Pause print on driver overheat?
-
@Veti perhaps... but is there nothing that I can do to pause the job should an overheat even occur?
Also - should I have installed the heat sinks on the underside instead of the top side????
-
@Gerrard said in Pause print on driver overheat?:
Also - should I have installed the heat sinks on the underside instead of the top side????
do you by chance have a cloned board? they cheap out on the copper in the pcb which means the heatsink design does not work as intended.
-
@Veti I do...
I'm assuming there's nothing I can do about pausing...
-
you could potentially write a macro and execute that on layer change, but that is more of a bandaid.
you need to address the real problem and cool the board.
using an case like this for example
https://www.thingiverse.com/thing:3721923 -
@Veti So the board is setup on a bench, with heat sinks and a fan blowing across - it's not ducted, but there's plenty of air flow.
What would this macro look like?
-
M906 X1600 Y1600 Z600 E1200 I30 ; set motor currents (mA) and motor idle factor in per cent
Your motor current on Z is set to 600ma but your link for the motors indicate they are rated at 2100ma. So your Z motors are very likely stalling, hence the warnings.
Target ~80% of rated, so 1600ma
Those errors may have been from after the reboot due to the hard power off during a stall. Before worrying about how to pause during an overheat situation try giving the motors the right current and see if they still stall. It's likely a non-issue.
Cooling is good idea either way, but I don't think it's actually an overheating issue in normal operation here.
By the way in your other thread about stall detection I did ask
"Motor currents should be around 80% off rated max. Did you change the config for the new motors?"
-
@Phaedrux Sorry for the confusion... those 2100mA motors were for the X and Y.
These are the motors for the Z axis.
I take your point on motor current for these too - I've now set that to 1200mA (80% of 1500mA).
Still doesn't give me a lot of confidence - if the Z motors fail again (and they could), then I have a guaranteed crash. But if there's nothing I can do about it, then so be it.
-
@Gerrard said in Pause print on driver overheat?:
if the Z motors fail again (and they could),
Well running them at half the ideal current is likely the biggest issue. Once given enough current they should be more reliable. However if the bed is large and heavy those motors may be a bit anemic.
Before trying a long test print again I would suggest running a dry print with no filament loaded. (turn off the heaters and enable cold extrusion (M302 P1) to prevent it from halting on error).
-
@Gerrard said in Pause print on driver overheat?:
But if there's nothing I can do about it, then so be it.
Well the thing to do about it is try to make it reliable in the first place. As I said, I don't think the errors you saw were indicative of what was actually happening during the print.
Give the motors the current they need, cool the board, and there should be no more stalls to detect or pauses on overheat needed. Besides, if there were an actual overheat or driver error I am pretty sure the print would halt anyway.
It's actually really hard to trigger overheat warnings on the drivers. We're talking over 100c.
-
I'm referring to an insurance policy, but i've made the suggested changes, so we'll see how it goes.
-
@dc42 what exactly happens if a driver overheat happens during a print? Or short to ground for that matter?
-
Reports of all 5 drivers on the DueX being over temperature suggest either that the drivers are actually cool and it is the SPI communication or power feed to the DueX5 that has broken down, or the drivers really are hot because a transient has killed all the drivers.
-
@Phaedrux said in Pause print on driver overheat?:
@dc42 what exactly happens if a driver overheat happens during a print? Or short to ground for that matter?
An overheat warning just produces a message. An overheat error or short to ground shuts down the driver until VIN power is removed.
-
@Gerrard
post a picture of your wiring -
@Gerrard I think its already been answered by dc42, but I quite often get those over temp and short to ground errors on my Duet WiFi when I hit the Estop. I have a separate 5v power to the Duet so the processor and web control stay on and I can see them. In my case there definitely isn't a short to ground and the temperatures are all low/stable.
The Duet is designed so that the driver chips conduct heat to the back surface of board (hence you cool the board) not the drivers directly. See https://duet3d.dozuki.com/Wiki/Mounting_and_cooling_the_board#Section_Mounting_1
One other benefit of this is you can get a reasonable idea of how hot things get over time (transients less so) by looking at the CPU temperature. If doing this though, I would recommend calibrating the CPU temp (follow this guide https://duet3d.dozuki.com/Wiki/Calibrating_the_CPU_temperature)Side note on the cloned boards. The blue heatsinks they ship with probably make the cooling situation worse. Most people stick them on top of the drivers (plastic surface so doesn't conduct heat very well) and they tend to block any good cooling airflow over the board.
Also, I recommend double checking all the soldering joints on the bottom of the board. I have a clone board and found it had a couple of bad/dry joints on the heater pins
