Notifiche
Cancella tutti

octoprint 1.3.8 causing stepper motor failure???  

  RSS
JuanCholo
(@juancholo)
Honorable Member
octoprint 1.3.8 causing stepper motor failure???

i'm having a really weird situation could use some feedback.

equipment list: 3 Prusa printers; 2 MK2S , 1 MK3 printer.
MK2S printer #1 is over 1 year old and was running perfectly.

MK3 printer #2 is about 3 months old and has its issues that prusa needs to work out.

MK2S printer #3 is less then 1 month old i put it together from a kit.

All printers ran normal until 1 week ago.

10 days ago i decided i wanted to start a print farm and i setup Octoprint over a couple of raspberry pies.

a few days into the setup Printer 3 (MK2S) Suddenly has a bad failure after running fine off the SD card as i had it running off octoprint.
the printer had made spaghetti on a print and i figured there was some clog and i went to reset it without watching it.
the nozzle rammed into the bed and damaged it. trying to figure out what happened i noticed the Z axis was Skewed bad to the right side. i reset it and that was when i found out my X-AXIS had failed. i worked with tech support and they were suppose to send me a new motor and the bed because the failure didn't make sense.
now the fact that the Z was skewed bothered me, but i tested the Z axis and it moved up and down fine.
so this was one week ago the printer is down.

tonight Printer #1 MK2S failed on a print making spaghetti once again as i was running octoprint . this printer has been a rock.
so when the failure occured and it looked just like printer 3's failure i took notice. i reset the machine and noticed Z was Skewed.
i tried to run a XYZ calibration and slowly watched as the Z axis stepper on the right side started to index or back step causing the Z axis to skew.

now i'm putting things together i had no issues running from the SD cards on the 2 machines for quite some time, i setup octoprint and run a few prints from local and not SD and suddenly i have multiple Stepper motor failures????

i took my Z axis motor connectors out and checked them, i did the LED tests on the connectors and both phases are working. i tried to swap the connectors and run XYZ again and once again right motor was back stepping.

i'm at a real loss here, because these failures really seem to be happening with octoprint in control of the printers. I have now lost 2 printers and i'm pretty nervous.

can anyone also give a link to the correct stepper motors that prusa uses.
i measured them and got:

1.8 deg 6.2OHM 42mm 2 phase 4 wire

1124007

42BYG40B-18W26

but i can't find them anywhere.

🙁

“One does not simply use a picture as signature on Prusa forums”

Postato : 24/04/2018 3:38 am
JuanCholo
(@juancholo)
Honorable Member
Topic starter answered:
Re: octoprint 1.3.8 causing stepper motor failure???

stilling working on it here.

tried swapping motors around again on Z always the Right motor from looking at the machine from the front lcd.

I attempted to reflash the firmware to the latest version and run the selftest and XYZ cal again.

Selftest passes. XYZ is ok on point 1 of 4 when it goes to point 2 Right Z axis starts to back step you can feel it making a certain noise and then the X carriage twists.

At this point i'm aborting any octoprint plans and as soon as my print completes on the MK3 through octoprint i will be shutting it down and going back to SD.

still have no idea what exactly is damaged now on 2 mk2s printers.

“One does not simply use a picture as signature on Prusa forums”

Postato : 24/04/2018 4:24 am
JuanCholo
(@juancholo)
Honorable Member
Topic starter answered:
Re: octoprint 1.3.8 causing stepper motor failure???

ok i took MK2S printer #3 apart for the Z axis stepper motors i swapped just the right side into MK2S printer #1 (that just failed tonight).
Printer #1 is now passing the XYZ calibration so the Z-axis stepper motor on the right side failed in some way.

