DSLR, Ha-filter och flats

[magnusl]

Hi!

I'm using a DSLR and have recently become owner of a Ha-filter. Now I'm curious about how to produce adequate flats for the subs I take with this filter. I guess the best idea is to extract the R channel and only work with this (as described in this thread: http://pixinsight.com/forum/index.php?topic=4702.0). However, what would be the best process here? Should I extract the R-channel from all flat subs, and then integrate them, or should I integrate first and the extract? It seems somewhat tedious to manually extract the R channel from 50 or 100 subs.....

After producing a R flat, I guess I extract in a similar way the R channel from the lights (again, a lot of subs...) and from my master dark, and then calibrate the lights. Right?

And when combining with the regular RGB-subs, I guess I would need to downsample them to the same geometry?

Or is there a better way to do it?

Magnus

[Niall Saunders]

Hi Magnus,

I would go for the non-tedious approach: Integrate and then extract. As far as the actual data is concerned, statistically-speaking, there is no difference in the order of the processes.

Remember, you will probably need quite long exposures for the HaFlats, so you will very likely need to gather a similar number of FlatDarks for this dataset, and would actually be Integrating the HaFlats after first running them through DarkFlat Calibration.

Hope this helps.

[magnusl]

Hi!

Great, thanks. FlatDarks.... what do you mean by that?

I have a regular dark library, and always calibrate the flats with these (optimize option checked). Should I do something else here?

Magnus

[Niall Saunders]

Hi Magnus,

Sorry - 'FlatDarks' is just a name that I use to describe one of the various collections of imaging data required for any imaging session, as follows:

Biases : Taken at the shortest practical exposure time (not always needed, especially if the imager can be run at a stable, repeatable - preferably temperature-controlled - level)

Darks : Taken at the same exposure time, and temperature, as the corresponding Lights

Flats : Taken using some form of reliable, uniform, light source

FlatDarks : Darks taken at the same exposure time, and temperature, as the corresponding Flats

Lights : Taken at an exposure time to suit the conditions and requirements, and at a stable, repeatable (preferably temperature-controlled) level

So, you will need darks taken to match your lights. However, if you can control the CCD temperature, then Darks can be taken ahead of time, and a library of such darks can last for a very long time (perhaps even years).

If you can control CCD temperatures, and have Darks to match your lights, then there is no need for Biases (otherwise, Biases are required to help match the Darks to the Lights)

Given that Flats are just a specialised type of Lights, you will still need darks to calibrate these Flats - I call these FlatDarks, to differentiate between the two Darks being used. It is just a name!

The only datasets that really need to be taken at every session are Lights and Flats. However, if your setup does not change between sessions (and that includes focus) then you don't eben need to re-acquire Flats. That said, even using a filterwheel can cause the optical train to require re-focusing every time a filter is changed (that is why I stopped working with mono imagers and filterwheels, in favour of OSC imagers - the time saved is significant, and when local weather conditions are often the biggest single limiting factor in image acquisition, every second saved can be critical).

Hope this helps.