I am trying to fine tune my imaging train to get the best performance I can. Please have a look at the image sections in the attached file. I can see slight differences in the amount of star elongation in each corner ranging from essentially zero in the lower RH corner to pretty noticeable in the upper RH and LH corners. I am wondering if I need to work on tweaking the backfocus of my field flattener, or whether I should try adding shims on one or more corners of the dovetail faceplate of the camera, or maybe a combination of shims and backfocus. What is the methodology for adding shims ? I have a supply of ring shims in 0.001" and 0.002" thicknesses. My Camera did not have any shims when it was delivered.
An STXL16200 is a fairly large chip, and it is normal to have some off-axis aberration even on the best telescopes... but your images don't look symmetrical. So I suspect you're right that something needs to be tweaked. We measure the tilt of the sensor at the factory, and provide shims if needed to level things out. Your camera did not require any. But it is of course quite possible for other components in the system to contribute a slight tilt. The easiest way to adjust that out would be to add shims. Looking at your images, the Upper LH and Lower RH images have the most out-of-round stars - and the angle is in the same direction. That does suggest a tilt in that direction. So yes, I would try adding some shims to see if you can level this out. It will take some trial-and-error; for one thing, it's impossible to tell which direction it's tilted until you try adjusting it. I'd try, for example, raising the Upper LH corner by .001" and see if it makes it better or worse. And iterate from there.
Thanks Doug So would the upper LH corner image would correspond to the upper RH corner of the camera faceplate (position A on the shim placement sheet), or are they reversed ?
LOL I always get confused about sensor orientation myself! Telescopes and lenses tend to invert the image, and the documentation is all relative to the front of the chip, which is looking out of the telescope. It's easy to get turned around. Pin 1 of the sensor is in corner A. Pixel 1,1 is therefore at corner C, i.e. the sensor reads out starting from the lower left corner. The image on the computer screen lines up with the pixels as viewed from the front of the camera. I'd start by playing with shims on the A-D axis.
Hi Doug I thought I would give you a quick update on this issue. I had a stretch of clear nights this week that gave me the chance to do some experimenting. On the first night (12/07/2020) I added a 0.002" shim to corner A. For the next two nights (12/08/2020 & 12/09/2020) I removed the shim and rotated the faceplate clockwise by 90 degrees each night. For the reference point I assigned corner #1 of the faceplate as the corner in position A on the shim chart (1-A). The attached file has some screen shots of the image corners with the different configurations described above, along with the baseline image from several weeks ago. From what I can tell, the images from last 12/09/2020 seem to have the least and most consistent distortion of the stars in the corners. Next time out I may rotate the faceplate one more time so that I can evaluate the images with the faceplate at all four corners, but from what I have seen so far the best performance is with no shims and the faceplate in the position 1-D relative to where it was when I started. I also need to mark it so that I will put it back in the correct position next time I remove it. This brings up a good point. I think that it would be beneficial to add a reference mark to the faceplates on your cameras that is indicated on the shim chart, and warn all the users that if they remove it, they should put it back in the same position. The camera manual mentions to put the shims back in the same position, but it should also mention that the faceplate orientation must also be in the same position. I have also attached a JPEG of one of the images from last night. Unfortunately, the fog rolled in just about the time I was getting started, and there were very few dim details captured in the images.
The adapter plates have tight machining tolerances and it generally doesn't matter if you rotate them. You are probably just seeing changes because of minor focusing differences. Concentrate on levelling via the shims. Also please note that if the tilt is external to the camera, then rotating the camera will impact the leveling.
