Quote from: kml on 06/22/2015 09:10 pm Most likely there is an RFI problem with the scale. I'm going to attempt to shield the scale with the unit suspended above it to prove that it is RFI.



I have to say that these are very intriguing results! While we are all waiting to see if you can eliminate RFI interference on your scale as the cause for these results, I would like to point out a couple of notable things:



1). The performance with and without dielectrics reflects Shawyer's and Yang's experience.



2). Much more interesting from my perspective are the declining lows in your first experiment with the dielectrics. This is a very similar pattern to the Eagleworks tests With a dielectric insert. It was speculated at the time that this might have been due to out gassing but I can't see that happening with a ceramic dielectric !!



So what the heck is happening with the dielectrics?



There is a more subtle pattern of declining lows without the dielectrics, but it is more pronounced with them present.



So either there is some type of ongoing electrical (electrostatic?) interference with the digital scale once the power is off, significant out gassing from somewhere, or we are looking some very interesting theoretical issues here folks - and evanescent waves are just not going to cut it...



Edit: After checking the scale, it is apparent that the unit is getting lighter and so we have to add ballooning to the possible effects, with the dielectrics thermal mass sustaining the effect.



Don't read too much into the overall declining slope. It takes a while for the scale to fully stabilize after placing the unit on it. The faster I start the test the steeper the slope is, usually. I have learned to wait longer before starting as you can see from the later test #'s.Keep in mind the "suspended" test is without any contact between the test unit and the scale. There should be absolutely no force registered in this configuration. The weight on the scale is from several ceramic flooring tiles used to simulate the weight of the unit. I did test with very little weight on the scale with the unit suspended above it and there was no change in weight indicated:This indicates that the RFI induced error may be related to the dynamic correction system used to offset heavy loads.:Also, the tests done in the "down" orientation show much less force, though still in the "weighs less" direction. This may be due to the better RF sealing on the fixed end which is down in the "down" orientation.