Why does simulation 2010 report very large normal free-body forces when using temperature external load on a body?  This is a completely fictitous force, but happens to be equivalent to the force required for to get the same displacement without a temperature change.

This is simple to recreate.  Just model a 1 inch cube.  Apply a temperature other then the zero stress temperature to the body (not the faces) and run.  Then, query the free body forces on the faces.  There is a very large normal force, which also can also affect two of the moment results as well depending on where you pick your point.

Any help on how to fix this is appreciated.

2010 SP0, 4, and 5 are all affected.  I suspect even the current build of 2011 is also affected since I've found no evidence of this being fixed, but I'd like some confirmation.