In article , Lloyd
E. Sponenburgh wrote:

Larry Jaques fired this volley in
:


Mill a hole and mike it?

In my followup, I explained that this is on a router -- cutting
structural foam, mostly.

The holes are never the size of the bit. Further (and I didn't mention
this), it's a ShopSabre 4896 bed router with a 'best' resolution of one
mil. So it's not going to make the same hole twice, even if working only
in Z. It's a very sloppy machine, compared to a mill.

Besides... this is a 'learning exercise', not something I absolutely must
do. Measuring on the lathe comes out +-0.0002 every time.

I don't want to invest in two whole sets of three-each v-anvil mics --
there are a wide variety of tool diamters, and it will take three
micrometers of each anvil angle to span the range of both the 3-flute and
5-flute cutter collections.

I've got to study that, too. It makes no sense to me that he was using
5-flute cutters on foam... But there might have been a reason I don't
see, yet.

My first thought was to cob together a jig where the cutter was held
between centers and free to rotate (by hand), with a micrometer head
positioned to advance perpendicular to the axis of rotation, allowing
one to measure how far from the axis the tip of each flute was.

This will probably work well for new cutters, but I think that many
regrinding machines hold the cutter in a precision collet (with the
cutting end flapping in the breeze), so the original between-centers
measurement no longer necessarily applies. Some new cutters are made
on collet-based machines as well.

Even so, is the possible error by sticking with the between-centers too
much? If so, a collet based jig is needed.

It's hard to beat commercially made ER collets for precision clamping
combined with easy mount and dismount of the cutter. One can loosen
and tighten the collet to rotate the cutter, but this will cause
errors. An ER collet on a spindle would be needed, but these are not
easy to make with sufficiently low runout.

Joe Gwinn