Extruder suddenly reversing, restarting, reversing - so on. During print

What could be the cause of my mendelmax 1,5+ (ultibots)?

The extruder stepper motor can suddenly start going back and forth (maybe something like 1mm) continously for a minute or so. Then everything is ok again... until next minute.

I have tried:
Cleaning the extruder.
Reconnecting ALL the wiring, checking for bad connections.
Adjusting the pot on the stepper drivers (now set to 0,25V)

Please ignore the bad bridging (/thickness) of the top layer...

Did you check the gcode to see if it is a slicer problem or a machine problem?
Start there and let us know.

Sounds like retractions to me. That's the only reason I can think of why the extruder would reverse. Otherwize it may be a slicer problem.

Thanks for your reply!

I went into the gcode, edited it, so it started at the layer of problems.

Printed that layer:



So.. Its not the gcode. I do not see why it should retract at this stage.

Other ideas?

Edited 1 time(s). Last edit at 09/06/2015 04:52AM by Itek.

If the stepper motor makes a knocking noise as it reverses, then it's skipping steps. Check that your motor is strong enough for the task, and that it isn't absurdly difficult to manually push filament through the hotend.

You might also need to calibrate your extrusion. If your printer thinks it's extruding 100mm when it's actually pushing out 150mm then it would be likely to jam the extruder/hotend.

Edited 1 time(s). Last edit at 09/06/2015 12:31PM by AlexY.

I have been printing for several 100 hours with this motor...

I do agree that the extrusion is a bit too high, Ill turn down the extrusion multiplier a bit. Thanks.

What if I turned the knob on the stepper driver board (ramps) too low? What would happen?

The temperature of the hotend also seems too low, this may solve the holes. You could try putting a dvm on the dir pin of the extruder. If it's told to reverse you can rule out hardware problems and troubleshoot firmware and gcode.

Also note that the generated gcode uses absolute positioning, when extruder is at say 100mm and then G1 E95 will retract. There are regular g92 e commands to reset it to zero.

Too low temperature? I do not agree. The filament is too much liquid to make bridges, that´s why it wont print properly over the infill in the top picture.

When you say "You could try putting a dvm on the dir pin of the extruder". Do you mean a digital volt meter? Also, between what pins? DIR and ___ ? Ground?

I expect the extruder need to be warm when I do this, but does the printer have to be runnning?

Thanks for your reply! (:

Or too fast if it's too liquid it would more stringy, I guess. Pla or abs?

Yes, a digital volt meter. Run the gcode and whenever it reverses, see if the voltage changes. From dir pin to ground.

I think that you will find that the pulse rate to a stepper is far too fast to get any kind of a useful reading with a DVM . . . This falls more into oscilloscope territory . . .

- Tim

It depends on how long it reverses direction. He could also hook up a low power led with 1k resistor to ground, the point is to visualize what is actually happening. The dir pin isn't pulsing exactly, not like the step pin.

I also have to be there when it starts reversing..

I was lucky to wintness it in that 6hrs print above. It only happened 2 times, and it was 4hrs into the print.

I have been printing recently, with success.

The reversing thing isn't repeatable, you can't reproduce it at will? Those are the harder things to troubleshoot. I just fixed some network printers hanging the spooler service in windows only around the afternoon. It took nearly a week before I could find a reproducable workflow of the problem.

It could be usb/serial related. If you own a sd card and the printer has a reader, long prints may be more reliable to print.

Did you get a chance to try manually pushing filament through? It still seems to me that the problem is with the stepper motor not being powerful enough. It usually happens when heat creeps up the hotend due to inadequate cooling and causes a jam, which would be fixed with better cooling.

Edit: Also, do you have a video of the problem?

Edited 1 time(s). Last edit at 09/08/2015 11:05PM by AlexY.

Quote
imqqmi
The reversing thing isn't repeatable, you can't reproduce it at will? Those are the harder things to troubleshoot. I just fixed some network printers hanging the spooler service in windows only around the afternoon. It took nearly a week before I could find a reproducable workflow of the problem.

It could be usb/serial related. If you own a sd card and the printer has a reader, long prints may be more reliable to print.

No, I cant reproduce it at will
Yes It makes it harder. I always print from my SD card reader.

Thats why Im asking you guys. Im glad you found out of your problem.

Quote
AlexY
Did you get a chance to try manually pushing filament through? It still seems to me that the problem is with the stepper motor not being powerful enough. It usually happens when heat creeps up the hotend due to inadequate cooling and causes a jam, which would be fixed with better cooling.

Edit: Also, do you have a video of the problem?

No, I didnt try pushing manually.Yes. I agree the stepper motor doesn´t seem powerful enough. Does anyone know if I turn the pot on the stepper driver to higher Vres voltage, the stepper will be more powerful?

No video. Not sure If I got a camera that can shoot continously good quality video for 4 hrs or more....

You can try a webcam or ethernet cam. It would be very helpful to further troubleshoot the problem. Some recording software allows triggers to record video, you may be able to start recording after x hours? There are freeware video security software packages that can be setup for this kind of thing.

