G28 Fine, then G29 ABL Probe hits bed

[csader]

Hey folks

Been sitting out for a while with a non-working printer. Had a meltdown and needed to replace some parts.

Getting around to replacing parts now and I'm trying to get ABL setup, but running into issues. I set the Probe Offsets in the firmware using these instructions: http://zennmaster.com/random-things/...-3-final-setup

Then I started following Clough42's instructions here: http://clough42.com/instructions/marlin-firmware-setup/

When I get to the Z Offset settings section, if I start at an offset of 2.00 in the Z-offset settings the LCD, then run G28/G29, I get a recommendation to put something like 15.49 in the Z Offset settings (I'm using a longer Z-probe arm than Clough's, so I thought this was normal). However, when I change the offset then run G28/G29 again, G28 goes fine, but G29 lowers the Z height too much and the probe hits the glass before it can extend all the way.

Any ideas? If I put the Z Offset back to 2.00 on the LCD, it stops doing it. But I'm assuming I'm going to have trouble printing if I leave it there (haven't tried yet), given that Clough's instructions are telling me to put in 15.49.

[Roxy]

Do you have your EEPROM turned on. If so, it sounds like you are getting some bad values and should clear it (set it to the default firmware settings).

[csader]

Do you have your EEPROM turned on. If so, it sounds like you are getting some bad values and should clear it (set it to the default firmware settings).

I do have EEPROM enabled. I'm using Clough's firmware from his website, which had it enabled by default.

By resetting it, do you mean running an M502? When I do that, and try running G28 and G29 again, I get the same result

[Roxy]

Yes, do a M502 (reset to default firmware settings) followed by a M500 (write current settings to EEPROM).

OK, it sounds like that is not at the root of the problem. I think I would start small and make sure all of my end stops and probes are reporting correctly. I would put the nozzle in the middle of the bed, up in the air and do an M119. Does it report all of the end stops and probes you think you have? Are the reported values correct? Are the values correct when you use your finger to trigger a particular end stop or probe?

Once that is all correct, I would do a G28. And then I would follow it with a M114 to ask where it thinks the nozzle is. Is the value reported reasonable? If not, we need to start tweaking the Z_OFFSET_FROM_EXTRUDER number until it is correct. (And each change of the firmware is going to need a M502 M500 pair)

[csader]

Does it report all of the end stops and probes you think you have? Are the reported values correct? Are the values correct when you use your finger to trigger a particular end stop or probe?

Yes, yes, and yes.

Once that is all correct, I would do a G28. And then I would follow it with a M114 to ask where it thinks the nozzle is. Is the value reported reasonable?

Yes, seems reasonable.

I went back into my firmware and re-ran the instructions from Zennmaster, followed by an M502 M500 pair, just in case. No luck.

I went back into my firmware and re-set the Z-offset to Clough's default, followed by an M502 M500 pair. No luck.

Using Clough's calculator, I'm still getting that I should set the Z-Offset to 15.1-ish, which seems reasonable given the extra length of the probe arm I have installed. But when I run a G28, it runs fine. When I then do a G29, it lowers the nozzle height too low for the probe arm to extend.

I'm clueless at this point.

[cperiod]

Your offset is 15? The default value to lift before extending the probe (edit: Z_RAISE_BEFORE_PROBING) is 10. After the G28 it knows where Z0 is, so it's probably going to be as low as possible before deploying the probe. Try changing that to 18 or so. But I'd definitely look at a shorter probe; the z axis is slow, and that extra offset sucks a lot of time.

[csader]

Your offset is 15? The default value to lift before extending the probe (edit: Z_RAISE_BEFORE_PROBING) is 10. After the G28 it knows where Z0 is, so it's probably going to be as low as possible before deploying the probe. Try changing that to 18 or so. But I'd definitely look at a shorter probe; the z axis is slow, and that extra offset sucks a lot of time.

You might just be a genius. Not near my printer at the moment to try, but this sounds like a winner. I'll report back for posterity.

Edit: Looking at the firmware right now, Z_RAISE_BEFORE_PROBING is currently set to 5.

[csader]

yep. Setting it to 18 kept it from lowering too far. Now after G28 (before sending a G29), it raises the nozzle a bunch, which is time consuming, but it's working and printing now, so I'm happy for the time being.

[cperiod]

Awesome. Now you can print a new probe....

[csader]

Awesome. Now you can print a new probe....

Not sure why I'd need to...this one is working fine, and I'm sure I can adjust the settings so it doesn't raise quite so much after the G28...

Something I'm missing?