RE:
I tried adjusting my belts using the newly added option in the Prusa app. I tuned both left and right belts to 85Hz with the print head dead center and all the way back, but now the Y axis won’t calibrate. I noticed that the linear rail on the back right does not sit flush against the back stepper unless depressed towards it, while the back left side sits entirely flush. If I reduce tension on the front right screw this corrects but then the belt will have far less tension than is required for 85Hz.
Does anybody know how to deal with this? Not sure what to adjust to square things up under tension, and there is no assembly manual yet so I can’t even figure out what to check that may be incorrect.
RE: Belt tension adjustment
If it's anything like the XL, when you bring the carriage all the way to the front it should be flush on both sides. If you apply pressure it shouldn't move at all. Sounds to me like the gantry is crooked after your adjustment.
Being such a new printer I'd recommend contacting support.
Here is a link to the XL instructions.
https://help.prusa3d.com/article/adjusting-belt-tension-xl_401793
RE:
That video shows calibrating the belt tension with the carriage moved about halfway forward. The Prusa app for the CORE One says:
Manually move the nextruder to the furthest center position (make sure you have disabled motors).
I interpreted furthest as towards the back as far as it would travel. There's no linear rails on the Y axis, and the case isn't _that_ rigid on the CORE One, so even though it was packed well, I hope it wasn't tweaked in shipment. I'll have to check if it sits flush at the front, it didn't at the back as described above.
I'm in the queue for Prusa support, so hopefully they'll have some answers. Thank you!
If it's anything like the XL, when you bring the carriage all the way to the front it should be flush on both sides. If you apply pressure it shouldn't move at all. Sounds to me like the gantry is crooked after your adjustment.
Being such a new printer I'd recommend contacting support.
Here is a link to the XL instructions.
https://help.prusa3d.com/article/adjusting-belt-tension-xl_401793
RE: Belt tension adjustment
I'm in the queue for Prusa support, so hopefully they'll have some answers. Thank you!
Please let us know what they say. I read the instructions as being all the way pushed back on the y and centered on the x axis.
RE: Belt tension adjustment
It seems that a lot of posts from the past day have been lost so I will rewrite mine for completeness.
I'm in the same boat with belt tensioning. I was getting homing issues (Auto-home took forever to complete or failed all together) so I looked into the belt tension. I noticed that the stock tension wasn't anywhere near what the app wanted so I adjusted. After a lot of tweaking I finally got both sides to be ~85Hz, however this resulted in the gantry alignment being out of skew, about 8-10mm on the right side between the frame and the axis when I brought the axis all the way towards the front, with the left side sitting flush. This then resulted in the y-axis calibration failing.
If I instead adjust for squareness of the gantry then my right belt is super tight and my left belt is super loose, both off the scale on the tuning app (Are the belts different lengths?). I can adjust them further to keep the gantry square and get one belt to be 85Hz, but this either makes the left even looser or the right tighter. I went for a square gantry with 85Hz on the right and loose on the left, which allowed the y-axis calibration to pass again.
I don't know if this was the right call, or if there is any way to actually make the gantry square with both belts being at 85Hz, but in the current configuration this was the best I could do. I don't want to disassemble the whole gantry to see if the belts are indeed different lengths, but its my biggest suspicion, either that or the printer itself just isn't square. Mine arrived with the box in almost perfect condition with no signs of damage, so I doubt its that which would cause such a big gap to form when the belts are tightened the same.
RE: Belt tension adjustment
Thanks for re-posting this. It's a pity that several discussions got interrupted by the data loss.
From what I understand, the CoreXY subsystem itself not being square should not be possible by design. It seems that it does not rely on the accuracy of the enclosure frame but has its own support frame -- the flat horizontal profiles to which the motors and idlers are mounted; one of them runs across the front to ensure stability and accuracy of the frame. I don't have my Core One kit yet (obviously...) and cannot make it out very clearly from the pictures -- but isn't that whole frame stamped from a single strong sheet of steel, so the 90° angles are not dependent on assembly accuracy?
Can you check your frame for 90° angles by fitting an angle gauge into the corners? Looks like the should be accessible with a bit of Yoga practice. 🙂
The belt lengths are defined during assembly, when the four ends of the belts are clamped to the Nextruder, I think. It's quite conceivable that the match could be less than perfect, say by +- one belt tooth. But the 8-10 mm you get seems clearly excessive -- unless the assembly technician had a bad moment and pre-cut one of the belts 1 cm too long or too short.
RE: Belt tension adjustment
Yes, but even if the outer frame is square the gantry can be skewed relative to the frame. From photos, it appears the gantry assembly consists of a steel rail bolted to a stamped plate to which the bearings attach at the ends. If the stamped bend is not exactly 90 degrees, the bearings will be skewed accordingly. My core one is still on order, but if I had the aforementioned problem, I’d be tempted to just give a good yank on the right side of the gantry to see if it can be bent in place back to the proper alignment (temporarily loosen both belts first). Don’t try this and then blame me if you break it, but it’s something to look at.
RE: Belt tension adjustment
Yes, but even if the outer frame is square the gantry can be skewed relative to the frame. From photos, it appears the gantry assembly consists of a steel rail bolted to a stamped plate to which the bearings attach at the ends. If the stamped bend is not exactly 90 degrees, the bearings will be skewed accordingly. My core one is still on order, but if I had the aforementioned problem, I’d be tempted to just give a good yank on the right side of the gantry to see if it can be bent in place back to the proper alignment (temporarily loosen both belts first). Don’t try this and then blame me if you break it, but it’s something to look at.
Sure, the gantry can be tilted -- as observed by David. My point was that this is much more likely to be due to belt length (determined at assembly time) than due to a non-square frame (determined by the presumably stamped part, which should never deviate by more than a few 1/10 of a millimeter).
