Advanced Search

I don't have such a setup so it is very difficult to say.
But it sounds like the current position is lost, and when you go back to csp mode the current position is set to zero no matter how the spindle actually is rotated.
aDm1N had something about tracking position didn't he? Did you try that?

Thank you Hakan.

This was very helpful. With your cia402pv and the files from aDm1N, I managed to get the servo running with GCode Commands both for rotation and positioning.

Here are the things I had to change in the hal-file:
- net c-mode-request and2.3.in0 -> spindle-ok-for-c
- net spindle-speed-normal spindle.0.speed-out mux2.0.in0 -> spindle.0.speed-out-rps
- and of corse some names and scale values

Now to my problem:
As long as I only use G0 and G1, the servo positions correct (example: C0 is at 12 o'clock). Then I let it run with M3 S1000 and after a while I stop it using M5. Now the servo stops at its position. In my optinion it should move to 12 o'clock, if I use G0 C0. But nothing happens. If I type G0 C90, it moves 1/4 revolution. So it looks like if the stopping of the servo sets somehow the zero position? 

forum.linuxcnc.org/27-driver-boards/3889...7c81?start=20#195102

video     That's what I did when I ran into the problem

2.9.7 I can't manage to start a configuration with inverted output ports.

pin 1-14-17 out-inverter-1 seems not to work

I can change state, but I can't start a setup with an inverted state

Unfortunately, in trixie image_2025-11-10-raspios-lcnc-2.9.7-6-trixie-arm64.zip the command mesaflash --device 7c81 --addr /dev/spidev0.0 --spi --readhmid doesn't work. No 7c81 board found.

Attached should be the config for that board.

Danke. Werde ich mal ausprobieren.
Gruß Andreas

understood, thank you.
I will look for tutorials as i'm beginner with this technology

Not sure exactly how that works. TIM2 and TIM5 share things, at least encoder pins, so only one encoder.
Maybe there can be TIM2 on one pin and TIM5 on the other for other purposes, not sure.
You can always do some surgery. Prototype in free-run mode and you don't need sync0 and sync1 which are
on TIM2 CH1 and CH2, so you can free them up and use them for encoder input for the 32-bit encoder.
Solder on tiny wires and melting glue. Has happened before.

Ok Thank you

Normally this is what is needed to read Mesa boards:
mesaflash --device 5i25 --readhmid
but for SPI i would guess
mesaflash --device 7c81 --addr /dev/spidev0.0 --spi --readhmid
should work, and it should be safe to try.
If you do have more questions about Mesa boards, use the "driver boards" section of the forum.

After looking at the arm64 Trixie repo it appears there are no packages available for mesaflash.
One option would to be build from source.
Or to manually download the bookworm version and install that.
As it is almost 12:00 am I fear I may not be able to respond in a timely manner.

With Imager version 1.9 it's ok.
Can I also ask what is the replacement for mesaflash --device 7c81 --addr /dev/spidev0.0 --spi --fallback --write.
I found:
halrun
halcmd: loadrt hostmot2
halcmd: loadrt hm2_rpspi
for reading (readhmid).
Or should I ask in a new forum?

Ich verstehe nicht. Wenn der Nullpunkt bei beiden Setups auf der selben Position ist, dann "müsste aber" alles normal funktionieren. Wenn der Nullpunkt für G54 für beide setups unterschiedlich ist (meine Beispielbilder oben) dann wäre es ja richtig, wenn die Maschine falsch fährt. 
 

Ist hier bei dem "operations" tab was verstellt?