Vignette on ASI 6200 Pro Mono

Bernardo

Hello friends !

I have an ASI 6200 Pro Mono with a ZWO filter wheel on a Newton 250/1000 mm at 100 Gain. All images are calibrated with Masterdark and Masterflat (Flatdarks subtracted). The connection of the camera and the FW is unfortunately only M53, which results in vignettes in the corners. I have tested flats with different ADU and gain without success. The screenshot shows the usable area of the image field. What can be done to get rid of vignetting ??? I have an ASI183 Mono and an ASI 1600 Mono, which deliver absolutely clean images over the entire field of view!

wvreeven

Bernardo I use an ASI6200MM with a 2" filter wheel and have strong vignetting in the corners as well. The solution is proper flats. If you still see vignetting after having taken flats then I am afraid that the flats simply aren't good enough. It can be very tricky to get the flats right so you need to be patient there and try many different sources of light before you may be able to get properly corrected lights.

I usually take my telescope inside and point it at a white wall or white towel from up close. That means about 50 to 80 cm from the wall or towel. Then I regulate the light by opening and closing doors and curtains until I get the proper illumination of around 55.000 ADU peak with 2 to 10 sec exposures. That way the full bit depth of the pixels gets used and I get the best vignetting correction. Though today I took OIII flats that peaked around 33.000 ADU and that worked as well.

With respect to the ASI183MM and ASI1600MM: those have much smaller sensors that fit well within the yellow rectangle in the images in your post which is why you don't have any vignetting issues there.

w7ay

Bernardo I have an ASI 6200 Pro Mono with a ZWO filter wheel on a Newton 250/1000 mm

Yes, the f/4 system will be problematical with M54 attachments (I think your 53mm is a typo).

At f/4, every mm that an aperture is away from the sensor surface requires 0.25 mm of extra aperture over the diagonal of the sensor, which is 44mm.

The inside diameter of M54 tubes are typically 50mm (I measure 49.9+ mm for the Blue Fireball tubes that I have).

That leaves 6 mm (50-44) of aperture. At f/4, it means that any M54 tube that is more than 24mm away from your sensor will cause vignetting. In practice, you are going to get aperture vignetting.

With an f/8 system, a 50mm opening can be 48mm away before vignetting, so can probably be flattened.

Basically, M54 and a full frame camera are really not compatible, especially so for faster optics.

Even if you can flatten the illumination, aperture vignetting causes a cats-eye effect (I have seen "almond shaped" term used also) that will cause bright stars to exhibit notches in its diffraction pattern. Even if you can flatten, your bright stars that are not near the center of your field will still look bad. Check the first (to see the effect) and the last post (to see the explanation) in this thread:

https://www.cloudynights.com/topic/547028-anyone-know-what-causes-this-artifact-on-bright-stars/

Chen

w7ay

Bernardo ZWO filter wheel

By the way, check your filters too, if you are using mounted filters. Many of the mounted filters only have a clear aperture of 43mm.

Unless you are using a reducer/flattener and its exit pupil is less than 43mm, you will be guaranteed to suffer from vignetting.

My FSQ-85's flattener glass is 40mm in diameter, so I mount a filter drawer as close to the flattener as possible, instead of closer to the sensor, since the 6200 sensor has a larger diagonal than the flattener. Do not go by the "rule of thumb." Find the ray diagrams for your specific optical train and base your filter location on that.

I had asked ZWO about filter drawers and filter trays for M52 filters (like the filters from IDAS), but they are not interested. They will eventually be forced to, if they plan on selling more full frame cameras. I don't think they had fully thought out their devices before manufacturing them.

In the meantime, try to use larger filters, trays and wheels from other manufacturers, or at least use unmounted filters. Borg, for example has a filter drawer (for M57 tubes) that holds IDAS 52mm filters. So, both the tube and the filters are larger.

My ASI6200 is mostly used as a special purpose camera. I use the ASI2600 (both Bayer and monochrome) for normal DSO work.

Chen

Bernardo

Hello friends, thank you so much for the interesting hints. The setup contains a korrector and a focuser of 3 inches, the only problem is the small passage on the camera and the FW. The filter is 2 inches and unmounted, but all adapters to M68 minimize the opening. I don't understand why ZWO have no M 68 adaption for that full frame camera. But we have to deal with. I will try the flats to 55 000 ADU, perhaps this is better, normally i use it between 30000 und 40000 ADU.
Another question for you is the exposure time, do you have short exposure of 60 or 120s ? My exposure in the example was 240s, is this too much to a sharp vignette ? Best wishes,
Bernardo.

wvreeven

Bernardo I usually shoot at 60s and only occasionally at 120s. I never thought about how the exposure time would affect vignetting and you may have a point there.

Bernardo

Okay, i thougt about the next night to 60s exposure as a attempt and i will also make flat with high ADU. Tomorrow i send a replay about the results. CS Bernardo

Bernardo

As promised, now my results. Thank you for your answers and they helped me a lot:
In addition to the longer exposures, I took 60s pictures and processed them with darks and the 50T ADU and 55 T ADU flats (flatdarks subtracted). Unfortunately without better results.
The corners remain and need to be cut off as shown earlier.
I have flats from 1000 to 55T ADU and the picture is always similar without removing the vignettes regardless of the exposure time. The pictures are best with an exposure of 180 or 240s but with the vignettes unchanged. I have to accept that the image field cannot be used in full size. The pictures will be good anyway.

It would be nice if the technical department of ZWO could give an opinion.

Kind regards and clear sky
Bernardo

Tech@ZWO

Bernardo It would be nice if the technical department of ZWO could give an opinion.

