forum.cncdrive.com

I have big toruble with probe operarions, if touch probe is tripped, it hits very hard on ballscrews, and throw servo fault ir some cases.
I think after firmware update, whole probe wizard now unusable. I try all range debounce values(0-255), wrote test codes, to be sure it here no problems with servo drives and other compenents. If probe not tripped, then machine movement is smooth every time, but if touch probe is tripped, then it hit very hard. I try reset emergency stop without decceleration, it looks very smooth, compared to G31 operation. Is here some possible bug?

Test code

G0 X0 Y0
G31 y-2 f50
G0 Y0
M30

We do not know about any bugs like that about probing.
But maybe you could explain the issue a bit more detailed, what do you mean by "hit very hard"?

I do quick servo plots, this positionin plot, without tripping probe.
https://ibb.co/c1HP1WT

Now plot with probe tripping.
https://ibb.co/6gcvyZY

Same code, mention above.

What does this plot show?
The position or the position plus the servo error or ... ?
Please explain.

Plot show position command from uc300eth (red line) , and blue line actual encoder possition (curves do not overlapp and offset because I jog axis when servo drive be into fault state) .
Let's see into plot 1 chronology(probe did not reach the target) .

G0 X0 Y0
G31 y-2 f50
G0 Y0
M30

1. Machine start moving with speed rate f50, from Y0.
2. Stops at 2mm with decceleration.
3. Move back to 0mm with rapid speed.

Now plot 2 (probe hit the target at ~1.5mm)
1. Start moving with speed rate f50, from 0mm towards 2mm.
2. At ~1.5mm probe hit the target.
3. How strange things happens. Machine moves TOWARD touch probe (at rapid speed?), until reach 2mm. It supposed to stop with decceleration imeadeatly at 1.5mm, instead of backoff, it increase speed and move toward touch probe.

This strange movement I notice years ago, but it accour rarely in older version. Maybe my custom textfields, or variables interfier with some uccnc internal stuff?

You didn't say what you are using for a probe. I assume its the tool bit making contact with a metal plate connected to the controller input. I chased a similar intermittent problem for a long time. 90% of the time the probe function worked fine. 10% of the time it would screw up. First fast probe cycle the tool would retract as normal but then on the slow probe cycle, would drive the tool bit into the touch off plate at fast speed. When I upgraded to servos I checked the tool bit ground path with an volt/ohm meter discovered I had a flaky ground readings. I was relying on the tool bit via the motor bearings and motor housing as the ground path. Big mistake. I changed to an alligator clip connected to the bit while probing hard wired to the controller ground never had an issue since.

Please see into graph 2. It's all about what controller do after probe trigger, here qoite from manual about G31 function - "If the probe
trips the coordinates at the trigger point are written into the variables #5061 to # 5066 (axes
coordinates X to C in order) and the axis stops with deceleration."

I downgrade firmware back to 2113. Issue goes away(but still exist in rare cases).

Which UCCNC version did you use which has this problem?

On version 2116 failure rate ~70-90%, on 2113 failure rate is about 3%, issue still exist.