the bolts providing only preload don't follow any fixed pattern. This provides incorrect stresses around the bolting and therefore incorrect stress ranges effecting the fatigue.
I've found that if I redefine the bolt elements exactly as they were, some work correctly but not all. It seems as if the bolt elements are unstable and hence untrustworthy.
Has anyone had similar issues with bolt elements? And if so, how did you resolve it?
Thanks in advance