At this point 2 printer failures in just a few days with stepper failures is telling in the gut something in octoprint is causing it. i've been in software and hardware too long not to see a connection here. Possible mis-timing between the octoprint and rambo as i was not watching Terminal to see the code and Ok messages going back and forth but something locked this stepper motor up and did the same on printer #3 (mk2s) last week. Fool me once, shame on you. Fool me twice, shame on me.

now this really sucks because i bought 3 raspberry pi and set them up with the printers and a 4th pi to run VNC to control the printer controllers with a little more security. but i will now be shelving the entire thing and i was having a blast checking the printers via my cell phone away from the printers today. plus i bought Logitech cameras to monitor and make Time lapse movies.

“One does not simply use a picture as signature on Prusa forums”

Postato : 24/04/2018 5:00 am
JuanCholo
(@juancholo)
Honorable Member
Topic starter answered:
Re: octoprint 1.3.8 causing stepper motor failure???

printer #1 is complete and back online. skew is 0.01. the USB data cables have been removed and octoprint will be removed from the equation for now.
i suspect it is communication timing issues between octoprint and the rambo. if the machine missed some Zhops it would explain how it got so messed up and blew a motor. i read some interesting things happening to other people and different 3d printers using octoprint. people talking about stepper motors suddenly overheating.

i know there have been communication issues in the past and that was part of the reason on the MK3 linear advance was temporally turned off in firmware.

I had some part anomalies when using octoprint that i again suspected were caused by either slow communication or missed steps. i will be re-running the same parts now to see differences.

So be careful. i don't know 100% that octoprint caused my steppers to fail but with everything happening from the time i started to use it till the next failure the odds sure dropped to 1:1.

“One does not simply use a picture as signature on Prusa forums”

Postato : 24/04/2018 6:14 am
JuanCholo
(@juancholo)
Honorable Member
Topic starter answered:
Re: octoprint 1.3.8 causing stepper motor failure???

Another thing to add I just realized.

I ran both mk2s printers at the time of stepper failure with linear acceleration.

I forgot all about that. Could of contributed with the documented LA issues and serial communication.

“One does not simply use a picture as signature on Prusa forums”

Postato : 24/04/2018 2:58 pm
JuanCholo
(@juancholo)
Honorable Member
Topic starter answered:
Re: octoprint 1.3.8 causing stepper motor failure???

i've read up a few papers on stepper motor failure and asked a few people on other forums.

99% of the people are running octoprint on a PC or mac, very few people by comparison are running it on raspberry pi.

the people on the pi have older machines usually and have different gcode handling. Prusa seems to be the odd man out in that the data has to be handled by octoprint differently to make it work. older printers are slower so they are less likely to have a sync issue. today with printers getting faster the need for real time sync is important.

most stepper motor failure is caused by the drivers very rarely is it mechanical. at this point i have high confidence that a timing issue in communication between the Pi and rambo caused the motors to be improperly index over and over and lead to premature failure of the coil phases.
there can also be studdering issues with the USB ports on the Pi where they get out of sync the B+ being faster was suppose to reduce this.

a modern PC is going to be much faster then a Pi so the chance of studdering or sync issues is small but still possible if some resource bogs down the PC. this is similar to an issue in the late 1980s when CD roms first started to appear and they would studder and skip and ruin discs because the pc was not fast enough to keep the buffer full for the writer. appears to be the same here.

So if you are running octoprint from your pc directly to the printer the chance of a problem is low. if you are running off a PI like i was you have a potential to cause harm to your stepper motors at least where a prusa is involved.

“One does not simply use a picture as signature on Prusa forums”

Postato : 24/04/2018 6:32 pm
Nikolai
(@nikolai)
Noble Member
Re: octoprint 1.3.8 causing stepper motor failure???

Hello Daniel,

I'm running Octoprint most recent version (currently 1.3.8) on the Raspberry Pi 2B since one year. Had no issues so far with it and MK2S. I don't see any difference between printing from SD card or Octoprint.

Using Octopie can't do any harm to your stepper motor. At the end the printer receives commands either from SD or USB, it doesn't make any difference in case of command execution.
Your failures are either hardware, user or software related (if you modified GCODES manually).

Feel free to post your finding in the Octopie forum, in case I'm missing something.

