Butterfly Stars on Tak FSQ-85EDX

[UKRoman]

Hi All,

Although I bought my first serious telescope for visual observing back in 1992, I'm still relatively new to Astrophotography. In the last 12 months I've played with a few small refractors to get the hang of things, including a WO ZS73 and a Redcat 51. I've used both of these with a modified Canon 800D and more recently a ZWO ASI2600MC Pro. Having found my feet somewhat in this fascinating branch of Astro I decided to take the plunge a buy a Takahashi FSQ-85EDX, a dream scope. First light was only brief break in the clouds, giving me an hour on the Horsehead nebula. However after a quick process in Astropixel Processor I managed to draw out a reasonable image for my first attempt with this gear (new ASIAIR pro and EAF too). I'm not totally unhappy with it, but have been surprised to see some strange artefacts to the stars around the outer third of the image (particularly noticeable on the bright star at the bottom left).  At first I thought I'd done something wrong with the processing, but after reading a number of posts on the forums it seems that 'butterfly' stars are the result of vignetting. So I stretched one of my Flats and sure enough there appears to be significant vignetting. To be honest I've not seen vignetting this severe before on either of my other small refractors. 

The imaging train for the attached image was FSQ-85EDX > 1.01 Flattener > M54 to M42 Adapter > M42 Spacers > ZWO Filter draw with Optilong LPro filter > ASI2600MC Pro. 

Although I haven't tried to image with my Canon 800D (also APSC) on the Tak, I have checked the vignetting with this camera and it appears to be around the same level, so I'm guessing I'd see the same artefacts around bright stars. I’ve also swapped the M42 adapters for M48 into the filter draw with the 2600MC Pro but the vignette looks unchanged.

Just wondering, does anyone else with an FSQ-85 see this kind of butterfly stars artefact with an APSC camera? If so, can it be prevented or maybe processed out in Post?

Thanks

 

 

Edited January 17, 2021 by UKRoman
Additional detail

[Earl]

id check that everything is threaded totally true and square as a starting point.

[Earl]

Top left hand corner looks like odd shaped stars also

 

[Seelive]

I would certainly be interested in understanding how vignetting causes that effect, but whatever is the cause, it does appear to act in a radial manor.

[UKRoman]

2 hours ago, Earl said:

Top left hand corner looks like odd shaped stars also

 

Agreed, and yes everything is tightened down in the imaging train. 

[UKRoman]

2 hours ago, Seelive said:

I would certainly be interested in understanding how vignetting causes that effect, but whatever is the cause, it does appear to act in a radial manor.

Definitely radial.

I’ve actually done some more digging around and there are a few threads on CN about ‘butterfly stars’ on the FSQ-85. There seems to be two competing theories. Firstly that this an artefact caused by the Petzval design. The second one is that it’s caused by severe vignetting. 
On the first theory I’m not so sure as I’ve not seen any images taken with an FSQ-106 that have this issue. But then again the image circle of the 106 is so large that maybe it’s not picked up by the vast majority of cameras as they only capture images from the centre of the field. 
On the second theory, this seems plausible as I’ve seen references to exactly the same effect with a Rokinon 135mm lens with a heavy vignette. 
That said, I’m starting to think it might be a combination of design exacerbated by vignette, which is itself caused by the field flattener.  
Either way I’m not sure there’s a way round it with my imaging train setup apart from good post processing. Happy to hear other opinions and suggestions though👍

Thanks

Edited January 19, 2021 by UKRoman

[steppenwolf]

13 hours ago, Seelive said:

I would certainly be interested in understanding how vignetting causes that effect, but whatever is the cause, it does appear to act in a radial manor.

Please note that this is not 'NORMAL' vignetting! To quote from a similar post on SGL in which I replied:-

I think this is a diffraction effect caused by 'aperture vignetting'. This is not the same as normal vignetting but appears because the edge of the image field doesn't receive a full circle from the cone of light through the aperture,  rather, it 'sees' an elliptical shaped aperture.

[UKRoman]

4 hours ago, steppenwolf said:

Please note that this is not 'NORMAL' vignetting! To quote from a similar post on SGL in which I replied:-

I think this is a diffraction effect caused by 'aperture vignetting'. This is not the same as normal vignetting but appears because the edge of the image field doesn't receive a full circle from the cone of light through the aperture,  rather, it 'sees' an elliptical shaped aperture.

That’s interesting, thanks. That ties in with other posts I’ve seen elsewhere.
 

So does this mean that the size of the image circle that isn’t subject to this effect is less than the size of the capture area of an APS-C sensor and a camera with a smaller sensor wouldn’t pick it up? Or is that over simplifying it?

Edited January 20, 2021 by UKRoman

[Stuart1971]

15 hours ago, UKRoman said:

Definitely radial.

