Auto_Bed_Leveling: Enhanced G29 command

[AbuMaia]

Thanks. It compiled just fine, but the Y axis still moves when the Z axis gets in position to extend the probe.

[Roxy]

That video you posted up above... What was the exact sequence of commands that you used? Did you do a G30 after a G28? Would it be possible for you to make a quick video of it doing a G28 followed by a G29 n 3 ? Depending upon where the nozzle is.... And where it is going to try to probe, it will move in both the X & Y axis to set up the probe over the correct location. Is it possible, that is what we are seeing? (Because you have the Y_OFFSET at -7.7mm)

UPDATE: I re-watched the video a couple of times. If the first movement is bringing the X & Y to zero... In order to probe at (0,0) it does need to move the nozzle to ( -29.6, 7.7) In reality, it isn't going to probe (0,0). It is going to move the probe to (40-29.6, 20+7.7) because of your Configuration.h settings. When the probe measures its first point, is it close to (x,y) (10.4, 27.7) ?

And one more question: Isn't your origin in the Back Right of your machine? You have it defined as Front Left in the Marlin_Main.cpp file.
That won't affect anything right now, but it will invert your Bed Topology Map and make it harder to figure out how anomalies with your print bed.

[AbuMaia]

For the video I posted earlier, all I did was send a G28, nothing else. The Y axis movement shown is not preparatory to a bed-levelling probe.

Yes, the LCD screen reports the first bed-levelling probe point is 10, 27. However, why should it move to probe for Z 0? And since it does move only Y to probe Z 0, why does it not report Y position is 10, instead of keeping it at 0?

As for back-right, front-left, it is the back-right of the machine from my viewpoint, but the printer prints as if it were the front-left. Everything I print is printed facing the back of the printer.

The video you requested is uploading. It will be available at http://youtu.be/oki4cO-dMeo. It demonstrates, during the G29 n3, the main problem with this erroneous Y movement during homing. The LCD reports the Y position as 187, but it's really 197, right on the edge of the glass.

[Roxy]

Tell you what.... How about we turn on 'Safe Bed Leveling' until we get this figured out. The reason is, I've always had that enabled and I'm more familiar with how things move in that mode. Once we figure out what is going on, we can turn the 'Safe Bed Leveling' off. Would that work for you?

And we don't need to worry about it now... But I suspect you want to set the Topology Map to having the Origin at Back Right. The reason is you can just visually map what it prints out on to the print bed and know which corner is high or low or where you have high or low spots on your bed.

[AbuMaia]

Safe Homing did seem to get rid of the bad Y movement. During the G29 n3, the last three probe points are no longer on the edge of the glass.

If I now try to use Pronterface to home X alone, then Y alone, then Z alone, it won't home Z, it says "Z probe out. bed". Perhaps that's the cause of the earlier Y movement during Z home?

[Roxy]

Safe Homing did seem to get rid of the bad Y movement. During the G29 n3, the last three probe points are no longer on the edge of the glass.

If I now try to use Pronterface to home X alone, then Y alone, then Z alone, it won't home Z, it says "Z probe out. bed". Perhaps that's the cause of the earlier Y movement during Z home?

Did you mean "Safe Homing did -NOT- get rid of the bad Y movement."

That is a valuable clue about Pronterface. What happens if you do a G28 and then a G29 ?

The problem is you have the probe in negative space. So if you home the X & Y you need to actually move back into positive space some amount before it can home the Z-Axis. But a G28 with Safe Bed Leveling will move the probe to the center of the bed.

[AbuMaia]

No, I meant what I said. There is no longer the 10mm offset in the Y axis with safe homing enabled.

[Roxy]

OK... Do you want to go back to unsafe bed leveling? If so, we can dig in and figure out where things are going sideways. With this new information, it shouldn't be too hard to figure out. But personally, I think the Safe Bed Leveling is worth having turned on.

[AbuMaia]

I'm going to try an experiment: turn off safe homing, and flip the sign on the Y offset, and see if that's the problem.

Edit: It didn't work. There was still the 10mm Y move during Z lift. So it's not moving to put the probe over the bed for homing.

[Roxy]

That's good thinking. But if that doesn't do the 'right' thing... Consider cutting the Y-OFFSET in half and/or double it. That would prove it is being caused by the Y-OFFSET.