In article , Alden Hackmann says...

Initially when making the wheel I put the sawn blank on a faceplate to turn it
down to the basic diameter, and to drill and
tap it (5/16"-18). There is less chatter in this situation, but of course the
shafts have a little runout, and the tapped
hole is never exactly perfectly square to the blank - so between these two,
there's always some runout in the wheel when it's
on the shaft. Consequently I make up matched sets of wheels and shafts: once a
set has been assigned to an instrument, it's
not interchangeable with another set.

Hmmm. I would stay away from a tapped hole to mount it to the
shaft, for the reason you say, this does not provide good
locational accuracy for the finished assembly. A better
design would be a straight bore in the phenolic, and a
round section on the shaft. The shaft could have a shoulder
at one side and threads on the other, a nut could compress
the phenolic up against the shoulder.

Tapped holes do *not* work well if you are trying to minimize
radial runout. Just taking the part off the shaft, and putting
it back on, will give you a few thousanths of error right there.

Jim

==================================================
please reply to:
JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com
==================================================