Bed Leveling Challenges Duet 2 Wifi + BLTouch
-
Changes made, fingers crossed for some good results. My intention is that g32 only gets run once in a very blue moon, maybe once a week/month or if the printer is ever moved.
-
@Phaedrux I think I've found the problem. For whatever reason, the BLTouch is causing some flex in the bed platform as it touches the bed in the x min and max. Is there a way to decrease the amount of force the BLTouch exhibits on the bed through the M558 or G31 commands?
-
@elli said in Bed Leveling Challenges Duet 2 Wifi + BLTouch:
@Phaedrux I think I've found the problem. For whatever reason, the BLTouch is causing some flex in the bed platform as it touches the bed in the x min and max. Is there a way to decrease the amount of force the BLTouch exhibits on the bed through the M558 or G31 commands?
It takes very little force to cause the pin on the BLTouch to retract.
If probing the bed is causing the bed to flex there is something else wrong with your setup.
Frederick
-
This post is deleted! -
This post is deleted! -
Why do you think you need to set the Z=0 datum by homing a point on the grid
as opposed to the center of the bed?Thanks.
Frederick
-
@fcwilt It has been my understanding that homing to a probe point (when using mesh compensation) has been the suggested guidance - now of course, saying that, I don't have any links on hand to back that up (although if you google "go to first bed probe point and home Z" you'll find of loads of results that are likely from the configurator)
If one were to home explicitly to the center of the bed, that homing point could be in between (and not coincident with) two probe points. I guess in theory when full bed probing is done we should have knowledge of our Z=0 datum (even if the physical location - center of the bed for example - not being probed during the height map creation). Perhaps I'm only spinning myself around in circles...
-
@sebkritikel said in Bed Leveling Challenges Duet 2 Wifi + BLTouch:
(although if you google "go to first bed probe point and home Z" you'll find of loads of results that are likely from the configurator)
IMO, the configurator should not be considered a reference on the best way to do things. I've never used it but have seen posts showing it generating some "wonky" code.
If one were to home explicitly to the center of the bed, that homing point could be in between (and not coincident with) two probe points. I guess in theory when full bed probing is done we should have knowledge of our Z=0 datum (even if the physical location - center of the bed for example - not being probed during the height map creation).
To my way of thinking the Z=0 datum point is a "characteristic" of the physical printer independent of the process of creating a heightmap.
From time to time I generate height maps with differing numbers of points. If I was required to always use a grid point the Z=0 datum would likely not be in a fixed location.
I have always used the center of the build area for setting the Z=0 datum and have never found a reason to do something different.
-
@sebkritikel said in Bed Leveling Challenges Duet 2 Wifi + BLTouch:
It has been my understanding that homing to a probe point (when using mesh compensation) has been the suggested guidance - now of course, saying that, I don't have any links on hand to back that up (although if you google "go to first bed probe point and home Z" you'll find of loads of results that are likely from the configurator)
Yes that has been recommended, but it isn't necessarily critical to getting good results. The benefit of having a matching point is that it removes the need for relying on correct interpolation between points. So on a very sparse grid and a bed that has some waves it might matter. On a flat bed with a higher density of points it matters less.
It also assumes that the bed is relatively flat to begin with, and properly manually leveled, that the probe is relatively accurate and repeatable and mounted correctly without tilt. That the XYZ offset between nozzle and probe is accurate. That the Z0 is set with G30 before mesh creation rather than an endstop (though that can work too if the Z offset is correct.). That the probing speed is slow enough to allow for accurate stopping positioning. That the Z axis resolution is fine enough to stop at a stable point of travel. Etc
All those things matter much more than using the G30 on a mesh probe point.
-
@phaedrux Closing the loop on this. Thank you for your advice and guidance on this. In the end, the final piece was finding that my right z axis belt was off center and grinding hard on the sides of the top bearing guides. I can see significant wear on the belt and will need to replace it soon, but after a morning of balancing the set screws, the printer is now operating the best I've ever seen it. I found lots of little mistakes and learned a lot about the duet framework and my printer in the process. Thank y'all so much!
-
