Fast Forward, in a Slow Kind of Way
Since you blinked, it is now February 1, 2015, just 18 months after the last post in this highly volatile blog – isn’t time travel a kick? Last post I promised that I would talk about the switch from a Cassegrain optical design to a folded Newtonian design, but its only fair to catch you up since our last post.
You’ll recall from August 2013 that we were trying to adjust the shape of the mirror so that we got a flat surface from the analysis of the Open Fringe program. We had lots of good sized bumps. Steve Follett, pictured here in a series of photographs by Len Nelson, along with Mark Hillestand and Larry McCune, have been working on those local irregularities these last 18 months with multiple weekly sessions that consist of measuring the surface after it has been left alone for at least a day, and then doing the briefest work – a few minutes – in reaction to the results of those measurements. You can see Steve here adjusting the interferometer with a DSLR ready to capture the interferogram as well as an image of the 40-inch mirror in its test stand.
This process was enhanced about a year ago when Larry came into possession of a digital projector that was being made surplus by the local astronomy club (Sonoma County Astronomical Society). The team began to use the projector to throw images of the processed interferograms directly onto the surface of the mirror so that the area of next action would be very clear – no errors in transferring from the report to the mirror.
We should say here that this approach is the opposite of what the ATM lists have recommended. We heard multiple times that with the astigmatism or the kind of aberrations our interferograms indicated, we should return to a sphere (back to grinding!) and try again. The contention is that it will take too long to reach an acceptable result and we won’t get the quality surface that we should be trying for. Our team decided that we should continue to make local corrections and slowly but surely sneak up on the correct figure. We are hoping that is what we have accomplished.
These three images need some explanation. The first is a sample inteferogram, the basis of the analysis that we do to measure the deviation of the surface from the ideal, a parabola. In practice, many of these inteferograms are captured at different rotation orientations of the primary mirror in its test stand, and then are combined in the Open Fringe software.
The second image is a 3-D representation of the surface of the mirror once these interferograms have been analyzed, taking into account the minute deviations caused by the mirror flexing on the test stand. You can see that the surface, which should be flat in this image if it perfectly matches a parabola, deviates from that ideal by about a half wavelength of light in the worst case.
The third image shows the profile of the mirror taken in cross-section at different orientations rotated around the mirror’s center. Each line is a cross-section, so we get to see all of the deviations in one view. This graphic confirms the visual conclusion from the one above it that the flattening work of the last 18 months has been making good progress. Enough progress to encourage the team to do a star test with the mirror back in the telescope.
What might that star test look like? Here is a simulated star test from the Open Fringe software.
If this image is anything like what we see when we do the star test at prime focus, with the new secondary in place, then we will call ourselves complete with the mirror and go on to get it coated.
This post has gone on long enough, so I will move on in the next post to the topic I promised back in August 2013 – what it means for us to change our optical design to a folded Newtonian.