xyz calibration failed. bed calibration point was not found
I recently bought my first 3D printer (Prusa MK3S i3) kit and assembled it. Everything went smoothly until the XYZ calibration. The calibration starts well with the Z-axis reaching the top, then the extruders lowers and starts 'searching' for the calibration points. It approaches the bed but stays 1-2mm away from the paper sheet and does not actually touch it. After searching for a some time it shows the error 'bed calibration point was not found' without reaching any of the 4 points. At this moment the PINDA is close to the 'reference cirkel' but not concentric.
To tackle this issue I contacted the helpdesk and gained the following tips: Tighten X & Y belts (done; X-270 Y-285 ), lower/higher PINDA (done), lubricate rods (done), check pulleys (done), check if all axes move freely (done).
Besides I tested the PINDA (works), checked if everything is level (axises, bed, etc.) and reseted the printer.
I also checked the Prusa help page regarding this topic ( https://help.prusa3d.com/en/article/xyz-calibration-mk3-and-mk3s_112351). Point 5 might be my problem (PINDA does not center when I auto-home). The issue here is misalignment of my axis assembly but I can't seem to figure out what is causing this...
After a few days of trial and error I am completely stuck and hope if some of you guys can help me with tackling this issue. I added some picture of the printer, maybe there is an obvious solution I am not seeing. I have a video as well, if helpfull (link: https://drive.google.com/file/d/1Cs5uSbEut6iq4VguwVaZB7bqQtiwR6M7/view?usp=sharing)
Hopefully someone can help,
Since your problem sounds Z axis related, I would start there.
First make sure the dust covers at the motor are not touching the motor. They should be up the shaft a bit so they don't drag.
Next to make sure the shafts are parallel. Loosen the screws securing the Z motors to the frame, and the screws in the top brackets.
You should be able to wiggle each shaft a bit.
Now, from the firmware, run the extruder axis from Z bottom to Z top a couple of times.
Now you know the shafts are parallel. Tighten up all the screws you loosened.
And tell us how you set the PINDA height.
Thank you for the tip! I will check it out asap.
For the PINDA I used a zip tie initially. Afterwards I lowered it as suggested by the prusa helpdesk.
Zip ties have a thinner tip. That is the part you use for spacing.
I would not go any lower than that.
I did the adjustments but the it doesn't help 🙁
Any other tips?
Did you check if the probe is being detected? In the menu go to "support", "sensor info". Then use something metal to trigger the probe, should read 0 when the red LED is on and 1 when the red LED is off. Something like a stainless steel spatula should trigger the prob at about 3 mm.
What the prob is searching for is a circle of copper in the heat bed under that white printed circle. In those initial searches it should be moving back and forth over that circle.
Hi, I checked the PINDA out and it works according to this test. The problem remains though. Thanks for the tip!
I'm not sure about it, but ...
...in the IMG_4212 i can only see 2 of the 3 zip ties that fix the textile sleeve (see manual Tigthening the textile sleeve). But those two are not shortened. When the first is not shortened, too, it can collide with the frame when the z axis moves to the top. This would give a wrong position to the calibration.
You right, they are not shortened correctly. Thanks for the info!
Unfortunately the did not collide with the z axis on top so this could not be the reason 🙁
It seems to be searching OK in Z and X, only question remaining is if it's searching correctly in Y.
Check for any binding in Y. Is the motion of Y smooth all the way from end to end when moving slowly by hand? Is there anything blocking the Y carriage from hitting the ends? On the rod with the 2 bearings they should hit both ends. Is the cable hitting anything when it moves all the way back making it think it's hit the end when it hasn't?
One thing I noticed in the video it didn't look like the Z right side was getting all the way to the top, but it could be the camera angle, the heads of the screws holding the black nut should be hitting the orange plastic on both sides when you hear the rapid clicking sounds.
Thank you for the reply!
When calibrating, should the nuts (1) of the z-axis top parts stop the x-axis? (see pic) or should it be the nuts on top of the trapezoidal nuts (black piece) (2)? I have put a picture as reference.
Additionally, I was able to perform a z-calibration! Maybe this singles out the issue...
Pity. Can you measure the length of the z axis rods? Should be 320 mm.
The Z-axis is 320 mm.
Ok. Then i'm out. Sorry.
No problem. Thank you very much for all the input!
Sorry, yes, the heads of those screws do stop the Z axis, mine is so close I never looked close enough to see the tiny gap above the trapezoidal nut hold down screws when it hits the end of travel.
Have you watched as it does the search over the circle to see if it's crossing in Y as well as X?
At this point what I would probably do is move the pinda probe down a few mm just for testing, to see if it can find that first calibration disk.
@mrschultz I tried this but it did not work (lowering the probe).
I had some more contact with the helpdesk. They told me that it might be a bug. Since the Z-calibration succeeded they told me to try to part and see what happens since everything might be okay..
Yeah, if all the hardware checks out it might be a firmware issue, did they have you do a factory reset or flash new firmware at any point?
Without the automatic XYZ calibration it should still print fine, but part sizes might not be accurate and you don't get any correction for possible frame alignment issues.
Yes I did both but they did not resolve any issue. A suggestion was also to load an old firmware version.
Nevertheless I am curious why it can fail or what the reason can be. It might be something very obvious that I am missing..
Is the bed heating OK? If it is then the copper traces on the bed should all be OK including the calibration disks.
I guess you could see if the calibration disk is there by moving the probe over the circle and see if the light goes out when you move down.
If you get a good first layer live calibration then print some calibration cubes and measure them.