Strange Flats - Atronomik Deep Sky RGB

[furgle]

I've been using Astronomik Type 2c RGB filters for a couple of years and just recently switched to the new Deep Sky RGB filters.

I'm getting these very strange flats where they all have a bright spot in the middle, instead of a steady gradient. It's evident in my light exposures as well (as a dark area). When I apply the flats to the light frames, the darkness isn't fully flattened out though, so I'm getting very frustrated.

Am I doing something wrong here? Has anyone encountered this before?

[Camelopardalis]

Adam, that is weird, I use the Deep Sky with my ASI1600 and I don’t see that problem.

[furgle]

Searching google turned up this thread: http://www.iceinspace.com.au/forum/archive/index.php/t-80648.html

There is a post by Terry B saying that anti blooming ccds like mine tend to need a much lower exposure. I've been aiming for 80% full well average in the centre 100x100 square. I think my flat frame exposures are too long.

Since it's cloudy again and everything is set up as it was, I'll give it a test tonight.


Edit: Set maximum to 9000 and still the same. Le sigh.

[furgle]

This seems to clearly be an issue with the Deep Sky filters.

I've now taken flats of all my filters and only the Deep Sky ones show the middle spot.

The Astronomik Type2c Luminance filter (attached below image #2) is fine, the Astronomik 6nm OIII filter (attached below image #3) is fine, but the Deep Sky Red (post above) shows a spot, as does the green (attached below image #1) and blue.

[gregbradley]

The purpose of a flat is to remove unevenness of illumination in your images.
Unless you have a bright white centre in your light than this flat is obviously not going to work.

I wold first - clean your filters they are extremely dirty (better to have a clean filter than rely on flats to remove dust donuts) and reduce the exposure.
A microfiber cloth is good for cleaning filters.
80% of full well is very high I think 30% is often a bit high but roughly what I am for. Some scopes do have a hot spot in the middle of the optics like that. I think its more common now as mirrored scopes almost always have a lens corrector.

Greg.

[baileys2611]

I've read only a little about well capacity. How does one go about calculating what your full well % is when taking a photo?

For instance, I'm going to be experimenting with a QHY10 in the near future (when it eventually comes on shore) and from their documentation, they state a 45ke- full well capacity.

Que?

[RickS]

The camera gain should normally be set so that full well in e- is around 65,535 ADU (maximum 16-bit unsigned integer.) So, half well will be around 32,000 ADU, 2/3 well is around 43,600, etc.

If you want to do the calcs from first principles, just divide the number of e- by the gain to get the value in ADU. As an example, the full well for a KAF-8300 is around 25,000 e-. Typical camera gain for this sensor would be around 0.4e-/ADU. To target 50% of full well, that's 12,500 electrons. 12500/0.4 gives 31,250 ADU.

IIRC, the QHY10 has some magic gain and offset numbers that you can tweak. There's a relationship between what they call gain and the actual camera gain, but I don't think they are the same.

Cheers,
Rick.

[baileys2611]

Ah! I think I see!

So when I'm setting gain, I normally do this based on an exposure on a white surface at 0.01 seconds exposure, then when the initial data signal shows a peak at around 500-700 on my histogram I generally set gain at that point.

Is it something like calculating the % is then the target for the 'stretch' point on that histogram (i.e. where the white point is set) taking into account the well depth of the camera per your excellent description below?

[Bart]

http://www.stark-labs.com/help/blog/files/GainAndOffset.php

[bigjoe]

Don't do much imaging myself..but heard these can be caused by ghosting from the big lens of a not so perfectly spaced reducer /corrector.
bigjoe

[baileys2611]

Yeah, I was way off. Thanks for the link.