[ATM] Full slit Foucault or the inverse wire test

["Marc J. Sibilia" <msibilia@ailr.com>]

I think I have found a significant improvement to the visual Foucault test 
that I will call full slit Foucault (as opposed to slitless) or the inverse 
wire test. I haven't seen it mentioned, but maybe I missed it.

First, an introduction. I am Marc Sibilia. I have been into astronomy for 
all of three months and only three weeks ago started my first 6 inch mirror 
from a blank my wife gave me 15 years go, so excuse me (and let me know) if 
I am going over old ground. (My apologies to ATM_free for reposting).

I read about Foucault testing with fixed and moving sources, pinhole, slit 
and slitless sources, and the wire test. I put together a very simple 
tester using a dial caliper as the stage and  measurement element. The 
caliper is held to two posts with rubber bands, the knife edge mounts to 
the moving anvil and a clip-on arm on the fixed anvil tilts the knife in 
and out of the beam.  The only down side is the measurements are in 
distance toward the mirror instead of away from mirror.

The tester:
http://www.atmlist.net/contrib/msibilia-at-ailr-dot-com/caliper_tester.jpg

I thought I would look at the 6 inch mirror in my Criterion first, and 
couldn't really see much in the way of shadows or zones. I thought I was 
doing something wrong, but a 6" f/8 isn't very different from a sphere. I 
found a 12.5" full thickness Pyrex blank on Ebay for $70 and I couldn't 
resist my "next" project. When it arrived, it had one very smooth side, so 
smooth, I could see myself. "Hey self, you look big in there". The thing 
was a fully polished unaluminized mirror.

So now I had to get really serious about testing to see if the figure was 
good enough to have it aluminized. Well, it is very easy to see shadows and 
zones on a 12 1/2 inch f/5.5 mirror.  but I have the same problem most 
novices do with judging where the null zone is.

I like the wire test because of the symmetry. You get a phi shaped dark 
circle with a vertical dark bar through it. You have to judge the center of 
a dark region instead of the 1/2 intensity point between a bright area and 
a dark area. And it is much easier to see when the knife (wire) is cut 
right in the middle of the return because of the symmetry. However, I used 
a pin for the wire (about 0.030") and had read that a very fine wire is 
best. I didn't have any, so I tried the hair trick but didn't like the 
contrast. The wire test is essentially right handed and a left handed knife 
edge superimposed.

So after that introduction, my technique (inspired by the reverse and 
subtract pictures on Michael Peck's semi-automated testing page):

I make two superimposed slitless testers by using two single edge razors to 
make a narrow slit (about 0.0005") in front of a high brightness white
LED, and view through the slit above the light source as you would view 
past the knife edge in the slitless. A good view is with the blades 
slightly diverging, so the viewing slit is about .002" to .003". The 
picture is like the wire test except the null zone is bright on a dark 
background.  It is difficult to see the mirror or the Couder mask viewing 
through the slit because it is so narrow that diffraction effects are 
serious, but the phi shape is very clear and contrasty, very circular, and 
looks like it will work really well photographically as in James Lerch's 
Robo-Foucault or Dale Eason's high-resolution automated Foucault.

However, the most surprising thing is how well this works visually if you 
do indeed use the mask. I cut the top and bottom off the mask so I can see 
the top and bottom of the "phi" to find the balance point (beam center). 
Now if I am measuring zone 3, I get some diffraction light through slots 2 
and 4 and the strange thing is that when zone 3 is just centered, the 
diffraction off the edges of slot 2 and 4 merge (dimmly) just below slot 3 
and the phi figure looks completely circular. It is very easy to judge. 
Also, when you move the tester longitudinally, the light jumps from slot 2 
to slot 4 smoothly but quickly. You can't see any change in slot 3, but the 
adjacent zones have a sort of amplification in the sense that the light 
band moves from 2 to 4 many times faster than the whole circle seems to be 
moving out, so centering the zone is very sensitive. I am a novice, and I 
was getting .001" repeatablility on zones 2, 3 and 4 and about .003" 
repeatablility on zone 1 and 5.

The mask:
http://www.atmlist.net/contrib/msibilia-at-ailr-dot-com/truncated_couder.jpg

Unfortunately, my mirror looks like it is overcorrected by a factor of 2 
and needs more work, so it has to wait until I finish the 6 inch. But I am 
excited about the testing technique. Please give it a try and tell me what 
you think.

Marc Sibilia

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