Bed Leveling
I've been curious about the bed leveling on my new Core One. It prints fine, but I wanted to know how flat the system thinks my bed is, so I downloaded OctoPrint and installed the bed level visualizer plugin. According to that software, I've got a nearly 1.5mm drop between the back and the front of the bed. That surprises me because I know there's a separate stepper in the back vs the front. Why doesn't that stepper drive the back up 1.5mm to level the bed, and then use the bed leveling to do the intra-bed flatness calibration. Is there a calibration I was supposed to have done when building my printer?
RE: Bed Leveling
I've been curious about the bed leveling on my new Core One. It prints fine, but I wanted to know how flat the system thinks my bed is, so I downloaded OctoPrint and installed the bed level visualizer plugin. According to that software, I've got a nearly 1.5mm drop between the back and the front of the bed. That surprises me because I know there's a separate stepper in the back vs the front. Why doesn't that stepper drive the back up 1.5mm to level the bed, and then use the bed leveling to do the intra-bed flatness calibration. Is there a calibration I was supposed to have done when building my printer?
There is the Z-axis calibration but some have found it necessary to install shims to start with a level bed. There is one of these projects on Printables. I’ve run the same on mine but I haven’t bothered with it since I haven’t had any bad prints to date.
- Bob Marley
RE: Bed Leveling
so I downloaded OctoPrint and installed the bed level visualizer plugin
I've done that too, it's a good tool to tune the bed leveling. As @hyiger mentions, there is a bedlevel correction mod to address the issue you found. In my case, I used an 0.4mm correction on the front left and the neutral one on the front right. I think the need arises from the modular construction of the printer housing.
Beware of that visualizer, a body can get way absorbed in fine tuning. Ask me how I know...
RE: Bed Leveling
I've done that too, it's a good tool to tune the bed leveling. As @hyiger mentions, there is a bedlevel correction mod to address the issue you found. In my case, I used an 0.4mm correction on the front left and the neutral one on the front right.
Or you can use M3 screws and lock nuts in the three threads which Prusa conveniently provided in the base plate, one close to each Z spindle. I assume the threads were meant for adjustable end stops; not sure why they don't get populated in the factory or kit builds.
I think the need arises from the modular construction of the printer housing.
Or it can be due to different pitches in the three Z lead screws, as discussed here: https://forum.prusa3d.com/forum/prusa-core-one-hardware-firmware-and-software-help/skewed-xy-plane-z-rods-heatbed/
In this case the bed cannot be fully leveled via the end stops, since its tilt will change gradually as it moves upward. My Core One is affected by this, but asking Prusa for a replacement motor & lead screw is still on my to-do list. The deviation is significant for tall prints, where it can cause up to a mm of skew.
RE:
Or you can use M3 screws and lock nuts in the three threads
I tried that, found the other to be close enough and needed less fiddling.
Or it can be due to different pitches in the three Z lead screws
I spent a lot of time looking at that, it wasn't the case in my MK4S upgrade unit. The paper-trace check that @petulf suggested is an easy way to check. A new Core One shouldn't have that issue.
RE:
I looked at those three threaded holes in the base. They actually line up with holes in the z axis gantry, just the right size for a hex wrench to fit through. It's almost as if they meant for them to be adjustments for the stops. That said, they're actually far enough from the lead screws that there's a decent amount of flex in those arms if you use them for stops. I'm not sure why they wouldn't have put them closer to the screws if they were supposed to be a long term solution. Regardless, I ended up using the little plastic covers instead. They get you to within 0.2mm (even tighter if you print a few at different layer heights and use calipers to determine the true height). I'm happy now as the overall levelness is well within the localized variance of the plate. If I wanted anything tighter, I'm probably looking at shimming the individual plastic standoffs that attach the heated plate to the gantry.
One question: Are the three stepper Z axis motors completely electrically separate from one another or are they all tied to the same driver circuitry? In other words, might it be possible (with a software update) for the printer to do this leveling calibration by itself?
RE: Bed Leveling
Are the three stepper Z axis motors completely electrically separate from one another or are they all tied to the same driver circuitry?
They are driven by one driver circuit. It's an common approach.
RE:
When I purchased my MK4, it replaced an older printer I had been using Bed Level Visualizer with for several years, and I put a great deal of time and effort into bed leveling -- that was very necessary with that older printer. Given the ability of the MK4 to compensate for any irregularities and having seen the (somewhat surprising) data from Bed Level Visualizer -- I gave up using Bed Level Visualizer, and I've never looked back. (That MK4 is now a Core One.)
I did recently notice, while printing an especially wide print, that the Z-axis would seem to shift ever-so-slightly up and down as the nozzle travelled back-to-front -- I made a few manual adjustments that I hope will help, but the system seems to be able to accomodate a slight variation, and I'm not particularly concerned.
RE: Bed Leveling
I did recently notice, while printing an especially wide print, that the Z-axis would seem to shift ever-so-slightly up and down as the nozzle travelled back-to-front -- I made a few manual adjustments that I hope will help, but the system seems to be able to accomodate a slight variation, and I'm not particularly concerned.
I recently ran bed level visually on octoprint and my bed variation is around +/- 0.4mm which I only noticed because I ran it. Probably wouldn't have noticed it at all otherwise. In a past life (i.e. couple of years ago) I would obsessed over it until it was as close to 0mm as possible (meaning days of tweaking with zero benefit).
- Bob Marley
RE: Bed Leveling
Mine was pretty bad when I started - mostly the left rear corner being 1.5 mm too low.
I spent a bunch of time with spacers under the heatbed and got it to +/- 0.5. Changing the build sheet shifts that a bit and it's not worth obsessing on getting it better (as mentioned upthread).
I think the heatbed droops a bit on the back right/left sides due to poor support.
RE: Bed Leveling
It does droop there because of how the lower support is made. They had to make that back part sort of triangular so it could pickup the leadscrew in the center rear but have clearance in the corners. Because of that, they couldn't bend a straight across strut out of the sheetmetal. Instead it's two 1/2 length struts along the triangular part. That means that it can flex quite a bit right where the rear leadscrew attaches, and that amplifies quite a bit by the time you get to the corners. If you really wanted to make it rigid, you could attach a small angle that spans from the full span from left to right just ahead of the lead screw, but honestly I don't think there's a whole lot of stress on the bed in the vertical direction, so it probably wouldn't do much.