I didnt think the calibration would be significantly affected by friction. Meaning that the rotating magnet generates a current in the non magnetic disc that attached to the needle. That also generates a torque on the disk. That torque is resisted by the springs on the needle assembly.

The faster the magnets spin the greater the EMF in the disk, the greater the TQ, the higher the needed reads. Since calibration does not rely on friction, I'm not sure how lube would change the reading by a factor of four.

I have never seen an explanation of how a analog speedometer is supposed to work, but that's what it looks like to me. Did I miss something?

Kevin