@o_lampe if you just want to know whether the filament is loaded, you can try using stall detection on the extruder for that.

I tested and had no consistent results with a structured PEI coated steel sheet. Moving the head just 1or 2mm sideways produced deviations of up to 0.2mm in either direction. I am back to BL-Touch with more consistent overall results whereas the IR-probe was more consistent on the same spot.

@andy Sorry you have a broken IR probe. We are happy to replace the board under warranty, but also happy if it's just a soldering issue and you want to fix it yourself.

If you would like to replace it under warranty, please send an email to warranty@duet3d.com and CC your reseller. Include a link to this forum thread and the details of your original purchase. You'll receive a reply with a form to fill out.

Ian

@dc42

Excellent. I actually did more digging around last night and found one of your posts in another forum from many years ago recommending this. I will try it and report back. Your sensor is absolutely the best one out there in my opinion, so I will be happy to get it working on this machine.

UPDATE: That did the trick David. It is now booting in digital mode. Thank you.

Problem solved under warranty, it works fine now.

Thanks Kees

@Lamerhouse yes you can. Just set them up as separate probes

@Phaedrux
This printer has been causing me a lot of quality problems since Dec of 2019 where I sort of let it rest (small single object TPU/TPE jobs printed fine) for a year after Covid happened before putting more (but random) time in to troubleshooting it.
Over the last few years replaced all the major parts including this last week replaced the Duet2 with a Duet 3 and 1lc tool board. Once I got everything for the most part working (ie printer could print), I started working on Auto Bed Leveling again, and was again having trouble that reminded me about this post. I reread your question a few times (asking if I run leveling multiple times in a row does it get worse) thinking it was a clue. But as I never reran auto bed leveling more then once in a row, I didn't see this happen when I was testing, as the bed would get so far out of level that the BL Touch would give up or already be triggered when it went to a location to probe for the bed.
Turns out this was the problem the printer was having after I moved to the Duet 3. Every time I ran it, bed leveling got worse in the same way. Focusing on what could cause this, I realized that the Z motors were not associated with the correct locations.

I am not sure on what caused the problems with this printer on the Duet 2, but think the board is OK and it was just some calibration that got out of whack when trying to print a handful of different filaments that ended up being bad... And wasn't able to figure out what I did. Going to the Duet 3 forced me to redo everything, gave me an Accelerometer for Input Shaping and all the other goodies I get when using an SBC. So I am not complaining.

And as always. Thanks for the help!

you could try roughing up the G10 surface with some 80 grit sand paper. that should result in more uniform top surface reflection.

Can you share how you have it wired and configured?

@ageis that blog entry hasn't been updated for many years. Documentation on connecting the IR probe (and other TYPES F z PROBE) is at https://docs.duet3d.com/en/User_manual/Connecting_hardware/Z_probe_connecting.

You don't need zprobe.mod for the IR sensor, it's only needed for the dumb modulated IR sensors supplied with later Ormerod and Huxley kits supplied by RepRapPro.

@greatlakes3dp you've made the mistake of treating the Z probe as an endstop. Remove the M574 Z line in your config.g file, then read https://docs.duet3d.com/en/User_manual/Machine_configuration/Configuration_cartesian#homing-z (in particular the bit about homing Z using a Z probe) to see if you need to change your homez.g and homeall.g files.

@ageis it has to be the underside because you want the top side to be PEI for your print to adhere to.

The reason for painting it is that PEI is transparent to IR so the sensor sees reflections from both the top and the bottom surfaces, which makes it less accurate. Painting the underside reduces the reflection from the bottom surface.

@ageis bear in mind that the IR LEDs that we use in the sensor are quite highly focussed, unlike the IR LEDs in remote controls. So you will to have your camera in line with the beam to see them.

@ageis can your camera see light from just one of the IR LEDs on the board, or from neither of them?

@droftarts I will add the G90 and change the pressure advance setting to the correct tool. The reason I am not using the 0 drive is that ithe driver was fried by my bad extruder motor. I had to switch drives once I got the motor replaced. Thanks for all the help

@ageis said in Ir probe all of a sudden triggering early:

Today when I homed the y axis it just trigged immediately.

Do you mean Z? That's what the probe is set up for. And it sounds like the probe is responding correctly. If it's Y that isn't homing, check the endstop isn't disconnected, then test the endstop is working correctly when you press it, by sending M119 before you press it, and again while you are pressing it.

Ian

@dc42 said in Idex second Z probe on the U axis:

@Proschi78 are the axes just shown incorrectly in the height map viewer, and the actual compensation applied is correct?
Please show the GCode commands that you are using to generate the height map.

I only use G29 K1

i have a Macro with the M557 U66:300 Y2:230
I start first, and then I give the G29 K1 command.
I haven't started printing yet to see if it's just rotated in the map.
I think it's better to have both on the same axis, unfortunately I don't have space on the x-axis for the additional BL touch.
Since I like to print on structured Pei, I need the Bl Touch.