MK3s XYZ Calibration failed! Check the axes and run again
 

[Solved] MK3s XYZ Calibration failed! Check the axes and run again  

  RSS
evangelos.p
(@evangelos-p)
Active Member

Hello there, 

After having succesfully printed for about 24hours (print time) we had issues with plate adhesion so started to re-calibrate the machine, but XYZ calibration failed at the very start (Z-axis). Z calibration fails as well, it is our 3rd full day of trying to fix the issue, would appreciate any help. Printer has been XYZ calibrating successfuly in the past months.

 

Repeatable Symptoms:

A. Auto Home correctly centers on left front dotted circle (pinda calibration target)

B. Z-calibration repeats auto homing, but this time it homes appx. -6 to 8mm front/left diagonal offset (from circle)

C. Z-calibration homes top up

D. Z-calibration moves to PINDA turn off Z position (which is offset from circle as described above) and error message appears: Calibration failed! Check the axes and run again.

 

It is still failing, even after trying the things below:

 

1. XY belt tensions adjusting, X259 Y269

2. hotend assembly position crosscheck

3. PINDA Z position check (tried with many heights except the "correct" one)

4. PINDA connection check (does not flicker, either in software or LED, 0 or 1 works as expected)

5. checked and fixed all cable routing to address homing issues.

6. Z-Axis Homing (reaches the top, collides with top Z bracket)

7. X-Axis Homing (reaches far left, extruder collides with Z slider)

8. Y-Axis Homing (left back bearing collides with back Y axis holder)

9. Self Test Passes all tests

10. Z-Axis Worm reseating (lose black worm retainer screws, move up down, tighten)

11. Firmware reflashing, versions 3.5.3, 3.6.0, 3.7.2

12. Factory Reset

13. correcting XY PINDA offset during calibration homing (we used spacers to force Z-calibration XY homing, forcing itto fall on dotted circle)

 

Pretty much we have tried everything we found in this forum and other sources.

 

things that we have not tried that other people said it solved their problem:

n1. replacing the fillament sensor

 

As we have allready spent 3 days on this, any help would be appreciated!!

 

thank you in advance

 

ps. if you need any clarifying photos/videos please ask, though the problem is repeatable and easy to describe.

Best Answer by evangelos.p:

problem seems to be resolved.
 
our assistant was downloading MK3s firmwares for a MK3 printer. that is why the problem started from one point and on.
this gave us the opportunity to check all the machine and calibrate tensions, homing, cable routing etc. as well as collecting all possible debugging in one forum post.
 
we found out about the firmware issue by insisting on the filament sensor erros (that we had not tried, as mentioned in the post)
since other users had wring firmware issues (mixing MK3 witsh MK3s), prusa could add some disclaimers about it in the firmware download section.
 
thank you all for your time, happy about using prusa printers and the immediate support, either via this forum or the official chat option.
This topic was modified 1 year ago 2 times by evangelos.p
Posted : 01/08/2019 3:22 pm
Tim
(@tim-m30)
Illustrious Member

I've read many reports of this yet haven't seen a single conclusive and common "This fixed it!" report.  

One reason for failure is the extruder is bumping into something as it travels. One case was the spool holder was blocking Z travel.  Another was cable ties blocking X travel.  Another the cooling fan screws the builder used weren't the right ones and the heads were limiting X travel.  Another had the extruder hot end built wrong (had added a volcano like heater) and again Z travel was wrong.

So check there aren't any obvious reasons limiting the X Y and Z axes movements.  Make sure they can travel at least 250, 210, 210 mm.

Regarding the home position: It generally isn't when the PINDA is directly over the white dot. Home is a bit left and forward of it and seems to vary by build (based on photos folk have posted). 

 

It is always wise to get more than one opinion......
Posted : 01/08/2019 7:51 pm
bobstro
(@bobstro)
Illustrious Member

Not that it every happened to me, but an errant bit of filament under the spring steel bed sheet can cause problems (though XYZ is done without it.) I also never had a stupid little decorative piece plugged into the hex holes on top of the Z brackets cause problems, or a misrouted cable do the same. Nor a bit of hardened filament ooze in the nozzle go bump. But it could happen.

You've done the factory reset, and that was the answer I resorted to when something similar happened to me.

If you suspect the filament sensor, have you disabled it?

My notes and disclaimers on 3D printing and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
Posted : 01/08/2019 8:35 pm
evangelos.p
(@evangelos-p)
Active Member

thanks for the feedback, yes, we have disabled the fillament sensor as well (forgot to mention it).

Posted : 01/08/2019 9:05 pm
evangelos.p
(@evangelos-p)
Active Member

thanks for the reply,

we have checked all collisions (prior to main parts hitting main parts) and our machine is stock.

allowed distances are X = 0 to 255, Y = -4 to 212, Z = 0 to 210, so the seem unconstrained.

the PINDA sensor perfectly aligns with dotted circle on Auto Home. It does no alignin only when Z or XYZ calibrating.

Is it possible we have assembled the bottom plate in a way that creates the offset?? trying to figure it out.

we have (by force) corrected the PINDA position (with XY spacers) which did not solve anything, as described above.

Posted : 01/08/2019 9:14 pm
Tim
(@tim-m30)
Illustrious Member

How about photos of the extruder from a few angles after a HOME.  Show the gap between the extruder and X-axis motor mount, and a view of the entire frame and bed. And an image of the extruder from the top, powered off, pushed all the way left.