I'm not sure yanking at the gantry will do much. If you do so after relaxing the belts, it should certainly be possible to pull the gantry to the desired position. But as soon as you tighten the belts again and bring them to the same tension, they will pull the gantry back to where it was before, right?
RE: Belt tension adjustment
Surely if the gantry was out of square due to bending etc. then loosening the belts wouldn't make it fall back into square? Or am I misunderstanding what you mean?
RE: Belt tension adjustment
Someone yesterday posted a picture showing the correct gantry position, but it got lost… Any chance of getting it posted, please?
RE: Belt tension adjustment
Surely if the gantry was out of square due to bending etc. then loosening the belts wouldn't make it fall back into square? Or am I misunderstanding what you mean?
Just to make sure we use the same terminology: By "gantry" I mean the linear rail which guides the extruder in the X direction, and the parts attached to it.
If the belts are quite loose, you should be able to move the gantry pretty freely. The only forces which give it some direction in that case are its bearings on the Y rods -- and in a "perfect" world of idealized mechanics, these would be bearings which do not constrain the direction of the gantry at all. The main constraint to define the gantry position and angle comes via the belts -- which is why you can "correct" the angle by setting an asymmetric belt tension.
RE: Belt tension adjustment
Someone yesterday posted a picture showing the correct gantry position, but it got lost… Any chance of getting it posted, please?
You mean the correct position for taking frequency measurements on the belts? Push the gantry all the way back (Y direction), position the extruder in the center of the gantry (X direction).
RE: Belt tension adjustment
Do you mean for doing the belt tuning process?
Push the x-axis all the way to the rear of the printer and place the extruder body in the middle of the axis. This should result in both belts experiencing the same tension when plucked so you can compare frequencies
Someone yesterday posted a picture showing the correct gantry position, but it got lost… Any chance of getting it posted, please?
RE: Belt tension adjustment
If the belts are quite loose, you should be able to move the gantry pretty freely. The only forces which give it some direction in that case are its bearings on the Y rods -- and in a "perfect" world of idealized mechanics, these would be bearings which do not constrain the direction of the gantry at all. The main constraint to define the gantry position and angle comes via the belts -- which is why you can "correct" the angle by setting an asymmetric belt tension.
Having stared at a diagram again, I am no longer sure I believe my own words here... 🙄 The belt tensioners can compensate for quite a bit of unequal length of the belts, so that parameter shouldn't be too critical during assembly. And if the belts are adjusted to the same tension, they should not exert any tilting force on the gantry.
So if the gantry sits askew in that situation, maybe there is something else that exerts a force onto it. Y guiding rods not mounted parallel to the CoreXY frame? Gantry's linear bearings on the Y rod not orthogonal to the gantry?
RE:
If the belts are quite loose, you should be able to move the gantry pretty freely. The only forces which give it some direction in that case are its bearings on the Y rods -- and in a "perfect" world of idealized mechanics, these would be bearings which do not constrain the direction of the gantry at all. The main constraint to define the gantry position and angle comes via the belts -- which is why you can "correct" the angle by setting an asymmetric belt tension.
Having stared at a diagram again, I am no longer sure I believe my own words here... 🙄 The belt tensioners can compensate for quite a bit of unequal length of the belts, so that parameter shouldn't be too critical during assembly. And if the belts are adjusted to the same tension, they should not exert any tilting force on the gantry.
So if the gantry sits askew in that situation, maybe there is something else that exerts a force onto it. Y guiding rods not mounted parallel to the CoreXY frame? Gantry's linear bearings on the Y rod not orthogonal to the gantry?
Looking at the (albeit out of date) gif on the Core Ones product page we can see the full gantry with the frame
(Click the image to see it full and uncropped)
This is indeed a single piece of steel with a square stamped into the middle, and it even has curved edges to add more rigidity along with it being screwed to the outer shell. I really don't see how tightening belts could pull this out of square, which means it must be the x-axis itself being pulled out of square with the frame along the smooth rods
RE: Belt tension adjustment
Correct, the X axis is what's actually being pulled out of alignment, I was not using the correct terminology earlier. The picture was very helpful. The X axis will end up with one of the ends being pulled further towards the rear/further towards the front than the other, at equal belt tension left/right, on some of our printers (sounds like more than a few), while others equal belt tensions is achievable. The question is now what causes this and how to correct it. I'm not sure if the gantry itself is a solid piece of metal or bolted together, I will have to look at mine later.
RE: Belt tension adjustment
So I just ran another test by loosening both belts until they were only tight enough to not derail the idlers, essentially removing all pull they have on the axis, and my x-axis is still out of square with the end stops by the same amount as when both belts were 85Hz, so that means its not the belts in my case, and my x-axis is just out of skew with the y-axis.
Between this and my other issues with z-axis binding, i've contacted support and think we are arranging for my core one to be sent in for a service
RE: Belt tension adjustment
Here is a video showing the loose belts and the gap my x-axis has on the right when bottomed out on the left
RE: Belt tension adjustment
I just got mine today and it has the exact same issue, only the gap is 2-3mm.
RE: Belt tension adjustment
I just got mine today and it has the exact same issue, only the gap is 2-3mm.
I expect the number of affected printers getting much higher. The problem may be caused by the metal brackets, attaching the X-axis to the Y-Axis bearings. If they are not bent to the exact 90° angle, the angle of the X-axis is skewed.
Maybe you want also take a look at this thread: https://forum.prusa3d.com/forum/prusa-core-one-assembly-and-first-prints-troubleshooting/failed-y-calibration-of-a-new-coreone/#post-737338