RE: [ATM] Which stroke ?

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

-----Original Message-----
From: Eric Royer

Thank you Bob and Jim for your help.
Even if the figure could be good enough for a telescope I would really like
to 
improve the figure. It's my fourth mirror and I can take my time to finish 
it. For the previous ones I wanted to finish them as soon as possible to be 
able to observe. This time I decided to make this mirror in order to learn 
more about figuring and not just to make a working telescope. I will make
one 
of course, but I'm not in a hurry.

So I'm going to try what Bob suggests (short strokes to remove the hole).
But 
I'd like a clarification about the edge. Jim says there is some TUE and Bob 
says there's some TDE. Does this depend on the way to interpret the other 
parts of the mirror ?
In other words saying "central hole, with TDE assuming the intermediate
zones 
are correct" is equivalent to "center is correct, a hill in zones 
2 and 3 and a TUE" ?

Thanks again for your help.
Eric
************************


Another opinion.... 

Your test data indicates you are using a tester with a stationary light
source. 

You have 5 zones and that gives 4 numbers to work with in the zone readings.

There is no reason not to boil those numbers down to the basic thing that
you have measured in the test. That is the DIFFERENCE IN KNIFE POSITION FOR
ADJACENT ZONES.  
If the difference from zones 1to 2 and 2 to 3, 3 to 4 and 4 to 5 are all
exactly as desired you would have as perfect as you can measure a perfect
mirror. But it is probably never going to be perfect. So you will subtract
the desired DIFFERENCE from your test DIFFERENCE. 

Zones                                  1-2     2-3    3-4    4-5       
test readings between zones           1.62     .9      .52   .42
Desired difference  between zones      .8631   .8127   .81   .8091    
Test readings minus desired readings   .7569   .0873  -.29  -.3891

No matter which program you are using for a Texereau type reduction program,
it will boil your test numbers down to the difference from one zone to the
next.  

You do not need the program to figure what to do.
>From zone 1 to 2 you are almost twice the desired difference. That is over
corrected there

>From zone 2 to 3 you are .0873 over your desired. Overcorrected there but
not much.

>From zone 3 to 4 you are .29 under the desired difference. That is under
corrected. 

>From zone 4 to 5 you are .3891 under desired. Under corrected by half
desired. This means the outer zone is turned up. 

Now it is about 1/6 wave P-V wavefront error. 

The programs that graph that error will make it look like mountains on the
moon. But 1/6 wave is not far from your desired finished mirror. It is not
so far from a parabola that you would need to work toward a sphere.

You can use a program to see what the shape is in relation to the best fit
parabola. But you won't really know much more about what to do to make it
better than what you can see by your test differences between zones.

The two inner differences are over and the two outer are under. In your case
if you work to increase the outer readings to the desired the inner readings
will change and if you do it right they will get closer to desired.

What has probably cause you to get the center over corrected with the outer
zones lagging behind is too wide a W in parabolizing. Or lingering (too many
strokes) on the sides of the W and not enough strokes in the middle of the
W.

Assuming a full size lap, what I would do to fix your mirror would be to
work TOOL ON TOP. Offset the lap to one side, I will say the left for the
sake of discussion, so that the right edge is about at the border between
zone 3 and 4. With light pressure on the right side of the lap do a W stroke
about 2 inches long and go to the left across zone 3 and 2 and then back to
the border of 3 and 4. The idea is to use the pressure to push zone 3 and 2
down and increase the correction from 3 to 4. After testing and seeing the
number from 3 to 4 getting bigger and 1 to 2 get smaller and maybe 2 to 3
stay about the same you will get where you want to increase 4 to 5. What you
do then will depend on what you have in the other zones. If you have fairly
uniform correction on the inner zones you may want to go back to MOT to
increase overall correction. But remember how you got what you have now.
Your parabolizing W needs more stokes in the middle relative to the sides of
the W or else you will end up over corrected in the middle again.  

When you do this TOT working across a zone you will increase the number from
the zone you stroke across to the one outside it. You will decrease the
number inside from the zone you stroke across.

You must try to do complete turns around the mirror. Don't do a partial walk
around the barrel. One time around is all you should do until you learn how
fast it will work. In general when you learn what to expect you would not do
more than three times around. I usually step so as to work across 9 or 10
diameter on a walk around the barrel. Don't "re-do" the first diameter you
stroked on by making the same diameter the last one you do on a walk around.
I also start on a different diameter on different sessions.

Excessive use of "accented pressure" can roughen the surface. There are
nuances that you can learn to blend the shape. Maybe not a good idea to get
too bold in the beginning but when you get the hang of it you can vary the
pressure starting light at the border of a defect and increasing across and
decreasing pressure on the other side. In any case in the beginning you will
find that you need to concentrate on what you are doing and where you are
doing it at all times when using this kind of stroke. If you don't
concentrate you might find yourself doing something different at the end of
a walk around than what you were doing when you started. Helpful towards
astigmatism. 

With some practice you will find that without reducing you test readings
beyond getting the difference from one zone to the next that you can get the
number just about exactly on. Then do the reduction to wave error to see how
good it is. The essence of figuring is to be able to make the zone reading
move in the direction that you want. You are measuring zones. You need to
work zones toward the desired differences from one to the next.  

I could elaborate.
Good thing that you are not in a hurry. Use the time and learn to make the
numbers move up and down when you want. When you have that down pat all that
is left is to learn to prevent or remove a dimple in the middle (not hard)
and to make a perfect edge (AAARRRGH!).
Jerry


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