Rigid peck tapping

So, much to my surprise the G33.1 tapping worked on my little Hardinge lathe. I should have tried it 10 years ago. 

It however doesn’t like nasty stringy tuff materials like 17-4 . Attempts at making a pecking program like below resulted in a blown out oversize thread. It worked ..kinda.

       G0 xo z.1
       G33.1 z-.2 k.03937
       G33.1 z-.3 k.03937
       And on until we get to a half inch deep.

     So.... is this a PID or drive tuning issue or would a separate routine that more accurately tracks the encoder need to be written. Some thing line a certain number of pulses have to be counted exactly after it see’s the marker channel ? I used peck tapping a while back on Mazaks and it was a game changer, being able to break  the chip every 2 threads is everything. Curious , is this doable?
 

I Don't have the exerience to say but these questions come to mind:

How much non cutting leadin distance do you have?
What RPM are you taping at?
What acceleration does the Z axis have?

Lead is 1 mm, lead in distance is .100 of an inch

I was running it at 500 rpm, about 35 sfm.

I’ll have to look at the Z accel , I’ll get it tomorrow.
Thanks

Max acceleration 30 , max axis speed 3.37 or 200 inch rapid

Why it isn’t working is in fact the question. Tapping can be voodoo on a good day. In the case of rigid tapping, the tap has to be dead on center or or will act as a boring bar. Chips getting caught in the hole will cause the tap to cut oversize and ultimately break. Tapping stainless is a bitch all the time. On a mill, I have come to thread milling holes . This being a lathe, not an option.

Only a bad carpenter blames his hammer, yet lcnc does have bugs and all you need is for some one with a different point of view to find one. There are also bits and pieces of info that are hard to find or forgotten in the documentation. Not blaming, it’s a big project.

Accel and decel on the spindle drive could have an effect, how does spindle pid effect its speed ramps? It would seem to me that it would just work, it can also be the material, it’s nasty. I’m in problem solving mode...

I would try increasing the lead in distance, if possible to see if it's something still accelerating. Though I would expect the thread to get better as it goes along.

Is spindle-at-speed signal connected to something so it waits for the spindle?