Collimation v3

Somewhere to ask questions about the best way to use SharpCap
Forum rules


If you have a problem or question, please check the FAQ to see if it already has an answer : https://www.sharpcap.co.uk/sharpcap-faqs
Post Reply
Tommohawk
Posts: 19
Joined: Thu Sep 28, 2017 10:57 am

Collimation v3

#1

Post by Tommohawk »

Hi. I've just got a "new" second hand SWQuattro and spent a while collimating this - I reset the focuser angle and collimated secondary and primary. I was quite pleased with the result using the cheshire, but when I checked it out with the SC collimation tool (with coma corrector) it showed some issues.

I couldn't quite work out what was supposed to be happening - it seemed to be cycling through several patterns: ellipse, hyperbola and something else I think. The ellipses on the screen were definitely off centre, and there was a hyperbola type plot which also looked off centre. Also I wasnt sure if I was supposed to be able to adjust and check collimation "live" so to speak. I tried a tentative twiddle of the primary and it shifted things sideways but not much change in the pattern. I guess that would need the secondary tweaking but a bit fiddly to do this in the field with scope pointing up.

Obviously I can't do this other than with clear sky, so I tried to replicate this with the test camera. It shows only a hyperbola plot and only if the smoothing is turned up to 60 or beyond. I guess thats because the test camera/scope is well collimated.

I realise this feature is experimental but it certainly seems to have some practical use. If you could give some guidance on what is supposed to be happening it would be good. From a practical POV the constant cycling through different patterns made it difficult to guage the effect of alterations becauee the hyperbola pattern was difficult to read. I had to wait til it came back to the ellipse before I could judge what was happening.

Also I wonder if it would be better to not use the coma corrector? If the CC isnt perfect this could skew things, and for the purpose of preliminary set up it might be better to include coma in the periph. Any thoughts?
User avatar
admin
Site Admin
Posts: 13173
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Collimation v3

#2

Post by admin »

Hi,

so the basic theory is that with a newtonian, the stars near the point where the optical axis hits the center will be sharpest and those away from that point will be fuzzy (hence larger) due to coma. In theory, perfect collimation would give the sharpest stars in the center of the image and you would see a nearly circular pattern that is perfectly centered on the center of the image.

If you have alignment issues then the pattern will still be circular, but offset - in theory you should be able to make adjustments to bring it back to the center and then you will be collimated.

If you have tilt issues then you will start seeing ellipses or other shapes (perhaps hyperbolae). If the tilt is left/right then stars to the left and right of the image would be bigger than those the same distance from the optical axis in the up/down direction - because the left/right stars will be slightly out of focus due to tilt. A slight tilt left/right should lead to a tall thin ellipse instead of circles.

I found it did work better without a coma corrector as detecting the coma is fundamental to how it works - if you make most of the coma go away then there is less data for SharpCap to work on. It will also likely work better with big sensors than with small ones.

I actually got rid of my newtonian telescope (it was too high-maintenance for me), so it's unlikely that I will be putting any more time into improving this feature in the near future.

cheers,

Robin
Tommohawk
Posts: 19
Joined: Thu Sep 28, 2017 10:57 am

Re: Collimation v3

#3

Post by Tommohawk »

Hi Robin and thanks for that- it all makes sense.
One follow on thought is that I did this very much basic testing for a new Quattro, and I didnt even have the mount powered up - I simply pointed near the pole to minimise any movement. I was only doing very short subs, maybe 5-15 secs - I cant remember TBH. Maybe even shorter. Of course what this means is that those stars nearest the pole (NCP) would move less than the others which would affect their FWHM - this may have mucked things up. That said the key thing is that the CC will lose (in theory) all the coma so as you say better to remove the CC if using SharpCap for collimation.
In fact one of the things I was most interested in is if the sensor (ASI1600) was out of alignment, and to that end I tried it again the following night with a camera lens. The collimation feature still showed some patterns though these were erratic and low strength, which I took to mean there was minimal error. The images I took showed pretty good star shapes, though not perfectly symmetrical in the corners. I think that's down to minor issues with the lens though. What I should have done was taken some images with the Quattro and simply given them a visual check.

FWIW I think this feature does have real potential, perhaps not so much for collimation of Newtonians, but more for checking sensor alignment/orthongonality esp with larger sensors becoming more widely used as they get more affordable.

Anyhow just notice the next version is available so going to try that out - keep up the good work, and thanks again!
dthomson39
Posts: 2
Joined: Tue Mar 20, 2018 4:48 pm

Re: Collimation v3

#4

Post by dthomson39 »

admin wrote: Wed Jan 17, 2018 10:47 pm Hi,

so the basic theory is that with a newtonian, the stars near the point where the optical axis hits the center will be sharpest and those away from that point will be fuzzy (hence larger) due to coma. In theory, perfect collimation would give the sharpest stars in the center of the image and you would see a nearly circular pattern that is perfectly centered on the center of the image.

If you have alignment issues then the pattern will still be circular, but offset - in theory you should be able to make adjustments to bring it back to the center and then you will be collimated.

If you have tilt issues then you will start seeing ellipses or other shapes (perhaps hyperbolae). If the tilt is left/right then stars to the left and right of the image would be bigger than those the same distance from the optical axis in the up/down direction - because the left/right stars will be slightly out of focus due to tilt. A slight tilt left/right should lead to a tall thin ellipse instead of circles.

I found it did work better without a coma corrector as detecting the coma is fundamental to how it works - if you make most of the coma go away then there is less data for SharpCap to work on. It will also likely work better with big sensors than with small ones.

I actually got rid of my newtonian telescope (it was too high-maintenance for me), so it's unlikely that I will be putting any more time into improving this feature in the near future.

cheers,

Robin

Hi _ am a new user of Sharpcap and like many I am really impressed - lots of hard work to get this far!

I note in the above post reply that you said you were not doing any further work on the Collimation assistant - is this the case? Is the current version usable and/or effective - I can see a real value to those of us using Newtonian scopes to run a routine like this at the start of a session just to do any tweaking that might be required - pity if its not going to be developed further if its not quite there yet.
User avatar
admin
Site Admin
Posts: 13173
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Collimation v3

#5

Post by admin »

Hi,

yes, sadly this feature is not under development any more - I only ever got decent results with it when *not* using a coma corrector, and even then it was hard to adjust - mind you I always found collimation hard....

I don't have a newt any more so I have no test bed to develop this feature further I'm afraid.

Glad to hear that you are enjoying the rest of SharpCap though.

cheers,

Robin
Post Reply