I am comparing hand calculations to the results of a simple FEA model of a hollow cylinder subjected to centrifugal force, temperature and internal pressure for use in a press-fit calculator. Almost all calcs match the simulation except for a few strange results from Simulation that hopefully someone can explain. The interior surface of the cylinder is restrained in the axial direction only, and inertial relief is turned on. The goal is to determine the radial/tangential/axial stresses and increases in internal and external diameter.

The issues are as follows:

1) I have noticed that when probing for tangential strain values in simulation, the thermal strains are not included in the resulting value. However, if I use a displacement plot and measure the inner and outer diameters, I get the expected results. In order to calculate the correct diameter from the probed strain values, I need to first add the hand calculated thermal strain to the simulation strain value to get the correct result. Is this by design? Has anyone else experienced this?

2) When exposing the cylinder to a constant temperature load only, with the inner surface axially confined, but without centrifugal or pressure loads, I get a resultant stress in the tangential direction through the thickness of the cylinder. This doesn't make sense to me, as there is no restraint in the tangential direction (the diameters are free to expand). The axial stress, however, matches hand calculations for plane strain. Can anyone explain this tangential stress result?

In the uploaded model, the configuration L=2OD is set up to thermal only and probing the sectioned plots demonstrates the issues. Thanks