[3.4.6] Odd behaviour of rotary axis

o_lampe

Hi gents,
In the past, when I put M350 before M92, it uses the microstepping value to calculate the real steps per unit. I could change the microstepping without having to change M92, too.
But for a rotary axis, I have to set it with the final value?

M584 A4 R1         ; set drive mapping 
M350 A4 I0          ; configure microstepping w/o interpolation
M92 A{200*4}     ; set steps per turn

Another thing I noticed was the Object model.
In moves.rotation.angle the value was always 0, although it showed the correct value in DWC.
Am I on the wrong path? I want to use several relative moves, but then query the OM to find out absolute position.

o_lampe

@o_lampe Found out, it has the same behaviour in linear mode...must've been a general change I've missed?

T3P3Tony

@o_lampe said in [3.4.6] Odd behaviour of rotary axis:

put M350 before M92, it uses the microstepping value to calculate the real steps per unit.

Are you saying that this does not work in 3.4.6?

For the OM field this was fixed recently:

The values of the fields of move.rotation in the object model were incorrect, typically zero even if G68 had been used

https://github.com/Duet3D/RepRapFirmware/wiki/Changelog-RRF-3.x-RC

o_lampe

@T3P3Tony said in [3.4.6] Odd behaviour of rotary axis:

Are you saying that this does not work in 3.4.6?

At least it doesn't work in my specific case:
I've modified a working config.g to test the rotary motor only. Maybe I erased too much and the missing information messes up the setting?
Here's the whole config:
config.g

Hmm, I haven't set any printer kinematics with M669, but which one would I choose with only a rotary motor on axe A?

T3P3Tony

@o_lampe can you try sending M92 and M350 in the console to report the existing settings, then change the microstepping for the axis using M350 and confirm that the steps/mm changes.

o_lampe

@T3P3Tony I did and it works.

The problem sat in front of the keyboard. Me thinking I only had to set full steps per turn with M92.
But it always was 200 x microsteps

Thanks for taking your time