Stepper motor lost torque
-
@bearer yeah I know, when active it feels a little stiff (I know this normally feels a lot harder to turn than this) and almost rough but completely smooth when off for both. Tearing my hair out at the moment as no idea what's going on.
-
So if you disconnect the motor is it the same? If you disconnect and short out the coils is it the same? Sounds like something is shorted, and those two should help point towards bad wiring or bad drivers.
-
@bearer nope, when disconnected the motor spins freely but when powered it's got such low holding torque I can turn the Shaft by hand with much lower amount of force to normal
-
@oliverracing said in Stepper motor lost torque:
@bearer nope, when disconnected the motor spins freely but when powered it's got such low holding torque I can turn the Shaft by hand with much lower amount of force to normal
What current are you running the motor at? Can you provide a link to the data sheet?
Send M906 without parameters to check that the current is what you think it is. Also M913 to check that it is at 100%.
-
I'm running the motor at 900ma - and have been for ~2 years, Tried M906 to set the current up to 1500ma just to check if that made any difference (definitely added torque, just not much). I'll try M906 and M913 to check tonight - at work at the moment so hold fire on this.
As for specs of motor - they are the basic ones that originally came on the Tevo Tarantula that my printer once was based on. I have also tried a similar spec motor fresh from the box and this shows the same issue making me think it's unlikely to be a motor issue. I've made a fresh wire, so making that unlikely and tested the new and old (in that order) on the E1 driver and that had the same issue.
Can't work out how to link so here are the specs...
Drive Shaft - - - - - - - - - - - - - - - - - - - - - - - - - - - - 4.5mm
Step Angle - - - - - - - - - - - - - - - - - - - - - - - - - - - - 1.8°
Step Angle Accuracy - - - - - - - - - - - - - - - - - - - - 1.8 ° ± 5%
Holding torque - - - - - - - - - - - - - - - - - - - - - - - - 420mN.m(I=1.5A)
Rated Current/Phase - - - - - - - - - - - - - - - - - - - 1.5A
Phase Resistance - - - - - - - - - - - - - - - - - - - - - - 1.6±10%
Voltage - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 2.4V
Inductance - - - - - - - - - - - - - - - - - - - - - - - - - - - 3.8mH±20%
Positioning Torque - - - - - - - - - - - - - - - - - - - - - 18mN.m REF.
Maximum load starting frequency - - - - - - - - - 1400pps
Maximum no-load operating frequency - - - - - 8000pps
Moment of inertia - - - - - - - - - - - - - - - - - - - - - - 57gcm
Ambient temperature - - - - - - - - - - - - - - - - - - _ -20 ~ 50 -
@oliverracing said in Stepper motor lost torque:
@bearer nope, when disconnected the motor spins freely but when powered it's got such low holding torque I can turn the Shaft by hand with much lower amount of force to normal
To recap, this was working fine but the fault suddenly started without you making any changes to the configuration files.
What you are describing (i.e. you can easily turn the motor by hand, even when the holding current is set to 1500mA) indicates that only one pair of coils is being energised. I can only think of three possible reasons for that. 1) The motor itself is faulty but you say you've tried another. 2) There wiring is faulty but you say you've replaced it all. 3)The driver is faulty but you say you've tried another driver.
One thing occurs to me but I don't know if it is possible, is that the stepper developed a fault which took out the driver chip. You then tried that motor on another driver but because of the fault, it instantly took out that driver chip too. @dc42 will know if that's possible.
If not, then it only leaves highly improbably scenarios. 1) The stepper you replaced it with is also faulty. 2) The wiring you replaced is also faulty. 3) The driver you swapped with is also faulty.
It's easy enough to test a stepper motor to see if both coils are healthy - just do an internet search which will reveal a number of methods depending on what tools you have available. Once you have established that you have healthy stepper motor, connect it to a known healthy driver - such as one of the axis drivers. If it doesn't work, then it has to be the wiring between the stepper and the driver. If it does work, then it has to be a faulty driver (or two).
-
@deckingman Spot on and and thank you for putting it far more elegantly than I ever could! I'll go though in that order to try and work out what's going on.
