M558 ignores quantity of probe hits
I'm running 2.05.1 firmware with a DC-42 optic Z-probe. Each probe point is hit 2 times only regardless of the value set on the "A" parameter. Here is my config.g probe code. And, yes, I did a re-load.
M558 P1 A4 H2 F300 R0.5 S0.02 T9000
Danal last edited by
S0.02 says "when you get another reading within +/- 0.2, move on".
If you really want to get a tight tolerance, add another 0 to your S value. I use S0.003 and A10 with a BLTouch and it probes twice most of the time, and occasionally 4 or 5. You can definitely get better than 0.02 out of an IR probe.
Thanks - mommy always told me to read the fine print.
.. However, what prompted this fiddling with probe quantity is the fact that I'm no longer getting satisfactory endstop convergence after years of proper performance.
.. Observing the DWC Z-probe sensor trigger values the readings jump around a lot (orange mostly mixed in with some red ones). Also the occasional "Z-Probe readings not consistent" message pops up.
.. I reset the A parameter to 5 and S to 0.005. Usually it gives 2 probe hits but occasionally jumps to 3, sometimes 4.
.. Does this smell like a faulty IR probe or perhaps the optics kicked out of proper alignment? The printer (delta) itself is solid and even now produces very accurate parts - no layer shifts or other anomalies.
Try giving the IR probe a cleaning?
Has anything loosened over time?
If you try forcing it to a tighter tolerance with A10 S0.003 do you get better results (even if it requires probing a few more times)?
I'm not really sure how age will affect performance. @dc42 might have a better idea?
No joy - same results. I'm thinking I inadvertently knocked things out of optic alignment. So, being tired of pounding sand, I'm going to order a new sensor and see what happens.
My one criticism of the design is there is no protection for emitter-sensor geometry. I'll give that a go when the new one arrives.
Thanks for your help.
@balrags said in M558 ignores quantity of probe hits:
My one criticism of the design is there is no protection for emitter-sensor geometry.
I'm not super familiar with the probe, so I'm not sure what you mean by that. Can you explain?
It will be interesting to see how the new probe works out.
.. An IR emitter (red) bounces a beam off the print bed in a targeted geometrical relationship toward two differential IR sensors (green & blue). A certain amount of precision of aiming ( over to DC42 ) is required. These emitter/detectors are not protected from external forces which might deviate the correct optic paths and re-positioning is, in my experience pretty difficult.
.. I'll post here when the new probe arrives.
![alt text]( image url)
Replacement probe arrived and installed. Works perfectly and converged at 0.003mm - good enough for government work. I must have somehow bumped or otherwise disturbed the optical alignment.
AND the good news is the latest I-R probes now have surface mount emitter/detector optics which are impossible to move out of alignment. Thanks DC42 or Tony or whoever.