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram

Postato : 04/05/2018 1:41 am
JuanCholo
(@juancholo)
Honorable Member
Topic starter answered:
Re: octoprint 1.3.8 causing stepper motor failure???

well i'm warning people anyway. if you are using octoprint on a pi and suddenly start having Stepper failures then feel free to post.

I had 2 separate MK2S machines fail while running octoprint in the span of 5 days.

both machines were problem free before hand.

the final dead stepper motor count was 3.

2x Z-axis failures on 2 machines and 1 X-axis failure caused by one machines Z-axis failure which also damaged the Bed.
Prusa support thankfully helped me out. i told them what happened.

a MK3 was running on octoprint at the same time and thankfully was not damaged. both MK2S machines were LA enabled and Running LA prints at the time of failure.

the prints themselves were missing steps on layers causing the layers to look like swiss cheese in some spots which you could visually see with a loop. this was not caused by an extruder clog. i believe it was related again to octoprint and a timing issue on my PI with the USB ports. it could of easily effected the extruder stepper during this point.

i repaired the mk2s machines and they are back to printing normally and have been running for over 100 hours since the octoprint failure.

me i will not be using octoprint in the future at least not on a PI.

hopefully others do not have to go through what i did.

i was running 1.3.8 waiting for 1.3.9 because of the Stop print bug. i got rid of everything after the 2 machine failures. slicing was handled with Slic3r so no manually edited codes. both failures also occured only when loading a Gcode into the machine from octoprint. Before this i was using octoprint to tell the printer to pull from the SD slot. the difference on the LCD was the machine indicating it was pulling data from the USB instead of SD card.

i posted here because it was only the MK2S machines which failed. Yes i did like the features of octoprint on a PI and wanted it to work. it just didn't work out and i'm not willing to dedicate multiple PCs to job.

“One does not simply use a picture as signature on Prusa forums”

Postato : 08/05/2018 9:41 pm
john.v11
(@john-v11)
Estimable Member
Re: octoprint 1.3.8 causing stepper motor failure???

I only used octopi on my MK2 for a relatively short time but I have over 2000 hours on my MK3 with more than 90% of the prints being via a raspberry pi3 over USB. No stepper motor failures here.

All the octopi system does is feed gcode, one line at a time as if the MK3/MK2 was reading from the SD card.

IMHO, I know you think you are trying to help people but you just don't seem to have the facts to back up your conclusion.

Postato : 08/05/2018 9:52 pm
JuanCholo
(@juancholo)
Honorable Member
Topic starter answered:
Re: octoprint 1.3.8 causing stepper motor failure???

that is fine also this is for mk2s machines i stated the MK3 i have was running octoprint also and did not appear to sustain damage.

there is obviously more going on here involving MK2 machines with possible LA enabled and PI usage with USB transfer of data.

there are other forums documenting issues with problems on the USB ports on PIs not involving octoprint.

take note i also did not use a B+ model which is suppose to have addressed some USB issues. this was a 2015 B model i was using.

I did not use a PC or mac in this case which would of most likely never caused a problem.

the failures appear to have effected Z-axis motors on both machines the rambo was checked after wards as well. both machines are running normal now from the SD slots.

Take it as a warning of a possible problem.

“One does not simply use a picture as signature on Prusa forums”

Postato : 08/05/2018 9:59 pm
JuanCholo
(@juancholo)
Honorable Member
Topic starter answered:
Re: octoprint 1.3.8 causing stepper motor failure???

"The octoprint issue has a thread on github and it boils down to the custom gcode that modulates the stepper drivers takes longer to process than octoprint expects. The "fix" involves either tweaking octoprint's settings to allow that command more time, modifying the gcode to alert octoprint of the delay, or simply running that test file from the SD card as other normal printing shouldn't cause a problem."

this explains the problem happening to me when i was using octoprint and switched from printing from the SD to octoprint controlling the print fully.

this is why my motors stuttered and caused a failure of 2 motors on two different machines.

“One does not simply use a picture as signature on Prusa forums”

Postato : 21/05/2018 6:29 am
Condividi: