The ganged axis (X in my config) refused to perform the homing

Both motors failed?

Even when setting motor current at standstill at 100% ($21x), the M122 report displayed 15/31 for the hold current.

Odd, works for me when I check.

                      X      X2
Driver          TMC2209 TMC2209
Set current        1200    1200
RMS current        1161    1161
Peak current       1641    1641
Run current       20/31   20/31
Hold current      20/31   20/31
...

Worse, after three seconds, the loss of torque on all axis was complete.

Can you post your original settings?
A recent change for the TMC2209 drivers was that the current pot was disabled, which it should be.
This could lead to the motor current been larger than before and the drivers shutting down due to overtemp (otpw flagged in the M122 report).

The $180-$183 and $190-$193 parameters have disappeared

They have been moved to the core from the Trinamic plugin and made generally available via an option flag in $22:

If enabled $24 and $25 will be hidden instead.

Hi Terje,

Here is the current config, i.e., the one I use for "build 20241127".
Just remove the $180-$183 and $190-$193 parameters to get the exact conf I applied with "grblHAL build/patch 20250103".
Paramétres_GRBL_V04.txt

For reference, I provide you with the firmware "grblHAL build/patch 20250103" the WebBuilder generated yesterday.
firmware_SKR3_H723_bootloader_241222.zip

Also, you asked me whether both motors failed : unfortunately, I can’t tell you because the X ganged axis has its motors running in opposite directions. The only thing I am sure of is that the axis didn't move and made harsh noise.

I don't think that the disabling of the current pot may have something to do with the loss of torque : the TMC2209 drivers I use are fully configurated through UART. Physically, I even believe they come without a potentiometer (EZ2209 from BTT).

Regards,

I am still not able to replicate using the STM32F4xx driver and regular Silent Stepsticks. I do not have a BTT SKR3 board so cannot test with that, perhaps @dresco can?
M122 output after homing fails could be helpful, also try with less current in case overtemp is triggered.

Yep, I can put the 2209 modules in my SKR3 EZ for testing.

Is just on the bench for development testing, but has 3 steppers attached. Will look for differences in M122 and apparent holding torque.

Am a bit confused by the version numbers above though, latest released H7 build was driver version 241221, and core version 20241222 (or are submodules getting bumped outside of the H7 repo for the webbuilder?).

(or are submodules getting bumped outside of the H7 repo for the webbuilder?).

No, perhaps I should put a note in the Web Builder about the H7 following a different update schedule?

Perhaps this commit is the reason? When I test with Stepsticks at max current (with heatsinks) they run too hot and overtemp is triggered.

Oh yes I see, I hadn't realised that build/patch description came from the web builder page...

@raoullector if you let me know what build options you want, I can send you a current binary with just that single commit reverted?

Great, I’m glad the issue has been identified.

@dresco , no thanks, I’ll pass and don’t have any urgent need. In fact, I was just taking advantage of a free moment to update the firmware; I will try that again later…

However, I’d be happy to get the info when the WebBuilder is operational again for STM32H7xx configurations.