RE: Skewed XY-Plane (Z Rods, Heatbed)
So over the weekend I built my Core One. [...]
It is a real bummer that there is such poor quality at the leadscrews. [...]
I take it that this was also a newly received Core One kit (i.e. you did not have it in storage for a few months)? If so, that would really be disappointing QC on Prusa's part, and apparently a total lack of feedback from their support team to production.
They should have learned from customer feedback in the meantime, and either have established dedicated checks that all three lead screws have the same pitch -- or at least stopped the practice of combining the pre-packaged MK4 motor kits with a separate motor, presumably from a different production lot.
RE: Skewed XY-Plane (Z Rods, Heatbed)
I ordered it Nov 20th during black Friday sale. When they announced the Plus version I hit the button. Received it last week. So brand new. Or maybe a kit which was in stock and they only added the Plus addon. I have to check the calibration sheet when everything was checked...
RE: Skewed XY-Plane (Z Rods, Heatbed)
Today I got the new motors. I printed 3 cylinders with 260mm heigth in vase mode near the leadscrews (left and right front, rear middle). Left and rear were 260mm, right was 261mm. Swapped the right motor and suddenly 260mm are 260mm. So my C1 is fixed. I also ordered some leadscrews (those also run smooth without binding) for my repair stock (we have 7 MK3.5/3.9/4.0 and one XL in the office and also 2 Core one, so it's nice to have some spare parts handy). What I really like is that many (unprintable) parts are similar over the different models.
I will check the Core1 of my colleague after the holidays and may also swap some motors and then I will deal with support and ask for a refund for the "defective" motors.
It is really a nice printer, but such issues with QC are spoiling the experience...
RE: Skewed XY-Plane (Z Rods, Heatbed)
Thanks, all, for documenting your experience. I am afraid I have the same problem. But I confess that I am VERY inexperienced with these machines and not terribly technically proficient, so I am DREADING the process of troubleshooting with tech support. I started my 3D printing journey in 2022 by building from kit an MK3S+. So much fun. Worked like a charm. Loved it. Then I did the full upgrade to MK4 as soon as it was available. Loved it. Same for MK4S upgrade. Then I ordered the upgrade kit for the Core One in February. It arrived in June (not sure why it took so long). I assembled it in August. And I have never had a successful print since due to first layer issues that sound exactly like the problems described in this thread (XY plane). Unfortunately, I just haven’t had sufficient free time to throw at troubleshooting. With the holidays newly here, though, I am hoping that I might have a little extra time over the next two weeks.
Assuming I’ve correctly diagnosed the problem, does anyone have any tips for how I might most efficiently go about solving it without having to spend hours and hours with tech support? Part of the reason I picked Prusa in 2022 (and am committed to sticking with it now) is how well established and generously helpful the Prusa community is. All tips and advice for a non-power user are most sincerely appreciated.
Happy Holidays, all!
RE: Skewed XY-Plane (Z Rods, Heatbed)
Looks like I'll be running a test seeing as I had a very tight leadscrew nut. That is, after I get a new main cable because the printer won't complete bed probing. I'm utterly disgusted with Prusa at this point. I bought these printers to support a company that I THOUGHT was doing the right thing. Turns out I not only wasted money, but time as well.
RE: Skewed XY-Plane (Z Rods, Heatbed)
I’m having similar issues. I have a mk4s conversion to Core One, which I received in August. I can visibly see the front edge rising about 1.5mm when first starting a print. I moved the Z axis to max, turned off power, and manually rotated all three Z rods - no movement. All were bottomed out. Moved Z axis to zero and there is a tilt of the bed toward the front of the machine. I did the paper and pencil trick and the thread pitch is identical on right and left Z rods, but different on the center Z rod. I’ll have to count threads to tell you by how much, but over the course of 250mm it’s about 1.5mm.
The two front Z rods came from my old mk4S. The center rod came in the conversion kit. It doesn’t really matter if the rod pitch is slightly different from batch to batch, though I would argue that a o.6-0.8% manufacturing error from batch to batch is unacceptable. What matters more is that all three Z rods have the exact same pitch. And mine don’t. The problem is that bed leveling compensation occurs the same amount over the entire Z range, but the bed tilt is worse at the top than at the bottom. A print a few millimeters thick won’t be affected much. But a 200mm print will be skewed.
Solution 1 is to have all three Z motors controlled separately. You can then compensate for different Z rod pitch in software.
Alas, the Core One Z motors are all connected in parallel. Thus, solution 1 won’t work. Solution 2 is to replace one or more Z rods so that all three Z rods have the same pitch. But for practical purposes all three rods should be replaced with a matched set else you end up shipping rods back and forth until you iterate into an approximate match.
Solution 3 is for Prusa to get a more consistent manufacturer and do in-house QCso they have some long term parts consistency.
Solution 4, after this experience, is to never buy a Prusa conversion kit that involves a mix of re-used and new lead screws.
I will contact Prusa tech support and see how it goes.
RE: Skewed XY-Plane (Z Rods, Heatbed)
During Black Friday, I bought a kit for One Plus, and while assembling it, I noticed that the nuts on the front left and rear screws were much harder to turn than the front right screw. During the initial calibration, I slightly loosened the nuts and lubricated them, and the calibration was successful. But now I notice that due to the uneven friction in the screws, the bed is actually tilted, and it is possible that the pitch is different. I just didn't suspect during assembly that the screws might have different pitches. It is also possible that the nuts themselves were manufactured with a negative tolerance.
