Hello Everyone,
First let me start by saying I have never collimated a telescope in my life, I understand the concept, have seen others do it by eye, but never done it myself.

A few months ago I purchased an AT6RC to use a longer focal length scope. I realized this scope would require collimation and for this task I decided to give CCDInspector a try. The presentations I saw made everything look very 1, 2, 3.

Last night was the first night the weather was clear enough since I bought the scope to set it all up and give it a try. Last night was my "shakedown" night, since I had never used the scope taking a picture was the last thing on my mind, last night was about being able to find focus, what adapters to use, and to try to use CCDInspector.

I have briefly gone over the CCDi help files and referred to them a lot last night as I got stuck. I needed to use 60 second exposures in order to get everything nice enough for CCDi to like it. The Star fields had just over 60 stars in each sub.

I only used the Real-Time Curvature Map.

Since I am using the factory included crayford focuser that a lot of people complained about and replaced for a FT, I decided the best way to avoid image tilt was to keep the drawtube in as far as possible. I did this by using spacers so that all of my image train with the exception of the 2" nose piece in the focuser is threaded. The focuser draw tube travel is about 10mm maximum. So it barely comes out...

Using CCDi I discovered I had some image tilt (+3.0, -1.7) but by lightly adjusting the brake and friction thumbscrews together I got it down to -0.1, -0.1 or so.

First question, Is that good???

I then spent the rest of the night trying to collimate the scope and this is when I packed it up and decided to reach out for help.

Using the Real-Time viewing window I also opened the collimation viewer which would display the three collimation screws, a line in the direction of the error, and a readout of I guess the arc seconds of error. My goal and maybe misguided was to get the error as close to 0.0 as possible.

My second question is using the real-time viewer and the collimation viewer, how do I know what screw to turn and in which direction?

Last night if the error was pointing at the "A" screw I would try to turn that one clockwise, recheck, counter-clockwise, check, and look for variations. I never turned more then 1/8 of a turn, more likely it was 1/64 turn, I was very light with it as the tiniest adjustment would make a change.

When that didn't work very well I would try doing "B" and "C" together and leaving "A" alone. By the late hours of the night I was hot, bugs where making me nuts, and I decided to give up before the frustration overcomes all progress.

I would like to know what are the "best practices" for collimating with CCDi so that with my next time out I can pick up where I left off and continue making good progress.

My scope is again an AT6RC, on an Atlas Mount, with a Starlight Xpress SXVF-H9, and SX filter wheel. I am using Maxim for camera control and all of the scope's specifications have been configured into Maxim.

Thank you in advance for your time and assistance.

Hi Jose,


A tilt of -0.1 to 0.1 is very good. Whether this is in arcseconds or pixels, this is negligible and will not be noticeable at all in your images. What this number is saying is that the maximum difference in star sizes from one edge of the image to the other is 0.1 arcseconds (or pixels -- you didn't say what units).

For collimation, the process is a little more involved than what you were trying to do. The knobs and the direction of error are not directly related.

The recommended procedure is as follows, using single-star defocused collimation viewer:

1. Start the real-time viewer, and the collimation viewer and start taking short exposures to get a defocused star as described in the documentation help file, and here: http://pk.darkhorizons.org/CCDI/1StarCollimation.htm

2. Pay attention to the number displayed by the collimation viewer. Your goal is to reduce that number as much as possible. It is nearly impossible to reduce it all the way to zero, simply because of seeing and other system errors. When you get to a number below 5, you are getting close to good collimation. I'm very happy when I get the number to around 1-2.

3. Note the direction of the collimation error. If the direction is approximately the same in each consecutive exposure, you should continue collimating. If the error direction jumps randomly between exposures -- you've reached the limit of what you can do for that night due to seeing and other errors. It's best to stop at this point.

4. When the collimation viewer is reporting an error value and direction that's pretty stable, you'll want to adjust one knob on your scope so that the star shifts in the direction that the arrow is pointing! You may have to try each knob in turn, initially, to figure out the direction that it moves the star. If the star shifts far away from the center of the chip, you'll want to re-center it, and then repeat step 4.

The knobs on the display are there just to help you remember the position and direction of each knob, but the arrow is not pointing to a knob that you need to turn: you'll need to do some initial experimentation to figure that out.

Hope this helps,

-Paul

Paul,
Thanks for your reply and sorry for not posting mine sooner.

I actually did not get another useful night to give your suggestions a try until 32 days after installing the software so I just ended up buying it.

I tried again and made more progress but the seeing was just too poor to make any real improvement.

I had a really hard time with the "star is to distorted" or "star is to dim". It was hard to find the sweet spot where CCDi seemed happy with the star in the center.

I tried to help a friend on Friday with his HyperStar collimation but it seemed to be even more testy with such a huge field/short focal length.

I'm sure it's just my inexperience with the software.

any pointers you can give me I would be greatly appreciated.

Off topic: I also recently purchased CCDStack, can you please tell me what I should do to install the CCDis plugin into CCDStack?

Thanks again,
Jose

Hi Jose,

Try to exclude other stars from the field of view. Center a sub-frame on the center of the chip, and size it to include just the bright star you want to collimate on. Leave enough room around it so that you don't lose the star when adjusting collimation, but try to make sure there are no other bright stars in the sub-frame. Try not binning the chip first, just find a star bright enough to get ADU level to 20,000-40,000 adu maximum.

If you have CCDInspector already installed on your computer, CCDIS will show up automatically under CCDStack image registration dialog as the first tab. You don't need to do anything to enable it.

Regards,

-Paul

Paul,
When you say to size the sub-frame to just include the bright star I want to collimate on do you mean that I should use Maxim to create a square around the star, much like I would do to single out a star to focus on?

I tried to do that but in CCDi the smaller frame was all white, I think that my have been cause by the poor sky conditions.

Thanks for al the suggestions, I'm going to give this another try soon.

About the CCDIS plug-in it did show up automatically, I just didn't check, pretty cool, thanks.

Clear skies,
Jose

Jose,

Yes, create a square subframe around the center of the frame (it's important that the subframe be centered in the frame!)

Make it larger than the defocused star by at least 15-20% (more is better) to leave some space for the star to move when you make adjustments.

Regards,

-Paul