[sansbury]

I'm doing a customer part that involves a 1/8-28 BSPP thread about .15" deep in a fitting. I'm using the T/C tap holder and a LinuxCNC subroutine which runs at 500rpm. I've used this subroutine many times on straight threads without any difficulty. But now I'm getting good threads on only about half of the holes, and since each part has three holes, my scrap rate is effectively 100%. Material is some sort of brass. It taps very easily by hand.

For those of you who've done tapered threads with the T/C head before, any suggestions on parameters, dwell times, techniques, etc.? I'm not sure whether the problem is happening on the way in or out, but watching it run, everything looks normal.

[SCzEngrgGroup]

What, exactly is wrong with the ones that you are rejecting? That should give a clue as to what's wrong. Do you get consistently good threads if tapping the same holes in the same material using the same tap by hand? If so, then it's probably not the tap, but something about your tapping code causing the problem, so you might want to post the code. Are you using plenty of lubrication?

Regards,
Ray L.

[sansbury]

By hand they're easy-peasy, so it's definitely something to do with how it's running on the machine. It's brass, and shallow, so I haven't been using lube, but I guess I could try that.

I don't have the code handy right now, but it's very simple. It starts the spindle at 500, waits 2-3 seconds for the speed to stabilize, feeds in at a rate of 500/TPI to the specified depth, stops the spindle, waits 0.5 seconds, starts the spindle in reverse at 500 and feeds out at 500/TPI.

I suppose I could run a test where I stop the spindle before it reverses and withdraw it manually to see if the problem is happening on the way in or out. I also have thought about running these without feeding the tap in (essentially letting the tap pull itself in and out) but it was getting late last night and I wanted to get some feedback from others first in case there was something else going on.

[FuriousGeorge]

Are we talking thru holes? Try eliminating the 0.5 second pause before the retract. The spindle has a built in pause when reversing directions, so the additional dwell time is not needed. What I believe is happening is that during that 0.5 second pause, the spindle is continuing to rotate clockwise a little, which is causing the tap to be pulled down farther into the hole. With a straight thread, this is not an issue as the spring allows the tap to descend at the proper feed rate and your just cutting air. With a tapered thread, you're no longer cutting air- you're tapping deeper than anticipated and marring your thread. Your code should look something like this:

M03 s500
G01 z-0.15 F17.857
M04 S500
G01 z0.1 F17.857

Again, Mach has a built in pause so it will not execute any commands until the spindle is up to speed. This makes pauses un-needed. You may also want to tweak your z depth and tap a little bit shallower than you need. The spring will pull the tap down during the built in pause. For 0.15" of thread, try tapping to a depth of 0.1"

[sansbury]

I'm using LinuxCNC.

[MichaelHenry]

What sort of a tap are you using? When I run standard NPT taps by hand they need to be reversed every 2-3 turns to clear chips from the flutes, so I don't think those would work well in a T-C head. Are there any spiral point or spiral flute NPT taps out there? They might work better with a T-C head.

I'm no expert, but have had good success with an NPT thread mill making large tapped holes. Here's one vendor for the size you need:

LakeShore Carbide.com Carbide End Mills-Carbide Thread Mills - NPT THREAD MILLS COATED

Mike

[sansbury]

Yeah, it's a Whitworth (55°) thread, so I'd need to buy a $90 threadmill for what might be a $150 one-time job. When I hand tapped these, I didn't need to reverse the tap, but maybe that's the issue. It's a really shallow thread (seriously, it's maybe 1/8" deep, I was doing .150 just to make sure it was fully formed) and brass so the chips are small and didn't pack the flutes at all. But hey, I'm the one with the problem, so that could be it.

[mike sr1]

It could be the upward spring pressure is too high on the return as a tapered tap pulls out of the thread right away and the spring pressure may be too much in the reverse direction and could damage the thread in such thin material possibly.