X-axis shifting - out of ideas!  

Page 1 / 2
  RSS
Newk
 newk
(@newk)
Active Member

Hi from Germany,

I'm out of ideas and things to try.

Before I upgraded my i3 MK3s with the MMU2S I've had a very reliable machine. Since then I've only had problems over problems...

The latest in a long line of issues is the X-axis shifting at roughly the same Z-layer height. This problem is reproducable (I've tried 3 times now with this model and a lot of other models before).

I've printed from SD to cancel out issues from octoprint. I've cleaned and lubricated everything again and again. The belt tension is 258 for X and 264 for Y which should be ok.

As far as I can tell there is no mechanical problem in my enclosure. Nothing gets in the way ...

For context here are a few more pictures of my setup:

This topic was modified 2 months ago 2 times by Newk
Posted : 26/11/2020 1:08 pm
cwbullet
(@cwbullet)
Famed Member

Have you tried everything on:

https://help.prusa3d.com/en/article/layer-shifting_2020

--------------------
Chuck H
3D Printer Review Blog...
Posted : 26/11/2020 5:30 pm
Newk
 newk
(@newk)
Active Member

Yes, I‘ve checked everything mentioned in any official documentation several times. I‘ve spend hours on troubleshooting already.

Posted : 26/11/2020 5:53 pm
Peter M
(@peter-m-3)
Prominent Member

Wen printing , check if nozzle is hitting model, could be a small blob on model, over extrusion or leakage or printing to hot.

Posted : 27/11/2020 12:47 pm
cwbullet
(@cwbullet)
Famed Member

Also, look at the grub screw one more time.  

--------------------
Chuck H
3D Printer Review Blog...
Posted : 27/11/2020 7:26 pm
Loak
 loak
(@loak)
Estimable Member

Hi from France,

My 2 cents : overheating of the motherboard (or its steppers) as the printer is in an enclosure ?

MK3s Stock, PS2.2 PS2.3 test, Fusion360, Debian, Windows 10....
Posted : 27/11/2020 7:36 pm
cwbullet
(@cwbullet)
Famed Member

@loak

Highly doubt that.  Mine has been running for over 2 years in a much tighter enclosure.  

--------------------
Chuck H
3D Printer Review Blog...
Posted : 27/11/2020 7:39 pm
Agkirk
(@agkirk)
Trusted Member

That is one nice hardware setup! Wow

Posted : 27/11/2020 8:24 pm
Walleye
(@walleye)
Eminent Member

Hi,

Yeah, that power supply inside the enclosure can be an issue. Heat degradation is an issue with power supplies in hot environments. Even the controller board is not supposed to be run over 40C to prevent damage. And the power supply can be damaged at lower temps.

Try printing with the doors open to see if heat build up is an issue. But remember that heat damage can be accumulative as well as instant.

Posted : 27/11/2020 9:34 pm
cwbullet
(@cwbullet)
Famed Member

@walleye

How would that cause and X shift?  I am just trying to wrap my head around that.,  

--------------------
Chuck H
3D Printer Review Blog...
Posted : 27/11/2020 9:50 pm
Walleye
(@walleye)
Eminent Member

@cwbullet

Hi,

By causing varying power out put from the supply. And "dirty" power will cause issues with the control board. I've seen it on $100,000 industrial machines where dropping .5V in output can cause problems. Heat is the enemy of electronics and they have a hard enough time shedding heat on their own without trapping them in it.

Even Prusa states that when using an enclosure that the power supply must be relocated outside of the enclosure and to not exceed 40C inside for the control board. They even recommend a ventilation fan to pull excess heat out if you are running higher temp filament like ABS or Nylon.

Running a print with the door open is a cheap and easy test. Mounting a thermometer inside your enclosure will also let you see just how hot it gets in there. If the issue goes away, that's a pretty good indication of excess heat.  Though it might not be definitive if the axis shift continues because there might be permanent damage already done.

Posted : 27/11/2020 10:20 pm
fuchsr
(@fuchsr)
Estimable Member

Interesting thread. After a long time of perfect prints I am suddenly seeing x-axis layer shifts as well. Of course I went through the usual trouble shooting -- belt (check), lubrication (check), grub screws (check) -- nothing.

