Bed deviation
Hi
So I've been playing around and retrieving the Bed Topography report from my Core One, and checking overall deviation at just under 1mm
I run the Bed Probing manually so the entire bed is probed, from what I've gathered after a bit of research 1mm is pretty good.
It seems the deviation I'm seeing is likely down to the gantry not being perfectly flat to the Z axis...I don't have any issues with the printer, so just wondering if anyone else has done the same and what kind of deviation they are getting.
RE: Bed deviation
I've been down this rabbit hole too. I found these useful. What do you see in the corner of the bed where the power attaches? Mine was being dragged down a bit.
RE: Bed deviation
I've been down this rabbit hole too. I found these useful. What do you see in the corner of the bed where the power attaches? Mine was being dragged down a bit.
Yeah my bed is showing some tilt towards a rear corner, I'm not 100% sure if I might've got the x measurements reversed...which could indicate a tilt towards the rear corner with the power connected.
What I did notice is if I do the normal Z calibration from the menu, the deviation is much greater than 1 mm, yet the printer still passes all checks and print fine. I can get to around 1mm deviation if I manually tweak the z motors with using a piece of paper.
At some point I'm gonna have to flip it over, check all the motors are all tight still...and check the top where the gantry mounts etc.
RE: Bed deviation
I found that my left front corner (0,0) was 0.2mm low, due to skew of the frame. I used those shims, the Z alignment calibration and the bed level visualizer to zero that out.
Can't remember how the visualization defaults the 0,0 corner, I put a dime under that corner of the print sheet and looked for the lump.
RE:
I found that my left front corner (0,0) was 0.2mm low, due to skew of the frame. I used those shims, the Z alignment calibration and the bed level visualizer to zero that out.
Can't remember how the visualization defaults the 0,0 corner, I put a dime under that corner of the print sheet and looked for the lump.
Yea the X axis was reversed, so the corner with the power cable is definitely being pulled down...did you correct that? or just live with it?
Just by moving the power cable I can see the bed being slightly pulled/lifted in that corner.
RE: Bed deviation
I have around 0.2mm of tilt in one corner. I just ignore it. If it were between 0.5 and 1mm I might pay attention to it. It hasn't affected my ability to produce large functional parts.
RE: Bed deviation
The slight steady slope would not concern me much; the software bed compensation must be good for something. 😉
That blue back corner is probably the bed's power connector cover getting in the way. If the rear fan shroud touches that cover during probing, the bed needs to be pushed up a bit further to touch the nozzle, resulting in a lower-than-normal Z height readout. This assumes that it is physically the back left corner and the X coordinate is shown mirrored.
RE:
Well I've found the culprit, out of the 8 expansion joints under the heat bed, 4 of them were now loose and pointing in weird directions.
Corrected that and I've no got <0.3mm deviation across the entire bed...I'll let the software compensate for that 😂 They've definitely worked their way loose since I built it...there's no way I would've assembled it like that.
I think the power cable being pulled was a bit of a red herring, that corner is within 0.1mm of the other corners now.
RE: Bed deviation
So I've been playing around and retrieving the Bed Topography report from my Core One
How are you getting the bed topography? Octoprint or something else?
RE:
Here's a question for you smarter folk:
In the assembly guide, you have to tighten the screws on the heatbed in a specific order...why is that ?
https://help.prusa3d.com/guide/4-heatbed-assembly_835514#842590
RE: Bed deviation
In the assembly guide, you have to tighten the screws on the heatbed in a specific order...why is that ?
https://help.prusa3d.com/guide/4-heatbed-assembly_835514#842590
I think the idea is that -- in case the heater circuit board has a curvature after manufacturing -- you attach it starting from the center and moving on to the outside. This way, the board can "stretch out" while the curvature gets flattened. If you started with the four outer corners instead, then forced the circuit board flat by tightening the central screws, you would create a lot of stress in the board.
RE: Bed deviation
Thats makes sense , Cheers !
RE: Bed deviation
So I've been playing around and retrieving the Bed Topography report from my Core One
How are you getting the bed topography? Octoprint or something else?
I've implemented it into my own esp based monitor...initially it was just to enable me to execute gcode commands for setting up some custom mods I use, it's grown a bit as I realised that a lot of useful info is spat out across serial.
You can get the topography data using the gcode 'M420 V' after a mesh probe has taken place, so I just implemented it as extra feature.
RE: Bed deviation
Even in large machinery, the rule about tightening multiple screws on one surface applies: Never tighten the screws sequentially around the circumference, but always crosswise! And even if the screws are to be tightened to a certain torque, they are always tightened less first (crosswise, of course) and only then tightened with a torque wrench to the prescribed torque (crosswise again). If you do not follow this rule, for example with cast iron flanges, with a probability close to certainty your flange will crack or not seal!
RE: Bed deviation
Even in large machinery, the rule about tightening multiple screws on one surface applies: Never tighten the screws sequentially around the circumference, but always crosswise! And even if the screws are tightened to a certain torque, they are always tightened less at first (crosswise, of course) and only then tightened with a torque wrench to the prescribed torque (again crosswise). If you do not follow this rule, for example with cast iron flanges, with a probability close to certainty your flange will crack or not seal!
I guess we all know that one from tightening wheel nuts!👍(Mounting cylinder heads is a somewhat less common pastime...) Same underlying issue, but the specific "algorithm" is a bit different for the heat bed since it also gets attached in the center.
RE: Bed deviation
If the screw(s) is (are) in the middle, then of course you start with this screw. Of course there is no dispute about that, it is just a matter of making sure that even beginners understand it and follow the assembly instructions. I think we have made it clear and that should be enough.
RE: Bed deviation
I've had cars where it was part of the 1000 or 5000 mile first service to retorque the head bolts. Should we be checking the heat bed after the first X hours of operation?
RE: Bed deviation
I've had cars where it was part of the 1000 or 5000 mile first service to retorque the head bolts. Should we be checking the heat bed after the first X hours of operation?
Some users here have recommended that you better check all screws on Prusa printers regularly, or apply Loctite during assembly...
I have done neither so far on my Core One. Had one screw fall out after just a couple of days, which I am pretty sure I did not tighten properly after I had done an adjustment. But I can't confirm or deny whether all other screws are still tight or whether some have come loose by now. Maybe I should check them, at least those which are accessible.
RE: Bed deviation
I've had cars where it was part of the 1000 or 5000 mile first service to retorque the head bolts. Should we be checking the heat bed after the first X hours of operation?
Not sure mate, I think this is where they use terms like 'tighten gently, but firmly...do not overtighten' and to different people these terms mean different things.
Unless they give torque measurements and we're all using appropriate tools then there's will always be ambiguity.
RE: Bed deviation
I've implemented it into my own esp based monitor...initially it was just to enable me to execute gcode commands for setting up some custom mods I use, it's grown a bit as I realised that a lot of useful info is spat out across serial.
You can get the topography data using the gcode 'M420 V' after a mesh probe has taken place, so I just implemented it as extra feature.
That sounds very interesting. I'm interested in that.
Is it easy to find on the internet, or could you share some more information?
Quick start printing for Prusa XL and Prusa Core One
Comfortable display working height on the Prusa Core One
Reducing metallic resonance noises on the Prusa Core One