Rigid Tapping

I have a spindle with an encoder 1pps for round, i'd like to do a G33.1 Rigid Tapping...

I tried to connect the thing but it does not seems to work.

it think it gets stuck at waiting for a spindle index pulse, and i'm not sure how should i connect my encoder

linuxcnc.org/docs/html/examples_spindle.html

i see that there is code for a 100pulse encoder and 1 pulse index

If i just want to use only the 1 pulse how do i change the code, any help?

Hi

see linuxcnc.org/docs/2.5/html/man/man9/encoder.9.html

setp encoder.0.position-scale [count]

where [count] is the number of pulses per unit ie 1 revolution

encoder.0.phase-z is index
encoder.0.phase-a is pulse

you can connect both to the same signal if you just have one pulse being used for both

regards

Rigid tapping requires Z tracking through reversals so you need A/B and perhaps index. (not sure if index is needed for rigid tapping)

With a 1 line = 4 count encoder (1/2 circle interrupter and 2 90 degree spaced sensors) A and Index can be the same so only 2 wires are used

I think its very very unlikely that rigid tapping would work with 1 PPR ) even 4 PPR may not do as Z tracking during the reversal is likely not going to be good enouigh

PCW

I think its very very unlikely that rigid tapping would work with 1 PPR

JT

I think Mack is the only one to claim that they can tap with 1ppr..

Indeed, whilst my answer is factually correct I hope, I have no idea if what you proposed will work.

I once set up a lathe to thread with a single pulse, feeding Z and A. It cut threads, but extremely badly, to the extent that they usually fouled up after several passes and were unusable.

regards

I got it to work with a single pps encoder connected to z and a

only problem i cannot do rigid tapping because i cannot count in reverse since this encoder is only one direction

i think i need A and B as PCW stated, and i'm going to put them...

I was just curios to know if there was a way to use the spindle reverse output to reverse the encoder count direction also, any ideas???

It cannot be reliable because the timing of the physical reversal is not known so there is the possibility of miscounts (pretty serious with 1 PPR!). I suppose some smart tracking hueristics could be used to interpolate better with 1 PPR but quadrature and more counts is a much better solution

If you want to try a minimal quadrature encoder, the 1/2 circle one might be a good choice (since it only needs 2 wires A/INDEX and B )

This is 4 PPR so 4 times more accurate with a very good velocity mode drive and 16 times more accurate with a torque mode drive!