Overheating motors with Duet 3HC and Duet 3 mini

pcarrier

I'm glad to see that so many people are upgrading their printers with the latest Duet 3 mini and HC versions!

However, on our side, it seems like we are seeing more and more cases of motor overheating. I'd like to check with you if we can find the source of this issue.

The motors comes from LDO and is rated at 1A RMS, and we specify users to set the current at 0.9A RMS (1.28A Peak).

100% of our motors are tested for pushing force and maximum rotation speed and a QC report is generated attached with each serial number. The temperatures rise is usually as specified, reaching about 20°C over room temperature under normal use. They are running on a A4988 stepper driver.

We have 3 machines running on Duet Wifi and never had issues with the current configuration.

As specified above, the new boards seems to behave differently. We have some users using settings as low as 450mA and still have motors impossible to touch due to the high temperature.

Is there a lower limit, or a loss of accuracy, for low current settings?

Thanks for your help!

Phaedrux

Sounds similar to this comment maybe?
https://forum.duet3d.com/topic/23050/voron-0-duet-3-mini/17?_=1624659025074

Which motor is it specifically?

dc42

@pcarrier said in Overheating motors with Duet 3HC and Duet 3 mini:

Is there a lower limit, or a loss of accuracy, for low current settings?

On the Duet 3 Mini the quantum for setting motor current is 74mA. The current you request will be rounded to a multiple of this so that it is close to the requested current and no more than 15mA higher.

When the motors are operated in stealthChop mode, the current used may vary widely from the current you set, especially if you did not follow the correct tuning procedure. So you may wish to change the mode to spreadCycle to see if the motors then run cool.

pcarrier

@phaedrux said in Overheating motors with Duet 3HC and Duet 3 mini:

Sounds similar to this comment maybe?
https://forum.duet3d.com/topic/23050/voron-0-duet-3-mini/17?_=1624659025074

Which motor is it specifically?

Indeed, it sounds very much like this, thanks for pointing that out! I've seen exactly the same workaround, which is to set the current to something around 40%~50% of the rated value.

@dc42 could there be something wrong with the settings or registers sent to the trinamics drivers for low current values? I can't test it myself, but it seems like low current motor receive more current than they should. I'd be curious to check the voltage on a scope to validate the settings behavior.

@dc42 said in Overheating motors with Duet 3HC and Duet 3 mini:

@pcarrier said in Overheating motors with Duet 3HC and Duet 3 mini:

Is there a lower limit, or a loss of accuracy, for low current settings?

On the Duet 3 Mini the quantum for setting motor current is 74mA. The current you request will be rounded to a multiple of this so that it is close to the requested current and no more than 15mA higher.

When the motors are operated in stealthChop mode, the current used may vary widely from the current you set, especially if you did not follow the correct tuning procedure. So you may wish to change the mode to spreadCycle to see if the motors then run cool.

Excellent, I'll update our documentation to start at a much lower current, and also change to spreadCycle. Thanks for your help!

dc42

@pcarrier said in Overheating motors with Duet 3HC and Duet 3 mini:

@dc42 could there be something wrong with the settings or registers sent to the trinamics drivers for low current values?

Only if the driver is somehow getting set into low sensitivity mode. All Trinamic drivers (including the TMC2660 used on the Duet WiFi) have low and high sensitivity modes. RRF uses the high sensitivity mode.

pcarrier

@dc42

Many thanks for your help and prompt replies! I've added the provided recommendation to our support section.

pcarrier

@dc42

It seems like the problem is still there. We specifically ask to start with a current set to 350mA, use spreadCycle, and check the stepper temperature.

Despite these validation, our clients using the latest Duet with the TMC5160 still burn our motors coils.

So far, I'm out of option for our support, as the behavior only happen with the Duet 3 hardware. Is there anything that can be done to evaluate if the problem comes from the Duet on your side and find a solution?

My current solution is to suggest our client to use an external stepper driver when using the Duet 3.

dc42

@pcarrier I have a few questions:

Does the problem of the stepper motors getting too hot occur regularly, or just occasionally? If regularly, do you know a way to reproduce it?
Are you using the Duet on-board 5V regulator, or are you using external 5V power? If you are using external 5V power, does removing VIN power (while still providing 5V external power) and reinstating VIN power trigger the motors getting hot?
When motors run hot, are all the motors affected, or just one?

There are very many Duet 3 board in the field, and I am not aware of any other reports of this issue. So I think there must be something about your configuration that triggers the problem.

pcarrier

@dc42 Thanks for your follow-up.

Forgive me, I thought our client was using a Duet3 6HC, but rather was using a Duet3 Mini with the 2209 TMCs.

Two motor have burned so far, and we are trying to find the source of the problem before going further.

The 5V comes from the Duet.

Only the extruder is affected.

The first motor was set to 0.365A RMS, the second motor was set to 0.5A RMS.
The second motor never over-heated like the first one did.