y and z self test failure after update to 5.1.0
Clear all

y and z self test failure after update to 5.1.0  

Page 4 / 4
Active Member
RE: y and z self test failure after update to 5.1.0

I don't think belt tension has any bearing on distance travelled unless it is that loose that it is slipping. The distance travelled should be determined by stepper motor and so will not change. I believe the issue with the Y axis fail is actually the positioning of the linear bearings within the pockets, rather than the belt tension.

I don't know the answer to those absolute values but assume it must be a relative value to something

Posted : 22/03/2024 4:34 pm
Estimable Member
RE: y and z self test failure after update to 5.1.0

Clearly you are not understanding the cause of the issue, and how the Y axis calibration works.

The Y axis calibration tests the limits of y axis movement.

On the Rear of the Y axis (the bed) it is moved to the rear limit, which is determines by the Belt tensioner striking the front of the Y motor pulley mount.
The front limit. is determined by the Front Y axis bearing (left side) striking the left side rod mount (rear).

The part giving trouble is adjustment of the front left Y axis bearing, which was given in the assembly guide to have the two bearings on the left side mounted as close as possible to each other while remaining inside their inset on the bed.  Which means the front bearing is moved a little backwards, and the rear bearing a little forwards.  The root cause of this while issue is the assembly guide lacks precision, and betrays no insight about the effect this has on calibration of the Y axis.

Technically, you do not need the shim at all, what's needed is to simply move the front bearing a little more towards the front.  By about 0.1 mm (or so).  But it's tricky to get into the screws that secure the bearings cover. So, someone invented the shim instead, so maybe you do indeed need it, for the convenience.

The firmware calibration test fails because the firmware programmer decided that the number of stepper motor steps needed to get the defined y axis movement was a certain discrete number. And made no allowance for the effect resulting from above reference in the assembly guide.  Left and right hand in Prusa not talking to each other.  But the previous version of the calibration test may have been a cause of (say) layer shifting or some such issue.  But thats just a guess, to fill the vacuum caused by the fact  they never said WHY the change was made, hell, they never even acknowledged it had been made at all. It was left to all the customers to discover the issue when they next ran the calibration test.  Bad mannered, rude, inconsiderate, and other uncomplimentary words go in here.

Your other questions seem irrelevant to me, perhaps you are overthinking the issue.
Bed tension per se has nothing to do with the Y axis calibration, but the length of the belt would have an effect, and the belt can stretch to a more or lesser degree. When the printer gets old.  Not new, fresh out of the box, printers made from kits.
If those figures are right, then your shim is far too thick. As said,the forward y axis position just needs adjustment by 0.1 - 0.2 mm.

Posted by: @bid3d

huh...MK4 fails Y axis calibration test. Do the manual shimming on front of left rod, Y axis calibration test passes.

Print the shims, install the largest shim on rear of left rod, Y axis calibration test fails.

Remove this shim, place on front end of Y axis rod, U axis calibration test passes.


Why does the rear or forward position of the shim result in differing calibration test results ?

With shim in forward location Y axis calibration test values displayed during tests are -4 then +150

With shim in rear position Y axis calibration values displayed during tests are -4 then +211

Two questions, in an attempt to understand the calibration test the test failure.

How would belt tension affect overall Y axis travel distance (assuming belt tension is the culprit behind the Y axis calibration test failure?)

Why is the same value of -4 displayed when the shim location is in either the front or the back location (I would assume rearward travel with the shim in the rear position would alter the -4 value in the positive direction)


Posted : 22/03/2024 11:11 pm
Estimable Member
RE: y and z self test failure after update to 5.1.0

This everything can be the combination of all of them. When at the calibration start the bed is close to the rear end the bump is not that strong.  It can be also related to the belt tension - the loosen belt can act as more flexible/ less rigid.  Next factor can be the changed stallguard sensitivity. Last but not least is the distance between the two bearings (as my post above). 

Posted : 23/03/2024 2:04 pm
Estimable Member

Prusa Research seem to have a culture of not admitting when they made a mistake.  The more complaints they get, the more they defend their decisions. They put up a wall of defence and do nothing. Particularly the case with this issue, for some reason.

And yet in the past they proudly proclaim when they make a concession and do something users asked for.

I think they cannot rightly claim that they are responsive to what users want.  They just do what they want to do, and then sometimes falsely claim users wanted it.

But for sure, nobody asked them to consolidate the self test routines for the Mini, Mk4 and XL into one thing (which "improvement" is the at core of this issue).
But judging from the operation of the self test routine, it no longer has functional integrity and a logical flow for the mk4.


This post was modified 1 month ago by bryn51
Posted : 17/04/2024 12:23 am
Page 4 / 4