Astrofest collection

[multiweb (Marc)]

Originally Posted by peter_4059

Thanks for the heads up Marc - how do you tell if it is a problem with the flat or the bias? I'm taking flats to get the histogram between 30-50% and the bias at 0.001 s.

If the bias is not subtracted from the flat prior to divide it into the sub then you get an over correction. If the flat ADU is too low or too high you'll get the same problem. It's hard to tell if it's one or the other but it is very easily fixed with pixel math. I'd use the Rho shot to test. Add or subtract 100ADU to start with from your master flat and divide in your sub until you get an even background. Once you got it right you can apply the values to the rest of your subs and you'll be able to push your data even further because the border will dim and bring more faint stuff while the center will brighten up slightly.

[peter_4059 (Peter)]

Originally Posted by multiweb

If the bias is not subtracted from the flat prior to divide it into the sub then you get an over correction. If the flat ADU is too low or too high you'll get the same problem. It's hard to tell if it's one or the other but it is very easily fixed with pixel math. I'd use the Rho shot to test. Add or subtract 100ADU to start with from your master flat and divide in your sub until you get an even background. Once you got it right you can apply the values to the rest of your subs and you'll be able to push your data even further because the border will dim and bring more faint stuff while the center will brighten up slightly.

I thought DSS was subtracting the bias from the flat before dividing into the sub - that's what the DSS help calibration flow chart implys?

When you say divide in your sub until you get an even background are you saying try different adu values in the subrtaction step or perform multiple divisions?

[Alchemy (Clive)]

Must have had some good nights by the look of it.

[multiweb (Marc)]

Originally Posted by peter_4059

I thought DSS was subtracting the bias from the flat before dividing into the sub - that's what the DSS help calibration flow chart implys?

When you say divide in your sub until you get an even background are you saying try different adu values in the subrtaction step or perform multiple divisions?

You add/subtract in increment of 100ADU from your master flat then save it as another version. Then take one original uncalibrated sub and divide it in (once) with the new master flat and check how your background evens out. When you hit the sweet spot use that flat to calibrate all your other subs.

[peter_4059 (Peter)]

Thanks Marc - I'll give that a try