Notifications
Clear all

Nozzle height cycling with 30 second period?  

  RSS
epzcaw
(@epzcaw)
Estimable Member
Nozzle height cycling with 30 second period?

I bought an Original Prusa i3 MK3S kit which I assembled successfully (after a few hiccups) and starting printing successfully in late August last year.  However, a problem soon emerged.  This was an intermittent fault where extrusion stopped some time during the print process though the printer continued to move round.  I had many “chats” with Prusa and also with forum members – see “Mulitple print failures” post.  No permanent solution was found.

In February, the printer was returned to Prusa for repair and service with the aim of eliminating the fault.  They found some faults which were repaired (though nothing that seemed to explain the intermittent fault) , and the printer was returned to me on the 25th March.  They enclosed some beautiful parts that had been printed with it.

I did a self test (ok) an and XYZ calibration (OK), but could not get a satisfactory first layer calibration. I had long “chats" with Prusa on Thursday and Friday, but was no further forward. I an printing a single layer rectangle (gcode sent by Prusa) and noticed that this appears to show a cyclic variation in the nozzle height even though that has been set at a specific value.

I did three rectangles in succession with the same fixed z value and you can see that the same pattern appears in each. I measured the time between the start of each “fringe” as about 30 seconds. A friend suggested that I should try locating the part at different parts of the heat bed, so I made a gcode with 5 rectangles at different positions and orientations. The rectangle which is at 45 degrees has a much more closely spaced pattern but the time between “fringes” was again about 30 seconds

As the print proceeds, the nozzle seems to get jammed up with unprinted material, so the print is not completed.

So it appears that the distance between the nozzle and the heatbed is oscillating with a period of about 30 seconds.

I assume something has been disturbed when the printer was in transit.  I will “chat” again, but in the meantime, I wonder if anyone has any ideas about what the cause of this could be.

Thank you.

A soft answer turns away wrath.

Posted : 29/03/2020 12:18 pm
Peter M
(@peter-m)
Noble Member
RE: Nozzle height cycling with 30 second period?

-Bed need to be very clean, this is very important, clean with dish soap(3 times), then try to print, do not use alcohol.

-Are you using the settings for your printer, do not use the mmu settings.

-Adjust the first layer, possible a little lower.

-Let us see pictures, settings , pla? abs?, give us more info.

Posted : 29/03/2020 12:42 pm
epzcaw
(@epzcaw)
Estimable Member
Topic starter answered:
RE: Nozzle height cycling with 30 second period?

The printer has just been returned from Prusa so should be in good order.  I have cleaned the bed several times.

The gcode was sent to me by Prusa so I can't (and don't want to) do any adjustments to it.  I am using Prusa PLA.

I tried a whole range of z values on the first layer calibration program and nothing works.  This is because, I believe, that this is is not maintained by the printer at a constant height but is instead moving up and down at a frequency of one cycle every 30 seconds.   There must be some explanation for why this is happening.

A soft answer turns away wrath.

Posted : 29/03/2020 1:59 pm
Jimbo70
(@jimbo70)
Estimable Member
RE: Nozzle height cycling with 30 second period?

To my eye, it looks like intermittent and very cycle decreased extrusion. Could the extruder gear be slipping on the shaft? Is there more tension in the filament at certain times? Maybe a sticking filament roller or filament binding on the roll?

Posted : 29/03/2020 6:59 pm
epzcaw
(@epzcaw)
Estimable Member
Topic starter answered:
RE: Nozzle height cycling with 30 second period?

Yes, there is certainly something cycling here with a cycle time of 30seconds.  It has just been returned to me from Prusa where it was repaired and services and where they printed several nice models which they sent to me, so this cannot have been happening there. So what has happened to it in transit that has made it do this?

A soft answer turns away wrath.

Posted : 29/03/2020 7:04 pm
nilok
(@nilok)
Estimable Member
RE: Nozzle height cycling with 30 second period?

Are you using the stock dual filament holder?  Also, what is the printer mounted on?  Carpet floor, concrete, on a shelf, on a dresser, on a table?

