On Thu, 21 Aug 2003 16:45:42 -0400, Mike Graham
wrote:

In article , wrote:

One stage further - mount a cheap laser pointer solidly on the
target wall and permanently fix a very small mirror on the mill head
with the mirror aimed to return the laser beam to a target on the wall
close to the laser.

This is getting better by the minute. 8-) Mind you, if you don't have
the mill bolted to the floor then you're screwed. If the mill is 'mobile'
then you'd need the target bolted to the back of the ram. Doable, but less
accurate. Probably more than accurate enough for a bridgeport, though.
Mike-I use an indicator holder which places the indicator about 8
inches from the spindle center. I use a tenth indicator and get it to
within .0003. The first shop I worked in insisted that heads be
trammed within .0003 and vises within .0001. Locating the laser on the
ram and bouncing it off a mirror back to the ram gets you about 6 feet
of light travel. 9 times the 8 inches I use. And I get the .0003 as a
total difference between indicator readings. So I guess that I would
need to see a .0027, no, .00135 difference from where the laser
pointed originally. I've got a HE-NE laser head that puts out a very
good beam compared to typical laser pointers and I can't get it that
close. I mean, I can't see the difference in location when it is that
small. But since my mill isn't bolted down I have to use the mill as
the mounting for the laser. There must be a way of amplifying the
error optically without bouncing the laser back and forth a bunch of
times. Using a laser would be so much more convenient.
ERS