On 2008-07-09, Jim Wilkins wrote:
On Jul 7, 11:42 pm, woodworker88 wrote:
On Jul 6, 10:14 am, Ned Simmons wrote:

Hardinge specs the thread as 1.238-20 in the "Total Workholding"
catalog. You may find the relevant section here...http://www.hardinge.com/?pageId=141

Yeah, I thought it was something like that.

The thread OD on my collets measures from 1.220 - 1.230 for Enco to
~1.243 for R-S-B. I made a collet nut for an S&D drill grinding
fixture that fits closely on a 1.240 thread. Some of the R-S-B collets
are too tight for it.

How would you measure an inside thread to cut it to 1.238"?

Well ... pitch diameter is what matters, and there is a way to
measure pitch diameters on internal threads somewhat akin to the
three-wire method for external threads.

You have a spring wound of an appropriate wire gauge for
measuring the pitch involved, and wound to 20 TPI -- but a little over
diameter.

You then bend the ends to form a pinch point outside the coil at
one end which will cause the diameter of the coil to reduce, so you can
put it in the thread to be measured.

Then, you measure the ID of the coil --- probably with something
like a tri-mike or something else which can let the spring wire pass
out between measuring arms.

You use the same math used with three-wire outside thread
measuring to determine how small the hole should be, and from that and
your actual measurements you can calculate the actual pitch diameter.

Cutting
out a fixed distance from the ID only works if the tip geometry is
correct and your lathe isn't worn. The spreadsheet gearbox chart I
made for my old South Bend gives the 29 degree infeed and tip width
based on Machinery's Handbook thread data but in practice the last few
thousandths are cut-and-try, mainly because I don't regrind the tip
for each pitch, especially the internal threading bit.

Here is where carbide insert tooling can help you. There are
inserts available to cut precise form threads, as well as the more
general sharp V ones. You don't have enough stress on a drawbar to need
to avoid sharp V bottomed threads, so you can calculate the infeed for
the sharp-V threads (straight or angled as you prefer). Note that the
angled infeed should be reversed for internal threads -- that is the
crank end of the compound towards the headstock instead of the
tailstock, but still the 29 (or 29.5) degrees for standard US and
metric threads, or half of 55 degrees for Whitworth threads. BTW -- the
drawbar and collet for WW and D sized collets use a buttress thread
instead of a standard V thread, so the infeed angle for those has to be
separately determined.

It's about
right for 32TPI so it makes coarser pitches undersize if I zero the
compound on the blank's surface. If I understand the book correctly,
the flat should be 0.00625" wide for an internal 20TPI UN thread.

This is why I suggested using your largest collet thread as a GO
gauge.

That is good advice anyway -- that is using the largest *thread*
OD, not the largest collet capacity. And if you have things like 3-jaw
and 4-jaw chucks mounted on collet shanks, check those as well. My
largest is a 3-jaw chuck which I got *after* I made the drawbar
extension for my drawbar when converting form 2-1/4x8 threaded spindle
nose to L-00 nose.

Enjoy,
DoN.

--
Email: | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---