Also, what was your LiveZ value for those prints, and does it do that during the LiveZ calibration?

This post was modified 5 years ago 2 times by nilok

Is your printer grinding? Bearings squeaking? Motors smoking?
Step 1) Wash your Steel Sheet
Step 2) Return to Step 1

Posted : 29/03/2020 7:08 pm
nilok
(@nilok)
Estimable Member
RE: Nozzle height cycling with 30 second period?

PS - before like 8 others show up and say so......clean that sheet!!

Is your printer grinding? Bearings squeaking? Motors smoking?
Step 1) Wash your Steel Sheet
Step 2) Return to Step 1

Posted : 29/03/2020 7:12 pm
bobstro
(@bobstro)
Illustrious Member
RE: Nozzle height cycling with 30 second period?

You've checked the filament feed path for friction and binding? Check the extruder idler gear for free rotation as well.

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 : 29/03/2020 7:13 pm
epzcaw
(@epzcaw)
Estimable Member
Topic starter answered:
RE: Nozzle height cycling with 30 second period?

@nilok

I've cleaned the sheet several times.  Soap and water, and iso-propyl alcohol

This post was modified 5 years ago by epzcaw

A soft answer turns away wrath.

Posted : 29/03/2020 7:32 pm
epzcaw
(@epzcaw)
Estimable Member
Topic starter answered:
RE: Nozzle height cycling with 30 second period?

The printer is in exactly the same location and environment as it was during the many hours of successful printing that I did since last August.

The Prusa "chat" agent had me check the gears - I even videod them and sent to him

It is not grinding, smoking or making any unusual noises.

Yes, the problem occurs during the first layer calibration.  A single z value does not give rise to a uniform pattern but shows the same sort of variation as in the rectangle. I am using the rectangle as a test piece on the advice of the Prusa agent.

I think I was hoping that someone would come up with a simple "Eureka" solution - a miracle??

I spent hours and hours following various suggestions trying to solve the previous fault, to no avail. I thought that by paying Prusa to repair and service it, I would be starting again with a new fully functional printer.  I've lost faith, and I think the only satisfactory solution now is to send it back again.

Will have another Prusa "chat" tomorrow and will report back.

Thanks all.

A soft answer turns away wrath.

Posted : 29/03/2020 7:42 pm
epzcaw
(@epzcaw)
Estimable Member
Topic starter answered:
RE: Nozzle height cycling with 30 second period?

Prusa has decided to send me a replacement printer.  They are as baffled as I am.  A friend has suggested that I may be detecting gravitational waves.  So maybe a Nobel prize.......

A soft answer turns away wrath.

Posted : 04/04/2020 6:20 pm
tj
 tj
(@tj-2)
Trusted Member
RE: Nozzle height cycling with 30 second period?

🙂

I'm curious to hear if you eventually solve this problem and what the source of it was.
My guess is that it is some mechanical problem causing filament to not feed properly, since the bad stripes changes width in your test prints it seems like after X amount of filament has been fed thru the extruder you run into problems.

A wild guess, is it possible that the filament catches on something periodically?
I once had filament catch on the prusament spools where you lock the filament in place around the edges, filament was pulled over the edge of the spool and eventually got stuck.
I know this is a long shot and unlikely to be the problem because if it was you would probably have heard the extruder grinding on the filament and already fixed it.

Posted : 07/04/2020 8:55 am
epzcaw
(@epzcaw)
Estimable Member
Topic starter answered:
RE: Nozzle height cycling with 30 second period?

@torbjorn-j3

Its an interesting suggestion but the pattern is much too consistent - everything I try and print shows this 30 second cycle.  I'm sure I would have noticed the filament jamming.

I videoed the extruder shaft while it was printing and it rotates smoothly and continuously.  I put a small piece of paper with a line onto the shaft and estimated the time when it reached 90, 180, 270 360 degrees:

No sign of any jamming!

But thanks for thinking about it!

A soft answer turns away wrath.

Posted : 07/04/2020 9:11 am
Share: