Help Needed With Noisy Flats

[russellhq]

I'm having a bit of bother with my flats, the seem to be making my image a lot noisier.

Here's a single flat:

Here's the master flat from DSS (made from 20 flats)

Here's the stretched stack without the flat

And here's the stack with the flats

To my eye, the stacked file with the flats looks noisier compared to the stack without the flats.

Any ideas what would be causing this?

[George]

Are you using BIAS & darks frames too?

The flats seem to be doing the job OK but a lack of darks & BIAS frames could be the cause.

[russellhq]

Are you using BIAS & darks frames too?

The flats seem to be doing the job OK but a lack of darks & BIAS frames could be the cause.

Nope, I've not taken any. In the past they seem not to have made any difference. But I'll give it a try again.

As for the flats, yes they have flattened out the image but when you look at the nebulosity around the horses head, the image with the flats seems to be grainy/noisier than the image without the flats?

[George]

At the very least you will need BIAS frames, the noise you are seeing will be minimised after BIAS subtracting your flats.

The value of darks with a low noise CCD/DSLR is open to debate, I dont use them myself with a CCD but its worth doing a with/without comparison on a DSLR.

[russellhq]

I re-did the flats tonight and took the bias frames, 100 of each.

I think this has improved the image:

I just need to work on my processing skills...

[Daniel-K]

i dont think theres any need for 100 flats really is there? i have been using 20, and have no issues, worth a try

[russellhq]

i dont think theres any need for 100 flats really is there? i have been using 20, and have no issues, worth a try

I was reading the help file in DSS and it was recommending 20 as a minimum but the more the better:

This is the reason why you should use as many dark/bias/flat frames as possible. On the practical side, 20 frames is a minimum if you want to not add too much noise, and 50 to 100 will give you really nice and (almost) noise free masters..

So I figured if I went for 100 then that would eliminate any doubt, it didn't take that long to capture or process either. And with the cloudy skies tonight, I thought I'd give it a go

[dph1nm]

Its not the number of flats that matter but the total number of photons you collect. Your example flat looks underexposed to me.

NigelM

[russellhq]

In APT there are 4 histograms, RGB and I assume

Luminosity. The luminosity peak was a third to 2 fifths across but R was about 1 tenth and B about 4 fifths.