DRO Zero position (a bit academic I know)

Advert

New Replies

Report Bug

Help/FAQs

DRO Zero position (a bit academic I know)

This topic has 4 replies, 4 voices, and was last updated 30 August 2025 at 14:30 by Lee Reynolds 1.

Viewing 5 posts - 1 through 5 (of 5 total)

Author

Posts
30 August 2025 at 11:38 #814449
Lee Reynolds 1
Participant
@leereynolds1
Hi all,

This is a rather academic and maybe pointless discussion but I’m sure someone knows the answer.

You find your position on the DRO and press zero as you need to return to it.

When you do wind whichever scale back to said zero position, there must be some error in your position.

I’ve often wondered if it is caused by the scale/reading head resolution, or if backlash still has a role for instance, or is there any other factor involved that can be used to mitigate any potential error.

Should you get the highest resolution scale you can afford, for instance.

Just wondering if anyone has any thoughts on, certainly in my case, a theoretical question.

Kind regards

Lee
Advert

30 August 2025 at 13:18 #814467
bernard towers
Participant
@bernardtowers37738
How have you arrived at that?
30 August 2025 at 13:18 #814469
Diogenes
Participant
@diogenes
On 30 August 2025 at 11:38 Lee Reynolds 1 Said:

Hi all,

.. …Just wondering if anyone has any thoughts on.. ..a theoretical question.. …

Kind regards

Lee

“Don’t lie awake at night worrying about it” – backlash plays no part because the scale/head should be ‘reading’ the actual relative positions of the ‘fixed part’ and the ‘moving part’, and not say, the position of the screw/handwheel.

Scant handfuls of microns are much less than the clearances between the moving parts of the machine, the effects of thermal expansion/contraction, or even the flexure of the main components, of most hobby machines.
30 August 2025 at 13:45 #814490
Robert Atkinson 2
Participant
@robertatkinson2
Any digital system inherently has a error of plus or minus one “count” or least significant bit. For the DRO scale example if we consider an optical scale and no other optical, mechanical or electrical errors or correction.there will be a directional error “backlash” in the order of half the resolution of the scale / reader. The sensor detects the edge of the scale mark. As you approach from one side it will trigger on one edge and in the other direction the other edge. So if the mark is 5 microns wide thare is likely to be a 4 micron error (it take a finite time to detect the edge).
This is a simplification as sensors use quadrature detection and other techinques to acheive their resolution but it is still a fundamental error.
Not an issue for most hobby machining applications but is in metrology and other high precision applications. I’ve been one of the designers of 3 axis “robotic” machines with 1 micron resolution and repeatability. To acheive that we always returned in the same direction. We also had to deal with bi-metallic thermal distortion.
30 August 2025 at 14:30 #814498
Lee Reynolds 1
Participant
@leereynolds1
Thank you gentlemen.

No I wasn’t lying awake worrying but it was just general interest.

It was obvious because anything with counters has an in between position no matter how many decimal places you are displaying. In some ways a slide rule is better as you can see the in between position (That statement shows my age and what I was brought up on).

But thank you Robert – that answer was what I was looking for. Not a Model engineers issue, probably, but it illustrates the reason, perhaps, why the DRO scales are usually to a much smaller resolution than I am capable of working to and, juggling it about to achieve zero is worse than approaching the position from the same direction that was used when setting it (if you can remember that is!!).

Regards

Lee
Author

Posts