Another cause is a spinning belt drive gear.  When the set screws aren't tightened in order, they will come loose and the gear spins on the shaft. This affects the "length" measured by the firmware.

This post was modified 1 year ago by Tim
It is always wise to get more than one opinion......
Posted : 01/08/2019 9:22 pm
bobstro
(@bobstro)
Illustrious Member

Kit or prebuilt? @tim-m30, the full range of motion would eliminate the ol' upside-down trapezoidal nut issue, wouldn't it? @evangelos-p can you post a pic showing the full printer clearly?

My notes and disclaimers on 3D printing and miscellaneous other tech projects
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
Posted : 01/08/2019 9:26 pm
evangelos.p
(@evangelos-p)
Active Member

@tim-m30 will check the axles, thank you, might explain the longer axes, here are some draft images

@bobstrowe attaching the images requested

Posted : 01/08/2019 10:04 pm
Tim
(@tim-m30)
Illustrious Member

Try this: take a fine Sharpie: mark a line on the face of the X and Y motor shafts and drive gears.  Run XYZ cal.  Check if the lines moved. 

 

It is always wise to get more than one opinion......
Posted : 02/08/2019 12:20 am
evangelos.p
(@evangelos-p)
Active Member

X, Y shafts and gears seem to work as one piece without slipping. 

Posted : 02/08/2019 7:37 pm
Tim
(@tim-m30)
Illustrious Member

Contact Prusa support using the CHAT pane on their store site.

 

ps: this step seems like you may have broken something:  adding spacers is never a good idea, nor necessary.

13. correcting XY PINDA offset during calibration homing (we used spacers to force Z-calibration XY homing, forcing itto fall on dotted circle)

 

This post was modified 1 year ago by Tim
It is always wise to get more than one opinion......
Posted : 02/08/2019 8:30 pm
evangelos.p
(@evangelos-p)
Active Member

adding spacers is the same with having cables or whatever blocking the axes which just leads to different homing. prusa mk2s and on use self homing motors so no need for external switches. if you read the order of actions, it is last on the list anyway so not the source of the original problem. if something was broken by spacers the machine would not function properly which it (apart from xyz calibration), does.

we will contact prusa chat as suggested, thanks for the tips.

Posted : 03/08/2019 3:56 am
evangelos.p
(@evangelos-p)
Active Member

prusa support suggested logic board failure. waiting for replacement to proceed with debugging.

Posted : 04/08/2019 8:14 pm
Tim
(@tim-m30)
Illustrious Member

I seriously doubt it's the board ... there isn't anything there that would cause the problem you are seeing.  Perhaps bad firmware (like 3.7.2 is broken) ... but any electronic failure would result in a more apparent failure mode.

Try flashing 3.5 firmware with a full data clear. See if that can get you past calibration.

It is always wise to get more than one opinion......
Posted : 04/08/2019 9:33 pm
Tim
(@tim-m30)
Illustrious Member

But what is a bit curious, is why you thought an adhesion problem was hardware related.  Can you explain your thinking on that?

 

It is always wise to get more than one opinion......
Posted : 04/08/2019 9:36 pm
evangelos.p
(@evangelos-p)
Active Member

you can reread our original post. we have tried all this. we had actually already tried everything before chatting with prusa, as prusa support mentioned before sending the board. and that was a 4 hour video/photo session...

we did not think adhesion was hardware related, just went to recalibrate Z-axis and we found out it was not working. it is mentioned in the original post i think.

This post was modified 1 year ago by evangelos.p
Posted : 05/08/2019 1:36 pm
Tim
(@tim-m30)
Illustrious Member

In my mind recalibrating Z is hardware related. Something had to physically change positions.

Lack of adhesion is most often simple contamination on the bed. Fingerprints the most common form, followed by PLA residue after dozens and dozens of prints. If the bed has been used for months, then oxides can form, but less common.  

So when you get the new hardware, might be a good time to do a serious cleaning of the print bed.  Most people argue their bed is clean, but when nothing else has worked and they try soap and water, they are usually quite surprised by how such a simple task works so well. Alcohol washes just make the issue worse by thinning the fingerprints until you don't see them - but leaves a microscopic layer of oils. 

 

It is always wise to get more than one opinion......
Posted : 05/08/2019 3:39 pm
evangelos.p
(@evangelos-p)
Active Member

thank you for the tips, you are right in all points. machine was very thourougly cleaned after each print, bed appears to be in mint condition and we will follow the procedures upon receiving the new hardware.

we would anyway need to recalibrate on every firmware update afawk = would run into the same issue sooner or later.

we also think that a board failure would lead to general malfunctions as you suggested and it is not the source of the issue, but since prusa support suggested so, we can only hope it resolves the matter.

Posted : 05/08/2019 6:02 pm
evangelos.p
(@evangelos-p)
Active Member
problem seems to be resolved.
 
our assistant was downloading MK3s firmwares for a MK3 printer. that is why the problem started from one point and on.
this gave us the opportunity to check all the machine and calibrate tensions, homing, cable routing etc. as well as collecting all possible debugging in one forum post.
 
we found out about the firmware issue by insisting on the filament sensor erros (that we had not tried, as mentioned in the post)
since other users had wring firmware issues (mixing MK3 witsh MK3s), prusa could add some disclaimers about it in the firmware download section.
 
thank you all for your time, happy about using prusa printers and the immediate support, either via this forum or the official chat option.
Posted : 07/08/2019 4:52 pm
Share:

Please Login or Register