Jonathan,
It took a moment but I see why.
Using the model from the MyAnalysis directory and editing the pressure load to be on the outside only I get exactly the same results as your image.
This is the rotations causing different subtended distances Z&Y Cartesian to the radial movement and inconsistent measurement location.
I added edges to the model to and measured as follows :
I got the following :
max and min Cartesian measures required rather than more lines (symmetry, and I didn't know which way the radial measure would go). Seems consistent.
The max_disp_x = RADIAL_GLOBAL_MAX_Z as you found.
The max_disp_y & z nearly equal to each other and don't equal RADIAL_GLOBAL_maxR as you found. These are affected by rotations and their locations not consistent as follows:
All seems good.
To Amit's questions ...
This is a 270mm diameter steel ring with a rim that is 8mm thick and a substantial hub.
The external pressure of 0.8MPa is small and will do little. The displacements are tiny approaching un-measurable.
NB, the CTE in the material definition should be 16e-6 mm/mm/degC, it is in as '16'.
Model attached. I changed the units of the model to mmNS.
It would be interesting to know where this fails.
atb