I’ve actually done some more digging around and there are a few threads on CN about ‘butterfly stars’ on the FSQ-85. There seems to be two competing theories. Firstly that this an artefact caused by the Petzval design. The second one is that it’s caused by severe vignetting. 
On the first theory I’m not so sure as I’ve not seen any images taken with an FSQ-106 that have this issue. But then again the image circle of the 106 is so large that maybe it’s not picked up by the vast majority of cameras as they only capture images from the centre of the field. 
On the second theory, this seems plausible as I’ve seen references to exactly the same effect with a Rokinon 135mm lens with a heavy vignette. 
That said, I’m starting to think it might be a combination of design exacerbated by vignette, which is itself caused by the field flattener.  
Either way I’m not sure there’s a way round it with my imaging train setup apart from good post processing. Happy to hear other opinions and suggestions though👍

Thanks

With your size of sensor you do not need the flattener, the flat-field of this scope should cover that sensor no problem, they brought out the flattener for the FSQ85 for full frame sensors...so try without that first...also what back spacing do you have from flattener to sensor...it should be 56mm if I remember...?

[andrew s]

4 hours ago, steppenwolf said:

Please note that this is not 'NORMAL' vignetting! To quote from a similar post on SGL in which I replied:-

I think this is a diffraction effect caused by 'aperture vignetting'. This is not the same as normal vignetting but appears because the edge of the image field doesn't receive a full circle from the cone of light through the aperture,  rather, it 'sees' an elliptical shaped aperture.

Can you provide a diagram showing the difference as I am struggling to work out what you mean? Any circular aperture viewed off axis will look elliptical so that's why I am confused.

Thanks Andrew

Edited January 20, 2021 by andrew s

[UKRoman]

1 hour ago, Stuart1971 said:

With your size of sensor you do not need the flattener, the flat-field of this scope should cover that sensor no problem, they brought out the flattener for the FSQ85 for full frame sensors...so try without that first...also what back spacing do you have from flattener to sensor...it should be 56mm if I remember...?

I will definitely give it a go. I've just ordered the correct adapter so that I can remove the flattener from the image train. I was advised that, given the small pixel size of the sensor on the ASI2600MC Pro, using the flattener was highly recommended, even though it's only APS-C rather than Full Frame. I've had the back spacing set to 55mm.

Edited January 20, 2021 by UKRoman

[Earl]

I thought the spacing was 72.2mm or was the the 106?

[UKRoman]

2 hours ago, Earl said:

I thought the spacing was 72.2mm or was the the 106?

The advice I received was 55mm. I’ve had a look elsewhere and some references mention 56mm. I can’t find 72.2mm anywhere though. Maybe that’s with the extender or reducer. 

[Stuart1971]

1 hour ago, UKRoman said:

The advice I received was 55mm. I’ve had a look elsewhere and some references mention 56mm. I can’t find 72.2mm anywhere though. Maybe that’s with the extender or reducer. 

I thought it was 56mm, I’m sure i ha e read that in a few places... 👍🏼

[UKRoman]

56 minutes ago, Stuart1971 said:

I thought it was 56mm, I’m sure i ha e read that in a few places... 👍🏼

I exchanged a few emails with the dealer about it when I bought the scope and I thought it was 55mm. But in any case I’ll give 56mm a go as well. Though my next stop is to try without the flattener. I guess trying these things out is half the fun 🙂.

[Earl]

17 hours ago, UKRoman said:

The advice I received was 55mm. I’ve had a look elsewhere and some references mention 56mm. I can’t find 72.2mm anywhere though. Maybe that’s with the extender or reducer. 

Ah yeah reducer spacing iirc

 

[ollypenrice]

I would certainly lose the flattener because there is no need for it whatever with an APSc chip. You're needlessly introducing more glass and more possibility for tilt. (My Tak 85 focal reducer introduced some tilt till I loosened the three radial screws, pressed the whole assembly down on a table and re-tightened them. Then it was fine.

I'm not really sure what is meant by 'butterfly stars.' I can see the familiar Tak 'inverse lighthouse beams' like this:

My FSQ85 did this and my 106 also does it. It never bothered me much. Greg Parker thought it was pinching of the optics but I don't know.

Could you post a close up of this 'butterfly' effect?

Vignetting: rather than guess at its extent it is easy to measure. Take a linear flat, or linear master flat, and read off the ADU in the corners, then in the middle. The light fall-off on my old fluorite FSQ106 is enormous, about 23%, which is made slightly worse by my using 2 inch mounted filters. A friend's similar scope with 2 inch unmounted filters is a few percent less vignetted but, after application of flats, this matters not at all.

Above all I'd like to know what you feel is wrong with the stars if it isn't the 'beam' effect.

Olly

[UKRoman]

1 hour ago, ollypenrice said:

I would certainly lose the flattener because there is no need for it whatever with an APSc chip. You're needlessly introducing more glass and more possibility for tilt. (My Tak 85 focal reducer introduced some tilt till I loosened the three radial screws, pressed the whole assembly down on a table and re-tightened them. Then it was fine.

I'm not really sure what is meant by 'butterfly stars.' I can see the familiar Tak 'inverse lighthouse beams' like this:

My FSQ85 did this and my 106 also does it. It never bothered me much. Greg Parker thought it was pinching of the optics but I don't know.

