Issues with Chamfer at end of thread on G76 cycle

[Chris RevNMR]

Hopefully you guys can help me with this one, I'm pretty stumped. I’m programming a part with a custom square thread on our TL lathe that’s 10 TPI and have to do it in multiple steps along the Z axis, turning down the OD and threading back, so I can keep the part rigid. When chamfering out using M23, with setting 95 at 1. And 96 at 45 degrees, the tool I’m using for threading is cutting into the previous thread on the chamfer out. It seems like the feed in Z is changing when it starts the compound move for the chamfer out, but the controller still says an active feed of .1 feed per rev.

Thanks for your time, any time I get stumped on something this place has been a great resource, but I haven't been able to figure this out so I figured it was time to make an account.

chris

 

[ewlsey]

That's just how it works. The Z axis has to decelerate to a stop at the end of the thread. During the deceleration, the pitch is deformed. Normally, the deceleration happen in the 45 degree chamfer out and it's not problem, you don't use the partial thread for anything.

You may be able to find a parameter to increase the size of the chamfer so that the tool is clear of the thread before it begins to decelerate. You could also program that long hand with G32 moves.

Slowing the spindle speed would also help. The deceleration is pretty much constant, so the slower the tool is moving, the quicker it can stop.

 

[Chris RevNMR]

Thanks for your help, I was worried it was just the deceleration at the end. I just wasn't expecting it to start so early in the chamfer, it seems like its beginning to decelerate right when the chamfer starts, right now as I'm trouble shooting it I'm running at 60rpm cutting threads in delrin so I can watch whats happening closely. I'll probably have to go to longhand and give that a try.

 

[Booze Daily]

I've never run a Haas lathe, so this is thinking out loud.

Shouldn't your pullout angle equal the angle of your thread form? In this case zero.
By pulling out at 45 in one rev. (.100) your X is pulling out .200 on dia. then straight up to the starting X in your canned cycle. If you're deeper than .200 (on dia) is it possible you're nicking the previous thread when your chamfer move is done and it rapids to your start X?

 

[Chris RevNMR]

I've never run a Haas lathe, so this is thinking out loud.

Shouldn't your pullout angle equal the angle of your thread form? In this case zero.
By pulling out at 45 in one rev. (.100) your X is pulling out .200 on dia. then straight up to the starting X in your canned cycle. If you're deeper than .200 (on dia) is it possible you're nicking the previous thread when your chamfer move is done and it rapids to your start X?

I was thinking similar when first working it, regardless of the angle, the issue is the deceleration of my Z axis as it chamfers. so you can see the final thread is tapering thinner as the Z feed slows.

I did a quick program with two separate G32 blocks that chamfer out to see how that works and it came out perfect, so now I'm writing up the code to just do it that way, using a subroutine with incremental movements to step down to my depth.