What's wrong with my threads?

[allamateur]

I'm whipping up a threaded cast iron backplate and I think I basically scrapped my work so far.
It's for a Leblond Regal, so 2-1/8" x 5 TPI internal thread for the spindle.

I got it started and had good flats on the internal threads. Took the chuck/plate combo off to test fit after I had gotten close, then put it back on to cut the thread more. To my surprise, the threading dial was off from where the threads were (whole numbers didn't line up with the thread). This may be from the chuck and its plate not seating exactly the same on the spindle, or something else I can't think of? If any of the chuck turning to remove it happened without the leadscrew engaged I imagine that would do it too? Anyway I tried to find the thread again just by bringing the cutter into the thread groove and making sure it had very little endplay in there. Cut more threads, now suddenly the flats were gone and the bottom of the groove was torn up something new. One final removal for test fit (it just barely fit, even in this state) and then final 0.001" pass.
Now the tips of the thread are absolutely ruined. My leadscrew has lots of endplay but I tried to ensure the slack was taken up when moving forward, and never ran it in reverse with the cutter engaged in the work. Not sure what was the most fatal error, but it seems I never had the same Z reference after chuck removals. Since I couldn't trust the thread dial, I had to find the thread manually by plunging the cutter in when the work was stationary. Let me know your thoughts, please.

Last pic shows the threads before the first removal, and the 3 other pics show it in its current state. What's more - the spindle thread tips feel a tiny bit rougher after the first test fit, so I know not to put it back on there for a final fit.

 

[Eddyde]

If the threads are unusable and you must reuse the backplate, you can bore it out and thread to a larger diameter then make a threaded bushing to fit the new thread and the spindle. You can locktite or pin it permanently in place.

 

[mattthemuppet2]

The crests or valleys of the thread are unimportant as long as they clear the mating thread. What matters most of all is that the threads and the facing of the back of the back plate (where it butts up against the spindle face) are done without removing the back plate from the chuck.

The thread flanks and the back face of the back plate are what keep it concentric to the spindle for the length of the bed.

 

[brino]

@allamateur ,

Dang I hate when projects go bad.... I feel your pain.....

One thing I have learned is to do the ****-ups on a practice piece, using exactly the same tools and procedure you want to use for the final piece.
Although I have done the "sample" in aluminum instead of steel, for ease and cost.....

This may be from the chuck and its plate not seating exactly the same on the spindle, or something else I can't think of?

You are right, that could do it.....some crap on the mating surfaces would change the z-axis.....

All those little ridges make it look like you has some significant chatter while cutting.
What did it sound like?

Stick with us, this group can help you sort out the root of the problem.
The threaded bushing idea from @Eddyde above is a great one; you can fix this!!!

Brian

 

[allamateur]

If the threads are unusable and you must reuse the backplate, you can bore it out and thread to a larger diameter then make a threaded bushing to fit the new thread and the spindle. You can locktite or pin it permanently in place.

Definitely a good idea, and I left lots of diameter on the thinner part of the backplate thankfully.

The crests or valleys of the thread are unimportant as long as they clear the mating thread. What matters most of all is that the threads and the facing of the back of the back plate (where it butts up against the spindle face) are done without removing the back plate from the chuck.

The thread flanks and the back face of the back plate are what keep it concentric to the spindle for the length of the bed.

Understood, thanks. But if my problem was an unreliable Z-reference for thread starting, isn't the problem that the grooves have become wide (just wide enough to accommodate the spindle male threads, not overshot thankfully) but not deep enough because the material was removed axially and not radially? I didn't get to size by feeding out on the compound, I think I got there from shifts in my thread-start which at the same time chewed up the thread crests. I think this might also explain the roughness I felt on the thread crests of the spindle after test fit, the irregular thread valleys on the backplate are both not quite deep enough but also not flat, so they chew into the spindle thread crests.

@allamateur ,

Dang I hate when projects go bad.... I feel your pain.....

One thing I have learned is to do the ****-ups on a practice piece, using exactly the same tools and procedure you want to use for the final piece.
Although I have done the "sample" in aluminum instead of steel, for ease and cost.....


You are right, that could do it.....some crap on the mating surfaces would change the z-axis.....