-
That didn't end well (the short to ground warnings). Pop and burning smell with this warning in DWC. Had previously checked M906 - came back with E900:900 and M913 with returned 100%. Unplugged and replugged all connectors to do with extruder to check and tried a 1mm extrude and pop.
For reference the e0 driver now has a little hole in it
-
Just looking for a replacement chip, is it a TMC2660-PA-T or TMC2660-PA that I need to be a direct swap and is it just a case of removing the old chip and soldering the new one on?
-
before you do any chip replacement you need to go on a hunt for what was causing the problem.
start by checking the resistance of each phase of the stepper that you had connected.
-
@oliverracing said in Stepper motor lost torque:
Just looking for a replacement chip, is it a TMC2660-PA-T or TMC2660-PA that I need to be a direct swap and is it just a case of removing the old chip and soldering the new one on?
Yes that's the one. I think the -T suffix just means that it is supplied on a reel. But before you replace it, how old is your Duet? If the driver failure wasn't your fault, replacement is covered by the warranty.
-
Yeah, I'm gonna do a proper debug this weekend as thought I had but must have missed something. The TMC2660 chips had a 3 week lead time so got a few ordered anyway ~£5 each so hardly a issue if they're not the original cause but one will need replacing anyway as it has a hole in it where the magic smoke came out!
Not sure the age of the board, I've had it nearly 6 months and it was used before I got it, the previous owner had given up with 3d printing but not due to board issues but will ask him anyway if it might still be in warranty, I suspect not as he'd given up for a fairly long time.
I got the board free so I might just invest in a new board if there's nothing wrong with the steppers and the rest of the wiring as massively prefer it over the stock board on the Tevo.
-
There will be a PCB revision on the board itself, that may give an indication as to its age.
-
So....
I am officially an idiot. The stepper I did my testing with the pin-out is different to the original one - making this the likely cause of the pop and smoke I'd have thought.
I reckon wiring was the cause of the original loss of torque as a quick resistance check shows high resistance on on of the 4 wires.
I'll replace the drivers and go from there! Any other bits I might need to change?
Thanks for all the help, I'm sure I'll be back here when this fix doesn't work but fingers are crossed!
-
@oliverracing said in Stepper motor lost torque:
Any other bits I might need to change?
If the rest of the board is working, then it's probably only the stepper driver chip that needs to be changed.
-
Replaced the driver last night and checked all the wiring and just fired it up...
Getting a short to ground on driver 3 (am I right in thinking that's the E0 driver?) and over temperature on 2 3 and 4. - likely that these are linked? All feel cool to the touch but only plugged on for a few seconds as didn't want too fry too much!
-
Over temperature warnings and errors on all drivers are usually a sign that the SPI data bus between the TMC2660s and the Duet isn't working. It's better to run M122, that gives you the complete status of all the drivers. Check whether you have a solder bridge between pins of the driver you replaced.
-
So I think my board is dead, I've replaced the driver chip again and checked under magnification and theres definitely no solder bridges but getting the same error. I decided in drastic measures to remove the chip completely so we if this helps and I'm now just getting a short to ground warning on that chip - is this normal with it removed?
Got another board on order so will see how this goes. Gonna do a ground up rebuild of my printer to check everything standalone to make sure I can rule everything out as the cause.
-
Yes it's normal to get a short to ground warning when the chip isn't there. You can suppress it using M569 P# R-1 where # is the driver number. If that's the only problem now, you could solder in a new driver, if you need all 5 drivers working.
-
So for anyone reading this - please if you are using a Tevo Tarantula (or an extruder from one), check the bowden tube isn't pulling back into the extruder gear on retractions!
Yep, almost 100% sure this is the issue that started this whole mess! Then swapping in a stepper motor with different wiring at the plug caused me to blow the driver! Ahhh. Lesson learned (£170 lesson sadly - new TMC2660 chips and eventually a new board). On my setup it's impossible to see into the drive gear so I couldn't tell this was happening and when extruding the mangled bowden tube then pushed back to where it was meant to be but almost completely closed stopping it extruding properly! No idea how I got away with it working for a few years as it seems a flawed design!
