RGB Filter problem - misaligned stars. Help Please!

[Tommohawk]

Hi all.

Well I've worked through (and resolved with the help of SGL members) a good few queries here over the years but so far this one has me beaten. 

Some time back I bought an Astronomik Deepsky blue filter, to try and resolve the blue bloat with my Sharpstar 61Mk1 scope (+ 0.79 reducer/flattener) and ASI1600 cool. This seemed to fix the blue issue nicely, but being a different thickness to the previous ZWO filters it was miles off parfocal. In an effort to resolve this I bought the matching R and G filters together with the L3 to sort the equivalent bloat in the luminance.

Last night was the first clear night to check this set up, and I decided to keep it simple and just do the Beehive, just for star tests really. The result is rather odd. With the red image, following stacking (with either APP or DSS) some of the red stars -  but not all - are slightly misaligned. When I compare the red image to the G and B, the brighter stars are not correctly aligned, although all the dimmer ones are fine. I thought I knew optics pretty well but I cant figure this at all. Ive attached an RGB image, which I've crudely processed to show the problem. If you blink between the RGB channels you can hopefully see what I mean.

I've checked the raw subs and the same applies - its not a processing artefact. I've also checked carefully and don't have this on any or my previous images, so it does definitely looks to be to do with the red filter.

Does anyone have a clue what's going on? The only thing I can think is that the red filter faces aren't quite parallel, such that there is displacement which somehow affects the brighter stars more.  

[vlaiv]

I don't think there is issue with red channel - I think there is issue with green and blue.

It looks like collimation issue - or something similar.

Look at these stars for example - this is montage of R, G and B channel (from left to right).

Red stars are nice looking, their peak is centered on star profile. Green and blue are distorted. They have center of the star displaced with respect to star profile. Stacking works properly and aligns star centers since those carry the most "mass" - peaks should be aligned.

Problem is "skirt" that is not symmetric in green and blue.

I've seen this problem before in simple photographic lens - as you move away from the center - you get stars that look like that.

To be honest - I don't think above is down to filters. To me it looks like poor optics / collimation issue. I remember @Ken82 having similar issues on his scope.

[Tommohawk]

24 minutes ago, vlaiv said:

I don't think there is issue with red channel - I think there is issue with green and blue.

It looks like collimation issue - or something similar.

Look at these stars for example - this is montage of R, G and B channel (from left to right).

Red stars are nice looking, their peak is centered on star profile. Green and blue are distorted. They have center of the star displaced with respect to star profile. Stacking works properly and aligns star centers since those carry the most "mass" - peaks should be aligned.

Problem is "skirt" that is not symmetric in green and blue.

I've seen this problem before in simple photographic lens - as you move away from the center - you get stars that look like that.

To be honest - I don't think above is down to filters. To me it looks like poor optics / collimation issue. I remember @Ken82 having similar issues on his scope.

Thanks for wading in -  that's a really useful observation. One thing about this hobby is it's so difficult trying to test anything because there just aren't enough clear nights, so goodness knows when i'll be able to do further tests - it's infuriating. 

I looked back over all my imaging for the last 14 months ie since buying the blue filter, and some was done binned, some in narrowband and some with a newtonian, but I only have two unbinned images done with RGB and the refractor. I dont think either of them shows this issue to any degree, even though they were both done with the same blue filter, together with  the ZWO red and green. 

Ive posted below the previous RGB images, M31 and M45. Also posted single subs of RGB for this this image of The Beehive - these arent aligned though. If you  could have a look I'd be grateful.

Maybe the optics were pinched or something last night - it probably didn't cool down for long, and I did the red last, so maybe that has a bearing on it?

Only other relevant point is I did red after meridian flip -  maybe something funny with image plane?

[vlaiv]

35 minutes ago, Tommohawk said:

Ive posted below the previous RGB images, M31 and M45. Also posted single subs of RGB for this this image of The Beehive - these arent aligned though. If you  could have a look I'd be grateful.

M45 looks fine indeed. M31 - shows similar effect on some of the stars. It's not easy to see - but I think it is there.

