Orient component revised

:blush:
github.com/LinuxCNC/linuxcnc/issues/725
:pinch:

andypugh wrote: I can see that you might expect M19 to set spindle-on, but it seems that it doesn't. It's one of those things that could be argued either way.

In fact spindle-on only goes true after M3 / M4 _and_ when S > 0

does this mean you must have the spindle rotating already to orient?
Another reason to not require S = 0 to turn off the spindle.

Chris

Hi

The orient process is standalone, and do not use the traditional spindle hal signal.
But imo i think strange to not see spindle status on with "halui.spindle.0.on" "spindle.0.on" using orient

For now using orient yes S=0 but "halui.spindle.0.on" "spindle.0.on" = off

Br

I think we should require that spindle be on to orient.
I don't think we should require the spindle to be spinning to have the spindle on.
I don't think we should require that the spindle be already spinning to orient.

cmorley wrote: I think we should require that spindle be on to orient.
I don't think we should require the spindle to be spinning to have the spindle on.
I don't think we should require that the spindle be already spinning to orient.

1 : i think like you spindle=1 with S=0 and orient enable=1 is more clear
2 : i don't really know for other stuff but in my opinion if S=0 and orient=0 spindle-is-on need to be false
3 : ask M3/M4 S>0 when orient is enable = canceling orient and start spindle, the inverse is also ok so i think all is ok regarding to this. (not tested with M3/M4 S0)

1)
I didn't say anything about the spindle number so I don't understand.

2)
Why do you think S=0 should automatically call a M5?
in my mind S sets the speed (speed includes 0)
M3/4/5 sets the state

3)
That seems fine.and I think already the default