[ATM] WG: Focault (sic!) Help

[Nils Olof Carlin]

Berthold Hamburger replied to
Anna Bergen:

> In order to cancel out test stand astigmatism you HAVE to physically 
> rotate
> the mirror on its support. Texereaux might be old fashioned and we might
> have acquired new techniques in the mean time, but he was certainly right
> about a few things ;-).

> Why not just MEASURE astig directly by tilting the aperture mask 45 
> degrees,
> nulling the knife edge, then tilting the mask 45 degrees the other way
> (without disturbing the mirror or tester) and measuring the change in the
> null.   This seems a no-brainer.   Of course you have to rotate the mirror
> 45 degrees in the test stand and repeat, to be sure of getting both the 
> sine
> and the cosine components of the astig.

I believe (without having tried this IRL) that with some modification, this 
might work: Placing the mask apertures along two diagonals at 45 deg to the 
KE (aperture edges circular as usual, but upper and lower edges strictly at 
right angle to the KE) you could do a Foucault test matching upper left to 
lower right aperture, then vice versa, and thus you get readings for the 
zonal R'iiC along 2 perpendicular diameters. You need of course repeat after 
rotating the mirror to avoid the case where the axes are 
parallel/perpendicular to the KE. This would avoid the problem of moving the 
mask, but also and likely more important of using the KE in different 
positions and directions (I expect it *could* be done practically, with some 
extra care).
(This would give, in the case of astigmatism present, 2 slightly different 
values of the ROC. Obtaining the astig term from these data is left as an 
exercise to the reader ;-).

Nils Olof