For example in this zoomed in patch:

Or maybe here on the other side of the frame:

37 minutes ago, Tommohawk said:

Maybe the optics were pinched or something last night - it probably didn't cool down for long, and I did the red last, so maybe that has a bearing on it?

Only other relevant point is I did red after meridian flip -  maybe something funny with image plane?

All valid points. Both temperature and gravity can have impact on the lens.

What sort of flattener are you using and does it have threaded connection to the scope? Just went back to original post - it is x0.79 FF/FR. I have this FF/FR and I had issues with 2" clamped connection - it would tilt as the scope tracks the sky.

Here is green sub:

That is bottom left corner - stars are defocused - clear sign of either field curvature or tilt.

Bottom right corner shows astigmatism:

Ok - blue shows similar thing - bottom left is defocused, bottom right is astigmatic - but to lesser extent than green. My guess is that you shot green first, then blue and after meridian flip - red.

There is a bit of play in FF/FR and I think there is a bit of tilt as well since even Red is suffering from this - but in top corners (meridian flip - 180 degrees rotation).

Here is red sub - top right corner:

Not as pronounced but still there - focused core and defocused skirt.

Top left corner of red sub is rather tricky - stars are faint and it is hard to tell, but I think there is a bit of astigmatism there as well:

I'd say - FF/FR is first to be checked. Maybe switch to threaded connection if you don't have it already or think of getting tilt plate to fix any tilt?

 

[Tommohawk]

I'll have a better look at this tomorrow. You're right I did green first then blue then red. 

Pretty sure the FR is screw connection both sides... I'll check.

Thanks for your insight.

[vlaiv]

Just thought of it - what is your focuser like on that scope, could it have any play in draw tube?

If your FF/FR is screwed in on both sides, I doubt that it is causing this - you need to check for anything that might move and focuser is next suspect.

[Tommohawk]

Trying to think through the implications of all this. Because the star centration seems reversed for R compared to BG, and I fIipped before doing R, I think this must mean that as you say its a gravity induced thing rather than a scope collimation issue.  

I've always considered the focuser to be very solid -  but today I have checked it and there is some play for sure. Not much but discernible. Funny thing is this focuser has 2 screws, a small silver one (blue arrow) aand a larger brass coloured one (red arrow)

I'd always assumed the smaller one is tension adjustment and larger one is focus lock. But heres a funny thing. If I tighten the small screw only, the tension becomes stiffer as expected, and there is minimal if any play in the draw tube. If I tighten the brass one this seems to have no effect at all ie it doesnt lock it at all. BUT once tightened there is then play in the drawtube, which reduces when the brass screw is loosened. Not sure how that can be!

Edit: Just one tihng to add. I'd forgotten about the brass screw underneath (I dont get much chance to get out one way or another, and also memory not so good!) and it was done up pretty tight when I checked it today. But I've been merrily focussing away with this screw "locked" for some time - many sessions. So maybe if this has a friction material on screw tip it's gotten worn smooth)

Edit #2. But I guess if its going to cause play somehow I dont want to use it anyhow!?

So I'm hopeful that I'm homing in on a solution, just need to figure how to sort the focuser. Any thoughts??

Edited March 8, 2021 by Tommohawk

[vlaiv]

3 hours ago, Tommohawk said:

So I'm hopeful that I'm homing in on a solution, just need to figure how to sort the focuser. Any thoughts??

I think that you are right - focuser seems to be the most likely candidate for tilt. Not sure if I can offer any sensible advice on how to adjust it as I don't have much experience in that department.

[Tommohawk]

Ok well I'm going to be completely pragmatic. The "tensioner" screw does a pretty good job of locking it when done up tight, so I'll try just using that and leaving the "lock screw" undone.

I have done some star testing in the past with an artificial star which seemed to work ok although that was with a Newtonian and maybe not so precise. I'll report back when done.

Thank again for pointing me in the right direction 

[vlaiv]

1 minute ago, Tommohawk said:

I have done some star testing in the past with an artificial star which seemed to work ok although that was with a Newtonian and maybe not so precise.

You can use this to quickly check for tilt or any issues - SharpCap and artificial star.

Slew scope so that star image ends up in each corner and use focusing aid that measures FWHM. It should be roughly the same in each corner - if not, you have tilt.