All those little ridges make it look like you has some significant chatter while cutting.
What did it sound like?

Stick with us, this group can help you sort out the root of the problem.
The threaded bushing idea from @Eddyde above is a great one; you can fix this!!!

Brian

You're right, I should have probably done a test piece, or even better a male test plug that was a clone of my spindle, but I had no stock the right size for a test backplate and had already removed the chuck by the time I considered a male test plug. It seems people say make a test plug for the ease and convenience of not having to remove a very heavy chuck (the giant cast iron backplate + chuck combo is probably at the edge of my lifting ability), but really it's also useful because removing the chuck can ruin Z references. Much more important.

I took the threading cuts below 100rpm (this geared Regal's max is 500rpm) because the hole is blind and I don't have much practice single point threading. As a result, I knew the surface finish would be poor using a carbide insert. I didn't mind so long as it came out decent dimensionally. It didn't scream or anything, but it did sound like it was digging, a low pitched sound (with a bit of vibration but far from rubbing). I also made sure to blow off the spindle with some duster after each chuck removal but I guess it wasn't enough.

 

[benmychree]

I have made a plug gage for every lathe that I have owned and needed to thread backplates for, any other approach is foolishness; the plug gage should be as close to exactly the same PD as the spindle, and the bored thread should be a snug, free fit on the plug gage, the fit of the backplate on the unthreaded part of the spindle should not be too tight in the backplate, making it too tight invites galling, it is the threads that make for centering the chuck closely, not the register diameters.

 

[markba633csi]

I have a couple questions: You mentioned removing the chuck/plate combo off to test fit- how were you mounting the backplate to the
lathe to cut the threads? - were you holding the plate with the chuck?
Secondly, did you disengage the halfnuts when you removed the plate to check the fit?
-Mark

 

[OTmachine]

Sounds like your threads are out of time. Hopefully, you are not ……..
If you are not to full thread depth yet, and your minor diameter is OK, you can, possibly, pick up and finish the thread. You only need the inboard thread face to be correct, since that is the one holding and locating the plate. Just a thought.
Sounds like you need practice picking up and chasing threads on a lathe.

 

[SLK001]

I have made a plug gage for every lathe that I have owned and needed to thread backplates...

+1 on this.

Having a plug gauge removes the need to unmount the chuck with the work, check the fit (with a 20 pound weight just hanging on), then remount the thing. But what is done is done. Just pick up the thread again and finish the job to see if it is salvageable.

 

[allamateur]

I have made a plug gage for every lathe that I have owned and needed to thread backplates for, any other approach is foolishness; the plug gage should be as close to exactly the same PD as the spindle, and the bored thread should be a snug, free fit on the plug gage, the fit of the backplate on the unthreaded part of the spindle should not be too tight in the backplate, making it too tight invites galling, it is the threads that make for centering the chuck closely, not the register diameters.

Too right, now I know it from experience. I'll clean up the bore since it's tighter than the chuck's backplate for sure.

I have a couple questions: You mentioned removing the chuck/plate combo off to test fit- how were you mounting the backplate to the
lathe to cut the threads? - were you holding the plate with the chuck?
Secondly, did you disengage the halfnuts when you removed the plate to check the fit?
-Mark

I was holding the plate in the chuck, backwards as in the pics. I don't believe I disengaged the halfnuts the 2nd removal, but I might have on the first. There was still a loss of Z positioning the 2nd time though.

Sounds like your threads are out of time. Hopefully, you are not ……..
If you are not to full thread depth yet, and your minor diameter is OK, you can, possibly, pick up and finish the thread. You only need the inboard thread face to be correct, since that is the one holding and locating the plate. Just a thought.
Sounds like you need practice picking up and chasing threads on a lathe.

I am very close to depth, it threads on but very tightly so I can take another pass probably.

+1 on this.

Having a plug gauge removes the need to unmount the chuck with the work, check the fit (with a 20 pound weight just hanging on), then remount the thing. But what is done is done. Just pick up the thread again and finish the job to see if it is salvageable.

Yeah it makes total sense to do, but it wasn't a thing I had found while looking up this process until I had already hit the critical bore dimension and thus couldn't remove things from the chuck. Definitely lesson learned.