View Single Post
Old Jul 24 2008, 12:39 PM   #39
StevenB
Bitsmith/Starsmith,
Master Sneak
 
StevenB's Avatar
 
Join Date: Apr 2007
Location: Scotland
Gender: M
Fan of: Skies of Pern: what else?
Now Reading: My computer screen
Default Re: The VISTA Telescope Thread (VTT)

Hello folks. It's 12:20pm here, and I'm up and getting ready to go to lunch and start another long day/night's work. Yesterday I worked from 11am until 3pm. I'm not going to do that again . We're gradually moving from day shift to night shift by extending the finishing time and getting up a bit later each day. When commissioning you need to work day and night - night to test the telescope and day to fix the problems you discovered the previous night. I need to be cloned .

I spent another night working on the active optics. Here are a couple of examples to show you what I was looking at last night:

Above is a diagnostic image from the VISTA High Order Wavefront Sensor. There are 6 different images. The top two are out of focus pictures of a real star as it appeared on the wavefront sensor. The left hand one is on the near side of focus and the right hand one is on the far side. These images appear like doughnuts because they are showing the secondary mirror unit (the big silver grey cylindrical thing on the earlier picture of the telescope) blocking the light from the centre of the primary mirror. The second row of two images shows the software's best fit to the two real images. The best fit corresponds to a set of distortions that the software thinks it has found in the mirrors. The bottom show shows the difference between the real images and the fit, and the picture below shows my High Order Wavefront Sensor control screen. The bottom right corner shows the set of mirror distortions calculated by the software.

In theory, the software then tells the telescope mirrors to adjust themselves and we get perfect images. In practise, this takes a lot of debugging because every parameter might contain a hidden sign error or rotation error. I'm still persevering.

Tonight we plan to take another look at the telescope pointing model. Our telescope expert comes back from his descanso (brief holiday) today. The pointing model tells the telescope what to do in order to point to a particular star. Last night we kept having to adjust the telescope manually, so getting the model up to date will save us a lot of time.

Time for lunch. I'll bring more news the next time I get some free time.
All the best everyone.
__________________
Steven
StevenB is offline   Reply With Quote