If you can't record video, maybe you can record the audio, maybe you can identify the problem by the sound alone. Skipping steps or fast back and forth motion can tell a lot what is happening.

Another idea is to hook up another thermistor onto the board and attach it to the stepstick chip or cooler. Repetier host will record the temperatures, maybe it is overheating after 4 hours and get some unexpected behavior. A fan could help in that case if you didn't use one already.

Maybe you can artificially push the printer to give this error sooner by turning the speed way up, maybe also increase acceleration. If it can show the error within minutes instead of hours would make it much easier.

I found this in another thread, this is when just extruding and is always happening, would this also illustrate your intermittent problem?
[www.youtube.com]

Just thought of something. Since it's not happening always and only after a long time unattended, could it be the filament being in a knot, the extruder motor is stalling? You could try blocking the filament and see if this replicates the symptoms. Just a thought.

Quote
Itek
No, I didnt try pushing manually.Yes. I agree the stepper motor doesn´t seem powerful enough. Does anyone know if I turn the pot on the stepper driver to higher Vres voltage, the stepper will be more powerful?

Yes, turning the pot clockwise will increase the stepper motor torque but make it run hotter. If you increase it too much the driver will overheat and restart, causing skipping as well. Try turning it a little to see if it helps.

Thanks for your replies!

Quote
imqqmi
I found this in another thread, this is when just extruding and is always happening, would this also illustrate your intermittent problem?
[www.youtube.com]

Wow! Yes, It looks exactly like that!

Quote
AlexY

Quote
Itek
No, I didnt try pushing manually.Yes. I agree the stepper motor doesn´t seem powerful enough. Does anyone know if I turn the pot on the stepper driver to higher Vres voltage, the stepper will be more powerful?

Yes, turning the pot clockwise will increase the stepper motor torque but make it run hotter. If you increase it too much the driver will overheat and restart, causing skipping as well. Try turning it a little to see if it helps.

Aha! I did turn the pot counterclockwise. If I understand correctly, this will deacrease the motor power. I think my stepper motor controllers are cold, they have their own heatsink, with thermal paste and a fan constantly blowing on them.

I´ll just start printing as normal again (I have an upcoming project 3D printing a racing quad).

Will come back to this thread if I have succeeded! (or not) (:


Thanks!

Quote
Itek
I´ll just start printing as normal again (I have an upcoming project 3D printing a racing quad).

If you're interested, I started a thread about my own 3D printed quad build, complete with a parts list: [forums.reprap.org]

Quote
AlexY

If you're interested, I started a thread about my own 3D printed quad build, complete with a parts list: [forums.reprap.org]

I guess you should have a look at my build thread:

Pictures, and google translate is your friend:
[fpv.no]!

Very nice!

Just curious, any reason why you didn't go for 6 inch props?

experience (:

I might go for 6inch later.

I am not sure it will with the porblem here but I experienced weird exruder behaviour some time ago as well.
Was during a complicated print when suddenly the snail-slow extruder started spinning so fast that it burnt through the filament and partially melted it into the extruder (fun to clean and fix).
The next test caused the same thing to happen but combined with the nozzle moving all the way to the x-endstop and back.
Mind you I was using Simplify3D, which should not cause such problems (or so I thought).
I nailed it down to communication problems.
The stuff mainly happened shortly after I manually decreased the print speed to aid with details at this layer height.
But also with some extremely slow print moves, like when I tested the bridging speeds.
The software floods the printer board with commands until the buffer runs over.
Using hardware flow control helped a bit but only after INCREASING the crawling speeds the extruder accidents stopped.

Quote
Downunder35m
suddenly the snail-slow extruder started spinning so fast that it burnt through the filament

Quote
Downunder35m
the nozzle moving all the way to the x-endstop and back.

Quote
Downunder35m
I nailed it down to communication problems.

This has happened to me as well lately! - Didnt think this problem was connected to the extruder skipping steps/ back-forth movement though. As some of these problems were gone after I reconnected all the wiring in the 3D printer.

I will turn up the voltage to the extruder stepper motor now and turn on a 6hrs print. Will let you guys know how it went.

Will be continued...

Edited 1 time(s). Last edit at 09/12/2015 04:59AM by Itek.

As I said I could not find any other reason than the flooding of commands causing the print to go haywire.
Simply check the speed setting after the next failure and increase the slowest speed a bit, or all if you dare.
I no longer go under 20mm/s for any movement and no problems ever since.

My lowest speed possible is 30mm/s
If the layer is too small, the extruder will pause.

Successful 6hrs print!

I also might have found the (or more than one?) errror. The extruder would not heat when I turned the heating on! - Just a tiny movement on the wire, and it started heating...

You might want to replace the heater cartridge if the wiring isn't reliable to protect your (more expensive) electronics.

Well, no heat is a good explaniation for extrusion problems ROFL
But as Alex said, inspect the wires and if doubt replace the heater.