RE:
The leadscrew should not require force to be moved. When spun it should move at least a little further.
Reach out to support to send you leadscrews. If you move the bed to the lowest and highest position there should be no stuttering (stepper skipping steps). It's a pity that there are not 3 individual stepper drivers (which could detect skipped steps)....
My center rear rod was also off. I ordered 3 motors with the long cable for MK4/C1, and those were fine. So it seems you have to replace the leadscrews and a motor.
RE: Skewed XY-Plane (Z Rods, Heatbed)
Are the lead screws removable from their motor? If so, I think I would just need three new matched lead screws.
RE: Skewed XY-Plane (Z Rods, Heatbed)
No, they are permanently attached. As mentioned in my earlier post, Prusa support did eventually send me three replacement motors & lead screws for free.
RE: Skewed XY-Plane (Z Rods, Heatbed)
If Prusa does send me three new (presumably matched) Z motors, I am not looking forward to disassembling my unit and reassembling all over again. I know I have to do it, but holy horsefeathers, what a mess Prusa has gotten themselves into.
If I was only going to print flat stuff, I would just leave the printer as-is. However, I’m printing nosecones and other rocket parts that take advantage of pretty much the entire Z height and I need the accuracy.
RE:
If Prusa does send me three new (presumably matched) Z motors, I am not looking forward to disassembling my unit and reassembling all over again.
It's not too bad. Place the printer on its back for this, providing some support to make sure it does not rest on the protruding WiFi module. You can then remove the trapezoidal nuts, and detach the motors and pull them out towards the bottom. But you will have to remove the frame's bottom profiles on the sides and front to get the two front motors out.
RE: Skewed XY-Plane (Z Rods, Heatbed)
I cut the foam feet according to the contours of the steel profiles. Lay it on the side or the back (take care of the wifi module).
Before removing the motors print test cylinders. Only my rear motor (spindle) was crap. Although the 2 front trapezoid nuts were also too tight, so I had to remove all motors at least a bit to install the new nuts.
RE: Skewed XY-Plane (Z Rods, Heatbed)
I think there actually may be a way to compensate for a lead screw pitch discrepancy in software, despite all three motors being driven in parallel. It relies on the fact that extrusion occurs at a single point in space, plus the presumption that any leadscrew pitch discrepancy is consistent through the entire length of the leadscrew. The bed leveling measurement just needs to occur at Z positions 0 and 275, or thereabouts. Then in software you can create a three dimensional skew map. Then every time the Nextruder is moved, the bed would need to be skewed to compensate for where the nozzle tip is supposed to be. Mesh bed leveling is already doing this in two dimensions. You just have to add the third dimension to the map. Granted, it's a much bigger math problem to solve and the added time to create two mesh bed leveling planes and then extrapolate the compensation required within that 3D space would add enormously to the print start-up time. Plus, there may not be enough memory or computing power in the current hardware to do this.
It would be easier for Prusa to just use leadscrews with consistent dimensions. That way, the mesh bed leveling would be the same at both 0 and 275 Z-positions.
RE:
I think there actually may be a way to compensate for a lead screw pitch discrepancy in software, despite all three motors being driven in parallel. It relies on the fact that extrusion occurs at a single point in space, plus the presumption that any leadscrew pitch discrepancy is consistent through the entire length of the leadscrew.
I think you have the same approach in mind as was suggested here? Real-time height adjustment as a function of the XY position of the print head is already done. They would just have to modify that and add a (linear) Z dependence as well.
I was expecting Prusa to release such a correction, since it would have spared them the hardware replacements. But there probably are just not enough users who actually notice this and ask for new motors; so it's cheaper to send out some parts for free. This might change if some major Youtuber picks up on this flaw and blows it up.
Edit: The linear Z-dependent correction should actually require much simpler math than the curved XY plane. Both regarding pre-calculations before the start of the print, and for real-time correction during the print.
RE: Skewed XY-Plane (Z Rods, Heatbed)
No brain surgery through rectum... just use correct leadscrews... those corrections complicate the software unnecessarily.
RE: Skewed XY-Plane (Z Rods, Heatbed)
Sure, a hardware correction is the clean solution. It just becomes costly when you have to offer it for every kit Core One. But as mentioned, that will not be necessary since most users don't ask.
RE: Skewed XY-Plane (Z Rods, Heatbed)
It would have been cheaper in the long run if PRUSA had QC'd their parts in the first place.
RE:
It would have been cheaper in the long run if PRUSA had QC'd their parts in the first place.
Hindsight is 20/20. (Sorry, could not resist when I noticed your user name. 😉)
What I find really baffling is that Prusa have apparently still not dealt with this issue. We had a very recent report of a newly purchase kit which still had the different lead screws. Ans when I was in touch with support about replacement motors, they told me that they don't know what screw pitch a given motor has, that they still have just one standard part number and don't qualify the motors for screw pitch.
RE:
They should have at least 2 part numbers because the cable length is different. I ordered the ones with the long cable and the pitch is correct.
This should be checked in the future during receiving inspection, at least some samples of the batch. Also the pitch and tolerances should be specified better. Till now a Prusa printer was a precision machine (for me). When I drew a part 100x100x100 it was exactly that. The tolerance of let's say 1mm off at 220mm may be okay for a normal thread, but definitely not for a leadscrew.
In the next days I will check the C1 of a colleague. It was one of the first batch and I am quite curious...