AnsweredAssumed Answered

How to explain seemingly random "pockets" of high error in simulation results?

Question asked by Daniel Landgraf on Jan 14, 2020
Latest reply on Jan 15, 2020 by Jay Seaglar

Hello all,

 

I am looking at a very simply weldment structure. The material is set to the Solidworks ASTM A36. I saved the weldment as a simplified part with one body. 

 

The part is constrained with Fixed Geometry to simulate its welded connection to a larger, relatively rigid body on either side.  

 

The applied loads are standard gravity and a remote load with rigid connections. The load produces expected torsion in part. The displacement plot passes the sanity check and the stress plot looks okay; however, the Error plot looks very strange. I have read several different thoughts on the use and application of the Error plot and the conclusion I came to is that the results should indicate 5% or less for acceptable confidence. 

 

Irrespective of this, I simply cannot wrap my head around the plot itself in this case. I have been including this plot in my simulations for some time now and have never come across something like this:

 

Does anyone or has anyone experienced a similar result? What would an adequate explanation look like?

Outcomes