Harold,

You guessed it! I assume you're speaking of a device that has a head of
given size----such as 1/2" diameter, and is held to the main body with a
spring? If so, they perform poorly at slow speeds, as I said. When it's
running at a good speed, it will stay on center when so placed with a
finger, and ride the edge of a part without moving away until such time that
the edge of the finder no longer has any clearance. At that point, it
smartly snaps to the side.

For my next session, I will run up the speed (after re-reading the max
RPM spec I saw) and give it a try.



If you run it too fast, it will pull away,
often causing some damage to the finder. When you're running at a good
speed, it will simply stay off center, riding the edge of your part. Set
your dial or DRO to zero at that point, reflecting the radius of the finder,
so the centerline of the spindle is at 0, not the edge of the finder.
Once you've done that, repeat the operation to insure that you are really at
0. It's easy to overshoot the first time, but the second time you have a
clue where 0 should be. If there's any doubt, repeat a third time. You
should be able to locate an edge within a thou repeatedly.

Dial backlash question: if I advance to the edge, back away to repeat
and then touch again with the dial locked in place from the first pass,
is it reasonable to expect the resulting two backlashes to cancel each
other? I think you've implied that, and it makes sense, but I want to
make sure that I am not getting the wrong idea. I understand that the
dial readings in the backward motion would not be correct.

Bill