Bill93, post: 454064, member: 87 wrote: The manufacturer chose to correct this to a degree that depended on the instrument specification.
This was the 4x encoder reading ??
My daily driver is 2", for control we have a 1/2" instrument, the back up gun is 3" and there is a 1" sitting around as well.
To have any meaning the question needs to be addressed in a holistic manner, taking into considerations the conditions, skill sets and deliverables you are working with and toward.
5" is fine for what we do.
Bill93, post: 454064, member: 87 wrote: If I recall correctly, Conrad was working under very well controlled conditions and found that a major source of error in his instruments was systematic, in that the error plots showed a sinusoidal dependence on the direction pointed.
this is correct.
The manufacturer chose to correct this to a degree that depended on the instrument specification.
that was my earlier hypothesis. i would now guess that the hardware is pure, and the 'error' is created on-board, by software or firmware.
Under those conditions Gaussian statistics don't apply.
If random errors dominate for your instrument or for your procedures, you will see results more like the Gaussian model.
without rigorous error mapping how would one ever find out? I don't think it's possible. when instrument specifications are listed as standard deviations i think we just default to thinking of normal distributions. but you can calculate a standard deviation for any set of data no matter how non-normal it is. the specs tell us nothing about the distribution. that is our assumption.
R.J. Schneider, post: 454091, member: 409 wrote: This was the 4x encoder reading ??
no, 4 read heads are employed in leica's top-of-the-line models. according to their literature it's to reduce the noise associated with the graduation readings (more sampling = better) and to eliminate, minor pi-periodic errors. the pi-period errors can exist in a system with 2 diametric read heads and i am sure they persist in the 'lesser' models that i've tested.