I was machining a bit of steel from hell the other day and
having no luck in achieving an acceptable surface finish. The steel
was of unknown origin (my local scrap dealer).It produced stringy
swarf and work hardened when light cuts were attempted. Heavy cuts
resulted in a uniformly rough surface. Cuts light enough to
potentially give a good finish skipped in and out of cutting leaving
random grooves more than 0.001" deep. I tried most of the usual tricks
but the only one that really worked was to machine oversize and take
the last few thou off with a fine single cut file. The penny then
dropped that, if the file works, then a file geometry variant of the
old skiving cutter might do the trick.

To cut a long story short - it does! The geometry I finished up
with was a 5/8 HSS blank with a corner to corner 45 deg flat ground on
the last half inch. In lathe tool terminology this is 45 deg NEGATIVE
side rake.

This was then given extreme front clearance by grinding 40 deg
clearance directly normal to the flat 45 deg cutting edge.

If the middle of this tool is presented square on to the OD
of the work piece, as far a the chip is concerned, it is a 45 deg
skiving cut with zero top rake. The cutting action is improved by
angling the whole tool holder about 15 deg anti-clockwise. This
results in a small effective top rake which allows the thin wide chip
to exit in long satisfying curls.

This tool produced consistently good surface finish with cut
depths in the range 0.0005" to 0.005. The finish and chip formation
seemed to be best when cutting in the 0.001" to 0.002" range. The main
drawbacks are that the tool shape makes it impossible to machine up to
a shoulder and the wide thin chip increases the tendency to chatter.
This means low surface speeds (sometimes backgear!). Power traverse is
advisable for uniform chip thickness.

The small included angle of the cutting edge (50 deg) is
well outside normal lathe tool parameters but is not a concern because
of the intended light cuts. It's perhaps worth remembering that the
normal cutting edge included angle of even a standard spiral twist
drill is close to 60 deg and that's happy with pretty heavy cuts.

These are results from a guesstimated set of cutting
angles on a very limited range of workpieces. The angles don't seem to
be very critical but I've no idea whether they are near optimum. I've
yet to discover how well the tool performs on other work materials.
It's hardly a tool for the professional because of the low metal
removal rate. But for an amateur, a tool that takes light cuts and
consistently leaves a good finish can be a pretty useful item.

Jim