I did some back of the envelope calculations of the latest experiment that seems to imply that most of the 'tapering off' could simply be inertia since the dire

I did some back of the envelope calculations of the latest experiment that seems to imply that most of the 'tapering off' could simply be inertia since the direct neasurement was deflection distance, and the apparatus is very heavy compared to the force measured. A part of it definitely looks like a heating related effect because of the nonlinearity, but at the same time the implied cooling rate seems to fast for moderately hot equipment in a vacuum. At this stage I don't think you can claim the measurements are obviously wrong without actually calculating heat loss rates and how heat expansion would affect measurement with the given equipment asymmetries.