For me though the layer shift is unpredictable. I had models where it happened after a few mm's, others were 14 hour prints with a shift close to the end. Confusingly, what I have also seen is a second shift in the opposite direction, apparently RIGHT back to where the model is supposed to print, so I end up with a model that looks almost normal but in which somewhere in the middle a region is shifted to the side. Can't figure out for the life of me why a second shift would be the exact opposite of the first.

I'm going to take the whole x-axis apart, just waiting for the SuperPINDA to arrive so I have to take things apart only once.

 

This post was modified 2 months ago by fuchsr
Posted : 27/11/2020 10:28 pm
cwbullet
(@cwbullet)
Famed Member

@walleye

I must have a magical board that has function well for 3 years at 35-45 celsius or warmer with zero shift.  I could see this with temperatures higher than that but not at the temperatures most hobbyists are seeing.  I will take your word that it can happen, I just don't understand why I am not seeing it.  

--------------------
Chuck H
3D Printer Review Blog...
Posted : 27/11/2020 10:37 pm
Swiss_Cheese
(@swiss_cheese)
Reputable Member

I'm with Loak & Walleye on this one, Newk's problem may not turn out to be that, however I've seen exactly what Walleye is describing on several occasions from my Heathkit Days long ago until now. heat creates resistance, varying/uncontrolled resistance mixed with electronic = No Bueno

I may be mistaken about this, but I think resistance is how the steppers determine a collision.

This post was modified 2 months ago by Swiss_Cheese
Posted : 27/11/2020 11:40 pm
cwbullet
(@cwbullet)
Famed Member

I have no doubt that heat can cause poor electrical performance, but most of these enclosures raise the temperature a measly 14 C.  That does not seem like enough.  22 C to 36 C.  In my intro to electrical engineering, the board should function well up to 110-120 C.  

This post was modified 2 months ago by cwbullet
--------------------
Chuck H
3D Printer Review Blog...
Posted : 28/11/2020 12:09 am
karl-herbert
(@karl-herbert)
Famed Member

Another option that can trigger layer shifting is a faulty data transfer from the SD card or via the print server, e.g. Octoprint. Additionally I would also check the voltage of the power supply.

 
Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen....
Posted : 28/11/2020 12:24 am
Newk
 newk
(@newk)
Active Member

Thank you guys so much for all the replies!

I don't think temperature is a problem. I've printed faulty prints with the doors open. I've also have ventilation in the enclosure which keeps it at 30°C.

The problem really is, that it does not happen constantly. With some models it does and it drives me nuts, but over night I printed these tests for example and (apart from the layers not being so pretty) there was no problem with shifting (the cut in the middle of the model is intentional). But this print also only took 4-5 hours.

Posted : 28/11/2020 7:19 am
cwbullet
(@cwbullet)
Famed Member

@newk

That looks ok to me.  Some banding, but it is ok

--------------------
Chuck H
3D Printer Review Blog...
Posted : 28/11/2020 11:00 am
Baklin
(@baklin)
Trusted Member
Posted by: @fuchsr

For me though the layer shift is unpredictable. I had models where it happened after a few mm's, others were 14 hour prints with a shift close to the end. Confusingly, what I have also seen is a second shift in the opposite direction, apparently RIGHT back to where the model is supposed to print, so I end up with a model that looks almost normal but in which somewhere in the middle a region is shifted to the side. Can't figure out for the life of me why a second shift would be the exact opposite of the first.

Could it be that you had a crash detected after the first shift? That would move cause the printer to home and start again in the right place like that first shift never happened.

Posted : 28/11/2020 11:10 am
Newk
 newk
(@newk)
Active Member

@baklin

Crash detection is giving me problems all around from the beginning to be honest. It's not working reliably and triggers a lot of times without any obvious reason.

So yes, I will try the same model without crash detection right now. I will report back tomorrow! Thanks!

This post was modified 2 months ago by Newk
Posted : 28/11/2020 1:25 pm
Page 1 / 2
Share:

Please Login or Register