Piezo z-probe on Duet Ethernet bahaving stangely

I'm just beginning to set up the piezo probe which I received some time ago from Simon, so I'm sure that there will be a few questions on the way over the next couple of weeks.

Due to a number of design considerations I'm currently planning to utilize an underbed piezo setup. Sensitivity at this point looks great, quite a bit better in fact than what I was seeing with the hotend piezo. The first iteration with 27mm piezos was likely too sensitive so I've redesigned the setup around 20mm piezos.


After a number of preliminary moves the nozzle begins a probe near the edge at about 7 o'clock but lifts again a few mm before it touches the bed. (My homed height places the nozzle just in contact with the bed and I'm using a -0.1mm Z offset as per Simon's instructions.)

Here's a video, as can be seen things appear to be working except for the probes nearest the camera. Also evident is that the failed probes are beginning from a greater z-height than are the successful probes.
I'm also curious, what are all of those slow transverse moves over on the left?


Not wishing to get ahead of myself ... but as of now seeing at least some good probes, I'm impressed! This will be my first time having a z-probe.

I'll be sure to do a complete write-up of the details once things are up and running, along with another describing the piezo end stops which are also waiting in the wings.


Thanks!

Edited 2 time(s). Last edit at 05/29/2018 01:32PM by papilio.

I'm not familiar with the Duet and its RRF firmware so I'll wait for someone else to chime in on any firmware related issues.

On the piezo front, it's not easy to tell from a video it looks as though there isn't even a flicker of a trigger over on the left side of the bed while on the other side you're getting nice clear probes and triggers. Is this true in person, can you see any flicker from the blue led when it's probing on the left?

Very neat printer BTW, I like it.

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

Hi Idris, and thank you!
The video is depicting the behavior quite accurately, there's no flickering of the blue LED on those probes at the 7 o'clock position.

The failed probes appear to be responsible for generating "Error: Z probe already triggered at start of probing move".

"Error: Z probe already triggered at start of probing move"

This suggests that you are getting false triggering. A couple of things to try;

Double check your wiring and connections. I had a very frustrating day chasing a probing problem that was in fact caused by a loose connection, only showed up when probing one side of the bed.

Reduce sensitivity slightly, are you using a Universal board or a PP20 board?

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

I have a Universal board.

Yes false triggering was my initial thought as well, even though the blue LED seems almost never to flicker now that I've tweaked the setup. But I've tried playing with sensitivity with no improvement so far. Just to deepen the mystery, that error message is no longer appearing.

Here's a new video with a different bed.g configuration showing things a bit more clearly -- all positions except the one seem to be probing properly. And there's an obvious difference, no idea what's going on but I'm seeing all successful probes beginning from a height of around 5mm while that one failed probe position begins at 40mm and then seems to give up before hitting the bed. Tapping the bed so as to generate a signal during that probe brings no response.

If it is indeed a wiring or connection fault I'm not finding anything, but TBH I don't have the experience to know just what needs to be checked and I may have missed something. Any suggestions based on the case you mentioned?

Thanks for the help!

If I remember correctly my issue was a single loose connection in the 3pin cable making intermittent contact, this was a nozzle sensor and when the effector was on one side it pulled on the wire very slightly and broke the connection. It was very consistent however which made it harder to pin down.

I can't work out what on earth its doing in that video, half the movements and probes seem unnecessary to me but then I'm not familiar with how RRF does a delta calibration. Have you tried moving the head to various positions around the perimeter of the bed and manually calling a Z-probe?

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

I've never seen such a weird probing behaviour with RRF before. Did this printer work before you installed the z-probe?
How does the z-probe config line look like? How did you generate the bed.g file?

Those are indeed very odd probing movements. You didn't say whether you are attempting G31 auto calibration or G29 mesh bed probing. If it's G31, please share the contents of your bed.g file. Also please share your config.g file and firmware version.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
Moriquendi
I can't work out what on earth its doing in that video, half the movements and probes seem unnecessary to me but then I'm not familiar with how RRF does a delta calibration. Have you tried moving the head to various positions around the perimeter of the bed and manually calling a Z-probe?
Idris

I know, it all looks mighty odd to me as well. Far as I can tell at this point, those repeated slow transverse moves on the left appear somehow related to the routine trying again and again to successfully probe that particular XY coordinate.

The failed probe occurs at X-73.6 Y-42.5, so is this the correct command?
G30 P0 X-73.6 Y-42.5 Z-99999


This causes the moves as shown in this video. That first good probe seen is not at these coordinates, the second probe attempt which is at those coordinates displays the same behavior as before.

Edited 1 time(s). Last edit at 05/30/2018 10:39AM by papilio.

Quote
o_lampe
I've never seen such a weird probing behaviour with RRF before. Did this printer work before you installed the z-probe?
How does the z-probe config line look like? How did you generate the bed.g file?

I know! So strange that I was expecting a quick solution from somebody more experienced, but better minds than mine are still working on it.
The printer is not yet fully commissioned.

Here's my config.g which contains the z-probe config, also attached is my config.g (not yet fully configured). The bed.g routine depicted in the latter video was generated by [escher3d.com] .

Edited 2 time(s). Last edit at 05/30/2018 10:49AM by papilio.

Quote
dc42
Those are indeed very odd probing movements. You didn't say whether you are attempting G31 auto calibration or G29 mesh bed probing. If it's G31, please share the contents of your bed.g file. Also please share your config.g file and firmware version.

Thank you for helping with this David, I've posted the requested files on my post on the Duet3D forum for you. I've been using auto calibration.

Edited 1 time(s). Last edit at 05/30/2018 10:23AM by papilio.

Can I just check you do not have deployprobe.g or stowprobe.g (or similar) files on your sd card on your Duet?

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
Can I just check you do not have deployprobe.g or stowprobe.g (or similar) files on your sd card on your Duet?

Hi Simon, thanks for the response. Yes deployprobe.g and stowprobe.g remain on the card. Remove them?

Edited 1 time(s). Last edit at 05/30/2018 10:42AM by papilio.

You solved it Simon, many thanks!!

[youtu.be]

Edited 1 time(s). Last edit at 06/01/2018 09:13AM by papilio.

No problem. they became "run on default" with a certain RRF version 1.19 I think, unless they are removed. My kossel XL was making huge moves when I asked it to probe, and David quickly suggested I remove these files. It was in the release notes but for that version.

Edited 1 time(s). Last edit at 05/30/2018 11:25AM by DjDemonD.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions