How about using a rotary encoder on the screw and a linear encoder running parallel to the screw fastened to the nut somehow? Feed the encoder outputs to different axii on a DRO readout and compare the readouts (DRO head or CNC software). Record the differences and positions of differences.
Even if the linear encoder was not as long as the screw you could reposition it along the screw to map the screw in sections.
This would not be as accurate as laser interferometry, but it should be fairly accurate with use of accurate encoders.
What do you think?
Alan |