Single Point Threading Issue - might be gearing?

[AGCB97]

I had a similar problem with the threading dial and the 1/2 nuts not wanting to go in at the line. Solved it by putting a magic marker line slightly right of engraved line and now it always goes in when I want it. You may be getting 'just before' one time and 'just after' the next which would not make for nice threads.
Aaron

 

[WobblyHand]

I had a similar problem with the threading dial and the 1/2 nuts not wanting to go in at the line. Solved it by putting a magic marker line slightly right of engraved line and now it always goes in when I want it. You may be getting 'just before' one time and 'just after' the next which would not make for nice threads.
Aaron

Thanks Aaron. I'm pretty sure that is what happened for this second set of threads (32TPI, not the wrongly geared 45TPI). What happens if you alternate just before, or just after? What does it look like? (Ragged edges?)

The first set of threads, although the wrong TPI, were much cleaner. I couldn't hit the mark, but what I did was consistent. The second set with the correct TPI had multiple instances of jitter around the mark.

 

[T Bredehoft]

Tom, I never heard of 10l12. Ledloy is usually 12L14. Was that a typo?

No it wasn't a typo, it was a mistake. I thought I remembered two grades of ledloy, musta been a nightmare. OK, ohter grades, 11L17 and 41L40. I'd like to try that, i use a lot of 4140.,

 

[mikey]

Okay, thanks, Tom. I can't even begin to count the number mistakes I've made here ...

 

[darkzero]

Hmm, forgot the picture of the awful threads.
View attachment 316670
How do you make nicer threads?

Are you feeding with the compound (modified flank method)? If yes make sure your compound is set correctly. Many small import lathes don't have a full scale on their compound, my 12x included. With your compound set parallel to the spindle axis, if your scale reads 0 & you set the compound to 29.5° on the scale that's incorrect for threading. This usually causes jagged rolled over funny looking thread forms. It would need to be set at 60.5° which the scale on most of these lathes don't read that far. You'll need a protractor or something else to set the correct angle.

If this was the case & you don't have a protractor forget the compound & try feeding in with the cross slide. 32 TPI is a fine enough thread where you'll be ok feeding with just the cross slide.

 

[T Bredehoft]

Just a small comment here. 60.5 or 29.5 is not absolute. The angle must be a tiny bit off of 30 or 60 so that there is cutting on both sides of the threading tool. A cut at exactly 30 or 60 will not clean up one side of the cut, that's why we offset the tool .

 

[homebrewed]

FYI, you don't necessarily need to disengage the half nuts on lathes that can be run in reverse. At the end of a pass, stop the motor, back the cutter out and run the motor/lead screw in reverse until the cutter goes past the work. Stop the motor, move the cutter in for the next pass, switch to the "normal" direction and do your next pass.

Kind of painful to do it this way but it will separate the threading dial issue from other things that might be going on. And this method is the only way to cut metric threads with a 16TPI lead screw.

 

[T Bredehoft]

Using a stop on the cross-feed, so you can run the carriage back in and just increment the compound, with the lathe going at a moderately slow speed you can (anyway I can) make a pass every couple of seconds. ) (short thread) I did it at work fourty years ago and still do.

It helps if the thread you are chasing can be chased using any of the gears (notches) on the leadscrew

 

[WobblyHand]

Are you feeding with the compound (modified flank method)? If yes make sure your compound is set correctly. Many small import lathes don't have a full scale on their compound, my 12x included. With your compound set parallel to the spindle axis, if your scale reads 0 & you set the compound to 29.5° on the scale that's incorrect for threading. This usually causes jagged rolled over funny looking thread forms. It would need to be set at 60.5° which the scale on most of these lathes don't read that far. You'll need a protractor or something else to set the correct angle.

If this was the case & you don't have a protractor forget the compound & try feeding in with the cross slide. 32 TPI is a fine enough thread where you'll be ok feeding with just the cross slide.

View attachment 316677

My setup is the modified flank (your first picture). I set the angle to 29.5 degrees using a protractor. Or rather, to 29.5 indicated on the protractor. I think the problem with the threads was dithering about the mark on the thread dial. I tried again using a practice piece of PVC pipe and the threads came out cleaner. Of course, stopping on time is still a problem for me. But the thread is looking quite a bit better. I was able to hit the threading dial mark consistently. (Actually, just before the mark, but it was consistent.)

Next, I'll try doing this thread on some 1215 steel. I had some hex stock that I machined to round. If the threads are good and clean, then I will try Joe Pi's method of threading with an upside down bit going from headstock to tailstock. What I am really trying to do is to make a replacement for this damaged part:

On this piece, the threads got destroyed on the right hand side. (This was a purchased part.) My reflexes aren't fast enough to stop the threading if I go from tailstock to headstock. Maybe Joe Pi's threading method will let me make this. The material is O1, the thread is 8-32. Center diameter is 0.1875".

 

[mikey]

This is the kind of part that a Sherline lathe excels at making because threads are cut manually by turning the leadscrew by hand. You might try spinning the chuck by hand to see if it works. If it does then it might be a good idea to make a handwheel to turn the spindle on these smaller threads.