berkeman said:
I dunno, seems like a pretty hard problem to me unless you can cover up individual mirrors somehow (which they can't). Anybody know how often they plan to re-calibrate the alignments? Once a "day", once a year, etc.?
According to the article below, they'll do a check every few days, it seems.
hutchphd said:
The alignment of each optical axis seems almost "easy". If fact at first cosideration the most difficult part if this design would seem to me to get the distance to the center of each mirror exactly adjusted. Each of those (d~1m) mirrors will produce a Rayleigh limit $$\theta =1.22\frac \lambda d$$ but for the coherently adjusted group of 18 where $$D\approx 4d$$ then the diffraction spot gets smaller by 4. How do they get that distance correct?
This article might shine some light:
https://svs.gsfc.nasa.gov/12753
During coarse phasing, engineers point the telescope toward a bright star and use NIRCam to find any large offsets between the mirror segments (though “large” is relative, and in this case it means mere millimeters). NIRCam has a special filter wheel that can select, or filter, specific optical elements that are used during the coarse phasing process. While Webb looks at the bright star, grisms in the filter wheel will spread the white light of the star out on a detector. Grisms, also called grating prisms, are used to separate light of different wavelengths. To an observer, these different wavelengths appear as parallel line segments on a detector.
[...]
During fine phasing, engineers will again focus the telescope on a bright star. This time, they will use NIRCam to take 18 out-of-focus images of that star — one from each mirror segment. The engineers then use computer algorithms to determine the overall shape of the primary mirror from those individual images, and to determine how they must move the mirrors to align them. These algorithms were previously tested and verified on a 1/6th scale model of Webb’s optics, and the real telescope experienced this process inside the cryogenic, airless environment of Chamber A at NASA’s Johnson Space Center in Houston. Engineers will go through multiple fine-phasing sessions until those 18 separate, out-of-focus images become a single, clear image.
[...]
The entire alignment process is expected to take several months, and once Webb begins making observations, its mirrors will need to be checked every few days to ensure they are still aligned