For 6200, there is always vignetting which is normal and you can only cut it in the post-processing but no better solution to eliminate it.

sHuRuLuNi

Tech@ZWO You cannot simply say that is "normal". How can it be normal? You are selling a flagship camera which costs several thousand euros but then you have no filter wheels or OAG or other products that actually support the said camera.
So, at the end you have paid 3 times more for a camera which performs like the cheaper one. So, what is then the point of buying the 6200?!

I too wanted to buy the 6200 because of its huge resolution, but when I read stuff like this I get very frustrated really, and that is the reason I am not buying it.

Didge

I just bought a 6200 ... I have to say it's a HUGE disappointment.

This camera displays HUGE vignetting - it is in the CAMERA - not the optical train.

I set the camera up on an Askar 130 which has an image circle of 60mm. All the intermediate elements are M54 - so I wasn't expecting any vignetting at all. After a few exposures, I was perplexed by the significant vignetting, so dismantled the optical train piece by piece to try to identify the culprit. I eventually got down to just the camera - no filter wheels, no OAG, no tilt plates - just the camera - and the vignetting was still there.

Then I set the camera (with no attachments) up on a flat panel and took a number of exposures at various settings.

THE VIGNETTING IS IN THE CAMERA.

The difference in brightness from the centre of the sensor to the corners is 30%.
THIRTY PERCENT !!
That is rubbish performance.

If course I know vignetting can be managed using Flats - and that is fine for vignetting introduced by elements in the optical train.

But I DO NOT expect vignetting in the camera.

Vignetting in a camera is NOT normal - that is rubbish.

Not for a camera that cost me USD 4,700.

I would NEVER accept this terrible performance in a Nikon camera - or even in a consumer camera that cost fifty bucks. Why should I accept it in a specialist 'Pro' camera that cost a hundred times a much ???

It is going straight back to the shop for a refund.

w7ay

Didge the vignetting was still there.

Image circle does not imply that you have 100% iluumination all the way to the edge of the circle. That is not how it is defined. Most of the time the image circle is some reduction factor (50%, 25% or even complete extinction) of the photons. Light falloff can come from aperture vignetting, for example.

See here (especially the text comments under the picture in that page):

https://en.wikipedia.org/wiki/Image_circle#:~:text=The%20image%20circle%20is%20the,of%20light%20–%20the%20image%20circle.

If you want no vignetting to the 44mm corners of the camera, you will need to find a better OTA. Even an FSQ-106 will not 100% illuminate to 44mm diagonals (you can find light falloff diagrams from Takahashi -- many other manufacturers don't even dare publish them).

Chen

Didge

Hi Chen,
Thanks for your quick reply ...

I unreservedly withdraw what I said before - of course I now realise that putting the camera on a flat panel will never provide an evenly illuminated image, because the corners receive less illumination than the centre due to the sensor boundaries and the inverse-square law ... it would be different if the light was collimated, or the flat panel was directly on the sensor.

And I take your point regarding the image circle ... although I had always assumed that it was defined as the boundary within which illumination was 100%.

I will re-test the camera using light collimated through the telescope, with the camera attached directly to the OTA with no intermediate items such as OAG or filter drawers.

w7ay

Didge And I take your point regarding the image circle ... although I had always assumed that it was defined as the boundary within which illumination was 100%.

Yeah, there is always a light falloff. Here is the plot of the light falloff graph for the FSQ-85:

(You might be able to write Askar to see if they are willing to send you the curve for their OTA.)

Notice that by the time the FSQ-85 gets to a 44mm diameter (the 21.7mm from optical axis in above graph), it has fallen off to 70% already.

The only times I can avoid taking flats on my FSQ-85 is when I take images of small galaxies (or when I don't care :-) :-). For any larger targets on the 455mm focal length scope, I need to use flat frames, otherwise I will inevitably get gradients on my images, or having to clip the dark end of the image.

Takahashi spec'ed the FSQ-85 as having 44mm image circle. I think Takahashi must be using 3 dB as the criterion, which corresponds to 70.7%. At least they are honest and publish what the falloff is like.

IMHO, light falloffs are not really a terrible problem, especially if the sky background is bright. The falloff will affect both background noise and the target equally. So, if you are sky noise limited, the SNR does not change at all once you apply flats. The SNR will change relative to dark current noise and read noise, but if you have exposed for long enough to have SNR of the target to be even just 10 dB or 13 dB over the read and dark current noises, you really have no practical SNR loss after applying flat frame corrections.

My recommendation is to know that the dropoff is there, but not to be overly concered about it.

Chen

Didge

Hi Chen,
Many thanks for that - yes it all makes perfect sense.

As my OTA is an Askar 130PHQ and I understand that ZWO sells these as their FF130-APO,
perhaps you can provide the light drop-off curves for the FF130-APO ?

w7ay

Didge you can provide the light drop-off curves for the FF130-APO ?

You will need to ask ZWO. I am a retired Apple engineer who lives in Portland, Oregon, USA. Never, ever even travelled to China (had airplane stopovers in Hong Kong a couple of times when it was still British territory.)

Chen

Didge

Thanks Chen,
Ah - I see ... apologies, I thought you were from ZWO.
Many thanks again for for posts - they were extremely helpful.
FYI - I used to work for Apple too - I was MD for Australasia Developer software.

w7ay

Didge I used to work for Apple too - I was MD for Australasia Developer software.

Ah, very good. I joined in 1988, worked myself up to a DEST at Apple in 1996, then retired in 2005. (Only Apple employees know what a "DEST" is :-) :-).

Chen

Didge

That would be a Distinguished Engineer, Scientist or Technologist 🙂

w7ay

Didge

Shhhhh :-) :-)

You can see the business cards here:

http://www.w7ay.net/chen/Others/More/apple.html

Chen