Could you post a close up of this 'butterfly' effect?

Vignetting: rather than guess at its extent it is easy to measure. Take a linear flat, or linear master flat, and read off the ADU in the corners, then in the middle. The light fall-off on my old fluorite FSQ106 is enormous, about 23%, which is made slightly worse by my using 2 inch mounted filters. A friend's similar scope with 2 inch unmounted filters is a few percent less vignetted but, after application of flats, this matters not at all.

Above all I'd like to know what you feel is wrong with the stars if it isn't the 'beam' effect.

Olly

Hi Olly,

The new adapter that allows me to remove the flattener from the chain turned up today, so the flattener is now out of the equation. I’m looking forward to trying another imaging session next time we get clear skies 👍 Yep, the effect I’m seeing is the ‘inverse lighthouse beams’. I got the name ‘butterfly stars’ from reading a post on CN that used the name to describe it. I haven’t measured the vignette, but will try the method you suggest, good tip; thanks. I have seen the stretched image of the flats as they were being taken by my ASIAIR Pro and they were a bit of a surprise; the corners were completely black compared to the bright central region. But yes, the flats did their job in post processing. 

In terms of what I feel is wrong with this effect, I guess it’s just personal taste and my perhaps naive expectation. Having paid £3k+ for a premium refractor, I wasn’t expecting to see such obvious artefacts around bright stars. However, the more reviews I read the more I’m realising that this may just be a Tak trademark, perhaps in the same way that diffraction spikes are for reflectors. Next time I’ll do more reading up in advance.

In any case I’ll try some more imaging to see if I can improve on my first attempt. 😊

Thanks everyone for the comments and suggestions.

 

 

Edited January 21, 2021 by UKRoman
Spelling correction

[Stuart1971]

10 hours ago, UKRoman said:

Hi Olly,

The new adapter that allows me to remove the flattener from the chain turned up today, so the flattener is now out of the equation. I’m looking forward to trying another imaging session next time we get clear skies 👍 Yep, the effect I’m seeing is the ‘inverse lighthouse beams’. I got the name ‘butterfly stars’ from reading a post on CN that used the name to describe it. I haven’t measured the vignette, but will try the method you suggest, good tip; thanks. I have seen the stretched image of the flats as they were being taken by my ASIAIR Pro and they were a bit of a surprise; the corners were completely black compared to the bright central region. But yes, the flats did their job in post processing. 

In terms of what I feel is wrong with this effect, I guess it’s just personal taste and my perhaps naive expectation. Having paid £3k+ for a premium refractor, I wasn’t expecting to see such obvious artefacts around bright stars. However, the more reviews I read the more I’m realising that this may just be a Tak trademark, perhaps in the same way that diffraction spikes are for reflectors. Next time I’ll do more reading up in advance.

In any case I’ll try some more imaging to see if I can improve on my first attempt. 😊

Thanks everyone for the comments and suggestions.

 

 

Hmmmm, I have the same scope, and no issues like the ones you see at all....this image is cropped...but stars are good...

 

 

Edited January 22, 2021 by Stuart1971

[UKRoman]

2 hours ago, Stuart1971 said:

Hmmmm, I have the same scope, and no issues like the ones you see at all....this image is cropped...but stars are good...

 

 

 

That's a wonderful image and encouraging for me to see what the scope is capable of. If I can get to this level I'll be very happy indeed. I'll try a session without the flattener to see what difference it makes. Thanks 

[steppenwolf]

On 20/01/2021 at 13:10, andrew s said:

Can you provide a diagram showing the difference as I am struggling to work out what you mean? Any circular aperture viewed off axis will look elliptical so that's why I am confused.

Thanks Andrew

This link may be of some assistance

[andrew s]

1 hour ago, steppenwolf said:

This link may be of some assistance

Thanks very clear but I don't  see how this is different from normal vignetting.  It looks just like an extreme form .

Regards Andrew 

[UKRoman]

Hi All,

So I  was able to get out for a very very brief imaging session last night, 21 minutes on IC443 (7 x 3 minutes - with L-Pro filter). Here is the result without the field flattener in the imaging train.  This is a quick stack and stretch in Astro Pixel Processor. I'm not worried about the image of the DSO itself, that's just lack of data. But thought I'd report back now that I have removed the flattener.

Basically the Inverse Lighthouse Beam effect remains. It's not easy to see as there aren't many bright stars on the edge of the field, but the top right has a few.  Also, the stars in all four corners do seem to be affected by the lack of the flattener. I was advised that this would be the case when I asked the dealer for advice on attaching the ASI2600 to the scope. Apparently this is due to the smaller pixel size, even though it's an APS-C format.

In any case I also attach a screen-shot of a stretched flat (from my iPad) as it was being taken by my ASIAIR Pro, just in case this is relevant for the Inverse Lighthouse Beam beam effect. I've just measured the actual drop-off of a linear Master Flat and the ADU fall-off is 20% from centre to the corners.

Cheers

 

 

Edited January 25, 2021 by UKRoman