RE: [ATM] wedge removal

["Jerry" <wa4guu@verizon.net>]

The answer for a flex mirror is the last two paragraphs at the end. The
description below is a method to remove wedge for a lens where you could
loose a significant portion of your aperture if you neglect removing wedge
from the beginning. It will help those that haven't done this before to
think of the things involved even if they choose another way to do it.

I have removed wedge making a lens. The following assumes that a wedge
tester that registers to the glass edge and one surface measuring thickness
at a uniform distance from the edge as the glass is rotated in the tester is
being used. If the glass is accurately round and one of the surfaces is
square to the axis of the cylindrical or conical edge this will result in
the thickness being measured at the same distance from the center of the
glass as it is rotated. Just edged round is not enough. Picture a
cylindrical piece of glass that is concave on one end but when you set that
side on a flat surface the cylinder leans. No matter whether flat concave or
convex on the other side we might or might not have wedge. But it isn't
right. We really want it to stand straight and not have wedge. Keep that in
mind in considering the following steps that I used to eliminate wedge in
the lens. 

1. Grind the curve for side 1 which could be either side in the beginning.

2. In the lathe cut the end of a short tube that is a little smaller in
diameter than the glass disk. Turn the inside and outside of the tube round.
Turning the inside round is necessary for a convex glass and turning the
outside round is necessary for a concave. Do not remove the tube from the
lathe. You cut it true with the lathe spindle axis on the end and inside and
outside and you do not want to disturb that alignment.

3. Center the glass as best you can on the end of the tube that is chucked
in the lathe. The better you center it the less diameter of the glass you
will loose. You might use a block of wood against the other side of the
glass and use a live center in the tailstock to hold the glass against the
tube. Also melt some pitch and drip some around the tube/glass joint. It
needs to be held tight enough not to move in relation to the tube during the
edging. Needless to say...but I will anyway... be careful.

4. Mount a piece of strap steel on the tool post that is wide enough to
bring against the edge of the glass and feed grit to grind the edge round.
The strap steel should have enough flex to give when the high side comes
around. You do not want the steel to push so hard on the glass that there is
any risk of moving the glass out of alignment or a catastrophe could occur
with the glass. Needless to say cover the important parts of the lathe to
keep grit out.

5. Feed wet grit in between the strap and glass and work the longitudinal
feed back and forth in the process to make a straight walled edge. You can
tell how round it is when the strap can be set to just barely contact the
glass and it stays that way all the way around and it doesn't flex from a
high spot coming around. Patience! Hurrying can be dangerous.

6. Remove the glass from the lathe and grind the curve into side 2. While
doing so measure for wedge and remove it from this side.


What this does is make the edge a cylinder, or maybe a cone, with its axis
on the axis of the curve of side 1 of the glass. Don't tell me that a sphere
doesn't have an axis. We want this one to have an axis and we want it to be
the same axis the cylinder of the edge has. The first curve and the edge are
together a figure of rotation. Now the curve on side 2 can be ground and the
wedge if any will be ground off of side 2. You would not want to remove it
from side one because the tube trued on its end in the lathe was put against
curve 1 and that trued the edge of the glass to that curve. As wedge is
removed by grinding side 2 it will be brought to be a figure of rotation
sharing the same axis with side 1 and the edge.


There is a bit of flexibility in the details. The main thing is the common
axis idea. You can grind both sides to some fine grinding stage before
putting it in the lathe and edging. And then remove wedge on side 2 with
fine grinding if the wedge is not too bad. A dial indicator will jump
considerably on a rough ground surface but as you get finer you will be able
to measure closer. You only need to be as close as you can easily measure at
each stage of grit. Mel is very correct as always. With a dial indicator
calibrated in .0001" increments and you can interpolate several divisions in
between those .0001" marks you can get to where you can't see movement of
the needle. Let's just say way better than .00005".  

Since it seems you are making a flex mirror I would suggest that if you have
the same radius curve on both sides then with no wedge the thickness should
be very close to the same even at slightly different distances from the
mirror center. So in this case you would not really need to edge the glass.
So I guess the simple answer on a flex mirror is to put the desired curve on
the front and the back and at each stage of grit remove the wedge "as best
you can" from the back side. At each grit stage you will be able to measure
it better. If you can measure the wedge you can remove it and it is hardly
any work at all. If you are working the glass on a turntable if you just
offset the thick side about 20-25mm (for that 189mm disk) or so toward the
center of rotation and stroke just as if it were centered, the wedge will go
away probably faster than you think. In fact you should watch out that you
don't wedge it the other way.

Jerry




_______________________________________________
ATM mailing list http://www.atmlist.net/