Thread T.I.R problem

[cheeves]

O.K. here goes, try to bear with me...

Machining a 1.25 x 16 LH internal thread. Using a LH threading bar (right to left), part has been faced, bored and a dia. turned on o.d. in one chucking. Other end of part has 3 bearing dia's (+/- .0003). After part is threaded it is mounted on a threaded arbor (machined in spindle with ZERO runout). For inspection, part is thread onto a machined fixture for TIR check (.001 tolerance). Checking fixture machined in One chucking to assure runout, etc.). Consistantly getting .003-.006 TIR. Part is checked while still mounted on machining arbor and checks DEAD Zero!
Have tried turning "stem end" first and colleting on this and threading last and same results. I'm running out of Hair to pull out. HELP

 

[MikeHalloran]

It's run up against a shoulder on the checking fixture, right?

I.e., it sounds like the checking fixture thread is a little smaller than the arbor thread.



Mike Halloran
Pembroke Pines, FL, USA

 

[cheeves]

same size, both arbor and fixture machined with same tooling and gaged with same mics (p.d. and major dia.) yes it is shouldered. Part seems to runout laterally (concentricity) not a perp runout issue.

 

[byrdj]

Is this a New setup or something that has worked for others in the past?

Just going by discription and my invision, I see that there will be a problem checking a part by screwing it on a threaded rod. I would think a straight shaft with a tight slide fit to the part's I (with bearing fits), could be used for verifing concentric machining acceptance

 

[cheeves]

Fixture is not "threaded rod" it it a precision machined thread with mating "face" for face end of thread to bottom out against. fixture is bored thru and ground for a nice slip fit on a ground "post" about which the whole fixture/part revolves. part mat'l is 4150H, all finishes including threads are mirror like, not a ragged torn thread.
When part is mounted (screwed) onto checking fixture and machining arbor it "snaps" into place, good square contact, Prussian blue shows a clean 100% "seat".

 

[cheeves]

A little history...I've made hundreds of similar parts, all the same design with different thread sizes and pitches (both RH and LH) and different "stem" or nose configurations. Never ran into this before. Also seems different supplier had same problem with this part. Personally this seems like a pretty bad design, expecting a thread to locate a part within .001 Tir without at least some sort of pilot. Hell I'd make 'em wrong to make 'em right if I could figure out a way.

 

[byrdj]

Is it possible to cut the thread and face first in the part. then install the fixture, use the fixture to center the part, and cut the diameters checked for runout?

(still just my gut feeling is that part may snug up on the fixture off center with out pilot bores)