TMC driver current settings... are we missing a mfr detail here?

I've been studying GSTAT issues (related to TMC drivers) that will leave our machines in a 'non-desired' state... (threads at https://discord.com/channels/431557959978450984/801826273227177984/1072065882811019315 and a future probable RatOS update to assist/assure this will NOT be an issue.... https://discord.com/channels/582187371529764864/859890291591217162/1072355483249279037
Question came up here , while reading several 'industry wide' facing forums regarding electronics design. (my 50+ years here in that field)
I stumbled on a statement as such.....
"...However, always keep in mind that even if you want to control the motor current digitally (through SPI or UART), the maximum current will always be limited by the potentiometer. So make sure it’s not set to a minimum value before using digital control"
I'm wondering if within Klipper in general, this has something to do with so many GSTAT faults (look at all your klipper.log files). Even a search here turns up GSTAT error=1 on decent hardware (off the shelf not 'diy' issues)
I've tested the power supplies (external data loggers attached to my board(s) and have not been able to detect Vref fluctuations as suggested in some of the above referenced threads/conversations.
I'm wondering now if even though we are controlling the current in software/firmware.... if the mechanical of the 'steppersticks' many of these boards use has something to do with the GSTAT errors.
I personally never got any (reviewing my klipper logs) until I switched t othe performance modes. My next test will be going back to the oem lower current (and slower speeds) to see if the GSAT issues subside.

(oh bummer just realized...maybe this should go in the Klipper discord #stepper-drivers)
Thoughts? suggestions? questions ? tia

miklschmidt•2/9/23, 2:23 AM

I'd try and monitor the board voltages. particularly 3.3, 5v and 24v. See if there are any dips when the error occurs. It could be a PSU brown out or a bad dc/dc converter causing the driver to be reset. It could also just be a bad driver slot.

miklschmidt•2/9/23, 2:24 AM

Also, if you're using fysetc drivers, throw those out and get btt ones.

miklschmidt•2/9/23, 2:25 AM

I get tons of GSTAT errors with fysetc drivers too. They suck. I couldn't tell you why they suck though, i just concluded they sucked and stopped using them.

pteglerOP•2/9/23, 4:08 AM

the fysetc stepper sticks (boards) don't have the 4.7uF caps on the power lines for the TMC chip like the real BTT units.
will have to examine the Octopus Pro board slots themselves as well.
Q for you Mr. Mikkel.... concerning the RatOS TMS5160 config files.... . It looks like you prefer uart config over the SPI (which is smoother/faster). Is It more a symbiosis of settings (hardware jumpers etc... ) across board models... or a Klipper software(firmware) based issue? tia

Pptegler the fysetc stepper sticks (boards) don't have the 4.7uF caps on the power lines ...

miklschmidt•2/9/23, 4:23 AM

There’s no preference, you don’t have a choice. 2209’s only do UART, 5160’s only do SPI, so that’s what RatOS does.

Mmiklschmidt There’s no preference, you don’t have a choice. 2209’s only do UART, 5160’s only...

pteglerOP•2/9/23, 3:57 PM

tnx.... I thought RO was using uart for the 5160... will have to re-read that (the .cfgs) for my own clarity. (which is how and what) but GREAT! (5160s using SPI in klipper)