G33.1 synchronized tapping problem

Interesting development after more testing.

I created a new config which does not use caxis.comp to manage the spindle.  Rather, it has a (poorly tuned) orient component so the spindle can be indexed.

Running a G33.1 cycle resulted in zero pre-synch movement.  No z-axis rapid movement before the spindle 'locked' to the Z-axis.

Just smooth, synchronized movement exactly as expected at all speeds tested - including 1000rpm with a 0.050" pitch.

So now I've got a conundrum:

• Caxis.comp - allows synchronized axis movement, but tapping and (likely) threading are dodgy
  • Non-standard component, no significant troubleshooting/tuning assistance available from community
• Orient.comp - no synchronized axis movement, but tapping/threading are as desired
• Something else?
  • I'd really like a working C-axis on my lathe, but I'm rapidly losing interest experimenting with LCNC.  I want a working lathe and not a perpetual software hobby.
  • My amp can't be externally switched between position & velocity mode
  • LCNC Devs do not appear to have spindle-axis high on the list of upcoming features.

Advice welcome.

Could you post your hal files (IE the one that uses caxis.comp and the one with the orient.comp)?

Both configs attached.  I included everything, not just HAL.

Thanks for taking a look.

Well, I had a good look and compared it to NoJo's config but I'm afraid I can't wrap my head around the problem.
If i understand spindle synchronized motion correctly then it really only depends on the encoder signals connected to the 'spindle.0.' pins.
I think the most likely way figure this out is to diagnose things with hal scope on the machine and find out what is happening to the index-enable signal and z-axis motion.
I'd love to help more but I have no hardware to test any of this.

Appreciate the effort.

I've had a think, and I've got one more config attempt left in me before I quit trying to have a working C-axis.

Plan is essentially what PCW suggested in my other 'freewheeling' thread.  I'm going to abandon caxis.comp, and instead use orient as a 'bridge' between velocity spindle and position-axis modes so the c-axis doesn't error out upon reconnection.

Flipping between spindle & axis will use M-codes, and I think it can be done without any halui stuff so I can do everything within the same g-code program.

I'll start a new thread when I run in to problems with this attempt.

Another (somewhat random) thing that came to my mind is that we can also change the 'ferror' values from gcode.
That might come in handy.

Good luck.