Rostock Max V3 Z-Probe Issues

Phishing Fish

@dc42 The probe triggers when running G30, but not G32. I bumped up the probing speed to 1200 however, it still is not stopping. It is a misconfiguration in my bed.g?

dc42

@Phishing-Fish thanks, that's important information. Two things:

Delete the M665 command in bed.g. You should have only the one in config.g, also there may be one in config-override.g if you are using that file.
When you run G32, after homing does the head move rapidly to a position over the bed and then more slowly down before crashing into the bed? Or does it move and crash in a single movement? If in a single movement, then it's likely that your M665 H parameter is too high. See https://docs.duet3d.com/en/User_manual/Tuning/Delta_calibration#running-auto-calibration-for-the-first-time and the following section.

Phishing Fish

@dc42 I commented out the M665 command in bed.g. I don't have a config-override.g, so the one in config.g should be the only M665 command.

When I run G32, it runs straight down into the board, afterward I quickly cut power to the printer to avoid damaging it. I followed the directions on the link you set, with the M665 H parameter being the height of the nozzle away from the board when all three end-stops are triggered. This value was very close to 400mm when I measured, and this lines up with the default value of 400mm when I got this config from SeeMeCNC's site.

With the removed M665 command from bed.g and no change to the M665 H value, as I measured it at 400mm, I tested again and unfortunately still no luck. Whenever the Z-probe is triggered (anything above the value of 0 it seems) a blue light comes on. I see this light come on when the nozzle collides with the board, however, the printer does not stop.

Again, thank you for the continued help

dc42

@Phishing-Fish please follow the suggestion in the link I posted to increased the H (dive height) parameter in the M558 command in config.g temporarily.

Phishing Fish

@dc42 Changed it from 400 to 300, still crashed. Changed from 300 to 200 and it did stop, about an inch away from the bed, never went down and touched it. From there drove itself into one of the vertical beams and I stopped it. I'm assuming this means my M665 H value is incorrect, however in the guide it says to measure the distance from the nozzle to the bed in mm. I get right near 400. Is there something else I should be doing?

dc42

@Phishing-Fish it looks to me that you ignored my previous reply. Also if reducing the M665 H parameter didn't help then I suspect that you still have another M665 command in one of your files.

Phishing Fish

@dc42 Apologies for the confusion on my end. I did increase my M558 H value but forgot to mention it. I bumped it up from 5 to 30, as the guide suggested, and still didn't have any luck. I bumped it up to 50 just to see if maybe it needed more space, still crashed. As for reducing my M665 H value, I double-checked both my config.g and bed.g files and there is only one line that has M665. Is there a reason that changing the M665 H value would prevent it from hitting the board, and is there a way to determine what value it should be? Again, thank you for the continued help, it is greatly appreciated!

dc42

@Phishing-Fish you can run M665 with no parameters to see what the current value is.

Also check what coordinates DWC reports after homing. X and Y should be zero, and Z should be about 5mm lower than the current M665 H parameter, if you are using our standard homedelta.g file.

Phishing Fish

@dc42 M665 results: Diagonals 291.500:291.500:291.500, delta radius 142.500, homed height 200.000, bed radius 145.0, X 0.000°, Y 0.000°, Z 0.000°.

After homing, DWC reports X and Y at zero, Z at 195. I would make my M665 H value 190 then, and see if it stops before crashing? Also, I'm unsure if I'm using the stand homedelta.g file, so I've attached it here.

; homedelta.g
G91                          ; relative positioning
G1 H1 X750 Y750 Z750 F5000   ; move all towers to the high end stopping at the end-stops (first pass)
G1 H2 X-5 Y-5 Z-5 F5000      ; go down a few mm
G1 H1 X10 Y10 Z10 F500       ; move all towers up once more (second pass)
G1 Z-5 F1000                 ; move down a few mm so that the nozzle can be centered
G90                          ; absolute positioning

dc42

@Phishing-Fish are you certain that you have the correct steps/mm set? For example, if you home the printer and then command Z to reduce by 25mm, does the effector drop 25mm?

Phishing Fish

@dc42 This looks like a potential issue. I brought the nozzle down to the bed and pressed the Z+25 button. Measuring it, it looks like it went up nearly 65mm. How do I go about corrected/calibrating this?

Phaedrux

@Phishing-Fish said in Rostock Max V3 Z-Probe Issues:

How do I go about corrected/calibrating this?

That's a pretty substantial difference. What are the motors you're using? 1.8 or 0.9 degree? Belt pitch? Idler tooth count? Microstepping?

Phishing Fish

@Phaedrux Motors are Automation Technology KL17H248-15-4A. Looking at what I can find online, I couldn't find anything related to 1.8 or 0.9 degree, but I did find a step angle of 1.8° ± 5%. For belt pitch, it's 2mm. Belts can be found here: https://www.seemecnc.com/products/gt2-timing-belt. Looking at the build guide for this printer, it seems like the idlers do not have teeth and looking at my printer don't see any. For micro stepping, this is what I have in my config.g: M350 X16 Y16 Z16 E16:16 I1

dc42

@Phishing-Fish your M92 command shows 200 steps/mm. With x16 microstepping, that would be correct if the motors are 0.9 degrees and the motor pulleys have 16 teeth. If the motors are 1.8deg then it would be 100 steps/mm with 16 tooth pulleys, or 80 steps/mm with 20 tooth pulleys.

If the correct steps/mm for your machine is actually 80 instead of 200, that would cause a commanded 25mm movement to move 62.5mm.

A Former User

This post is deleted!

Phishing Fish

@dc42 That worked! Changing the steps/mm to 80 in config.g and now the auto-calibration stops when it hits the board. Thinking back on it now, I should have known the motors were running faster than they should have, but it never crossed my mind. To confirm, I should run the auto-calibration, then M500 to save the results in override-config.g, and then put M501 at the end of config.g, so that override-config.g is used?

Thank you both so much for the continued help!

Phaedrux

@Phishing-Fish said in Rostock Max V3 Z-Probe Issues:

To confirm, I should run the auto-calibration, then M500 to save the results in override-config.g, and then put M501 at the end of config.g, so that override-config.g is used?

yes