E3D Roto + Duet Roto Toolboard - Stepper motor stalling

arturolaz

@dc42 My bad, there is no connection between them as the way I created the fan shroud requires the mounting screw not touching the toolboard but instead coming from the shroud. I have now read the wiring notes from the docs and will be trying it.

How can this cascade to a motor stall?

dc42

@arturolaz extruding filament builds up static charge. If it is allowed to build up on the extruder motor body, it may flash over to the stepper motor wires. From there it flows into the stepper motor driver and may trigger the short circuit detection in the driver.

See also this thread https://forum.duet3d.com/topic/37882/extruder-stopping-mid-print-fixed-after-reboot.

arturolaz

@dc42 Understood. I did the test grounding the metal body of the Roto through a wire soldered to the V- terminal of the power connector in the toolboard with a 100k serial resistor but it didn't work, it failed just like before even without reaching 70C

arturolaz

@dc42 I've been reading about CoolStep and I think it is the next step, unless you have another idea for me to try before.

Are there any special considerations when implementing CoolStep for the extruder? Most of the documentation is oriented for the other motors

arturolaz

I haven't been able to conrrectly configure stall detection for the extruder, I just found this post from some years ago https://forum.duet3d.com/post/102570 so I'm running out of options if CoolStep can't be implemented...

dc42

@arturolaz I doubt that CoolStep is the solution because I am not convinced that the driver is overheating. How quickly after the driver stopped extruding did you take the M121? If the driver really is shutting down because of overheating then it will cool down fairly quickly.

BTW your M121 report says that the driver on the tool board is stalled:

Driver 0: pos 0, 2682.0 steps/mm, stalled, SG min 0, temp 69.9C, read errors 0, write errors 0, ifcnt 52, reads 40354, writes 0, timeouts 0, DMA errors 0, steps req 0 done 251658

If you haven't set the stall detection threshold using M915 then that doesn't necessarily mean that it has stopped moving, but it does suggest that it was under high load when last commanded to move.

arturolaz

@dc42 I sent the command as soon as I realized it was failing, 2minutes difference top, at that point I hadn't enabled anything related to stall detection.

Quick update of my tests:

Disconnected the Roto motor from the Roto toolboard and rewired it to use a Duet3 stepper driver - Still failing
Printed the exact same test print but now with PLA hoping that the ABS ambient temperature was the trigger of the motor failure but it also failed
Inspected the whole motor assembly but found no issues apart from maybe insufficient grease on the gears (waiting for the grease to test)

At this point I'm almost sure the motor or some configuration is the problem but I can't find the root cause of it

dc42

@arturolaz what mode are you running the driver in? Send M569 P121.0 to check.

arturolaz

@dc42 these are the current settings:


M569 P121.0
Driver 121.0 runs forwards, active low enable, mode spreadCycle, ccr 0x00053, toff 3, tblank 0, hstart/hend/hdec 5/0/0, pos 8

dc42

@arturolaz said in E3D Roto + Duet Roto Toolboard - Stepper motor stalling:

How can the CPU temperature can be calibrated?

Turn the printer off and give it time to reach ambient temperature. Read the ambient temperature from a thermometer. Turn the printer on and immediately use M122 to get a report from the board of interest. That report includes the CPU temperature. Compare it with ambient and calculate the correction required.

Currently, only the main board CPU temperature calibration can be stored, using the M912 command.

arturolaz

@dc42 the Duet3 control board is not installed inside the closed chamber as the roto toolboard is. How can this calibration affect the outcome of the printing tests?

Is there anything off in the M69 P121.0 report?

arturolaz

For anyone following.

I just measured the coils resistance in "hope" for them to be out of spec but they are not. Both measured 7.6 ohms at room temperature.

I haven't been able to print ABS nor PLA. I will keep trying tweaking the configuration but I feel that I have tried everything.

elmoret

@jay_s_uk said in E3D Roto + Duet Roto Toolboard - Stepper motor stalling:

@arturolaz there is this as well about the Roto itself

That's a PLA limit, to avoid jams in the hotend.

Notepad

@elmoret Can confirm, E3D testing is always a "worst case" recommendation. The max motor amperage is only if they are pushing ultra high flow high pressure, in a 50+C heated chamber, for extreme lengths of time.
You may want to raise your amperage as that would give you more force available.

arturolaz

@elmoret This problem started while printing ABS but I have since reverted to testing with PLA with the same results, the motor keeps stalling after 15-20 minutes of printing when coming from a cold start and sooner on later prints.

@Notepad I have also tested with the absolute max current E3D recomends, 0.6A peak with the same result, I think anything beyond that might risk damaging the coils.

I will open a case with E3D to be sure the motor is not damaged at this point.

infiniteloop

@arturolaz

the motor keeps stalling after 15-20 minutes of printing when coming from a cold start and sooner on later prints.

Looks like a classical heat creep. Initially, this might have happened at high temperatures when printing ABS. Then, with PLA, remainders of the ABS can partially block the filament path so that the melt zone expands upwards…

Have fought with that before, cleaning the filament path had only a temporary effect. In the end, I had to swap both the nozzle and heat barrier. Finally, I improved the cooling (of the cold end of my hot end ), which was the root cause of my problem.

arturolaz

@infiniteloop I hadn't thought of that, I will try a fresh nozzle and upgrade the fan, thanks for the suggestion

arturolaz

@infiniteloop I managed to try a fresh 0.4 nozzle and replaced the original extruder cooling fan with an slightly bigger one, still not working.

Also, I saw on the TMC2240 datasheet that Spreadcycle can compensate to some degree for the back EMF coming from the stepper, switched to that mode but didn't work as well

More ideas are appreciated before I take the leap and go for an smart orbiter.

Phaedrux

Can you share some photos of the extruder setup to help with context?

arturolaz

@Phaedrux For sure