I got to spend a few more hours with the lathe last night, and learned some
things (I expect to learn stuff at practically every lathe session for the
foreseeable future).
First, I learned that PVC plumbing pipe cuts easily and fairly cleanly with
a carbide tool bit designed for cutting steel, but that it will very easily
distort if cut too thin; I also note that the finish would probably be better
with much more back rake than the (I think) 5° built into the carbide
bit I'm using. Fortunately, I wasn't trying to actually make anything
with the pipe, just taking some trial cuts to see how things worked and practicing
at meeting a diameter from both ends after turning a piece end for end in
the chuck to turn the entire length.
Second, I got some experience with work deflection, turning a shape from
a 3/8" aluminum rod (sorry, I don't know for certain what alloy) extending
only about 1.5" from the chuck jaws; it took four passes at the same cut
settings on the cross slide to stop cutting, as I worked through the deflection
of the part with successively lighter cuts (the first was only .010" in feed).
Again, this might be better with a tool specifically ground for aluminum,
since that would want more back rake and the resulting sharper edge would
cut with less pressure against the work surface. The surface finish,
however, was better than I'd been led to expect from brazed carbide tooling
in a small lathe. I also got to play with the power feed; at the stock
setting, which amounts to 256 tpi (or a little under .004" feed per revolution,
if you prefer) it wasn't hard at all to disengage the half nut at the proper
location to avoid overcutting the shoulder or crashing the compound into
the chuck jaws, and surface finish was beautiful. Once I got enough
shims under the tool (this nominally 1/4" tool needed .090" of shims to come
to center height for a nominally 5/16" or 8 mm tool holder -- I'm guessing
the tool shank is a tad under nominal, since this is a brazed carbide tool
rather than a precision ground square blank), I was able to make nice facing
cuts using about 5° of "drag" on the tool -- point of tool closer to
chuck than shank -- and cutting from center out, as suggested in one of my
1941 machining texts. This causes the tool to take a very thin, but
wide chip and even with a very small radius corner gives an excellent finish.
Work deflection is important in producing a true cylinder (as opposed to
a taper); if not accounted for, the finished part will be larger at the free
end than the chuck end (which is why machining texts strongly recommend supporting
the end of the work with a center for anything that isn't short and very
stiff). In addition, it's important for turning to a specific diameter,
because if you measure the part at the end, where deflection has left it
oversize after a relatively heavy roughing cut, you'll wind up cutting it
too small, either at the chuck end where there's less distortion (and the
part was smaller than at the end that was easy to measure) or after a finishing
cut, which because it's lighter will deflect the work less and simply cuts
more than intended.
And relative to supporting the work with a center: I don't yet have any center
drills, but I thought of a way to center a piece with the tools I have. First,
I can use the 60° threading bit that came with the tool bit set I have
to cut the countersink (careful attention to setting the tool angle will
be needed to get proper bearing inside the countersink, but I think I can
match it against the center in the tailstock -- don't have a center gage
or thread gage yet, either), and then use that to keep a twist drill centered
to drill the pilot at the bottom, which prevents exceesive pressure and wear
on the sharp point of the center. This can't be done accurately in
the opposite order because the jobber length twist drills I have will wander
and drill off center, oversize, or might even break as they're advanced in
a tailstock chuck into an off center hole in rotating work. Countersinking
with a threading tool will require care to cut only on the front side of
center, since the bit won't cut properly (relief is wrong, at a minimum)
on the back side of the work, where rotation carries it upward instead of
downward past the cutter. I think the best way to proceed will be to
set the compound to 30° and cut from the perimeter of the countersink
in toward center, feeding with the angled compound. I'll have more
information here when I have a chance to try this technique.
