Discrepancy in Jog distance (Resolved)

[Insect]

Hi,
I am doing some screw compensation, and driving the machine 25mm at a time using macro 1,2,3,4 keys.

When I use the script below which orders it to move 25mm, it actually displays a DRO move of 25.0019 each time. This requires some manual compensation on my end, as the effect is cumulative. All of these macro scripts are doing this.

Also when going to the WCS origin (a common macro I normally run) it may show 0.0000, 0.0006, 0.0000 , etc, it shows it landed a submicron level away from the desired coordinate. This doesn't bother me much as I am chasing tens of microns, not fractions of microns, sure, but I feel it may or may not be related.

Any idea what is causing this?

The program below is written to do the last 0.5mm slow to get repeatable non-jarring moves for the purpose of the calibration, nonetheless its still 25mm.

Thanks,
Eric

G91 G21
G1 X-49.5 F2000.00
X-0.5 F200.00

[suntravel]

Related reading:

https://centroidcncforum.com/viewtopic.php?f=63&t=1801

Your setup is 25mm / Rev with 6400 steps = 0.00390625 mm / Step

No way to make the CNC move from 25mm to 25.001 mm

Your deviation is caused by rounding from the setup operating with a to coarse resolution

Uwe

[Insect]

You are correct.

I fancy a way to call some math function into the gcode, identifying final position, then rounding it, without overstepping and inducing backlash, i.e. always staying in the lead direction. But I'm not certain it can be done. It's OK. Thanks