Flats for Ha, L, R, G and B when there is focus difference.

[Jannis]

So, I probably know the answer already, i just don't like it... 

I had my first go this night with my new observatory and new LRGB filters and electronic filter wheel. I was hoping for a easy imaging session if everything worked well, and for the most part it did - except for my optics. 

I'm using an old Olympus OM 135mm F/3.5 lens, and i've previously used this for Ha and OIII without issues. However, i noticed once i put on my CLS or L filter that this lens would simply not work easily. The focus difference between the red and blue spectrum is huge. So huge that even stepping the lens down to F/8 didn't result in pinpoint stars. This naturally means that i would not be able to use the L or CLS filter, but stick to just RGB with re-focusing after each filter switch.

So then is my questing (exclude any dust bunnies, temp difference etc), taking flats is normally straight forward as you simply leave the focus and camera position untouched. I usually take my flats by using the sky the next day.
Normally this is no problem, but how critical is the focus difference between the filters? I can easily run trough every filter and take flats like i should, but naturally in daylight i can't re-focus each filter. I guess for Ha and R the focus is roughly the same, but what about B and G?

Here is a picture with the CLS to easily show the focus difference wide open and at F/8.

Also, first RGB picture from this night, so far without flats or darks though...   

[ollypenrice]

I would be astonished if the change in focus were enough to have a visible impact on the final image. Yes, the dust bunies will change size - but by how much? My guess would be 'too little to matter.' A number of the imagers who come here with their own kit will simply use a luminance flat for all their filters and I often do likewise. Sometimes it doesn't work because of a dust mote on one filter but, quite honestly, it usually does work.

You could try a test: take two master flats, one from each of the filters most widely separated in terms of focus. Make a copy of one and then apply this to the flat of which it's a copy. You'll get an absolutely flat screen. Now apply one of the masterflats to the other and see how close you get to an absolutely flat screen. Any non uniformity in the resulting screen will show you the level of miss-match between filters (apart from any dust bunny unique to the filters in question. Ignore those.) However, any bunnies common to both, and the vignetting of the lens, will show the level of mis-match due to the filter shift.

(I often demonstrate tha value of flats by applying a flat to a flat in my introductory tutorials. It is enough to persuade most people to take flats!!!)

Olly

[Jannis]

Thanks! 

I will give it a try and see how it goes.

 

[geoflewis]

15 minutes ago, ollypenrice said:

You could try a test: take two master flats, one from each of the filters most widely separated in terms of focus. Make a copy of one and then apply this to the flat of which it's a copy. You'll get an absolutely flat screen. Now apply one of the masterflats to the other and see how close you get to an absolutely flat screen. Any non uniformity in the resulting screen will show you the level of miss-match between filters (apart from any dust bunny unique to the filters in question. Ignore those.) However, any bunnies common to both, and the vignetting of the lens, will show the level of mis-match due to the filter shift.

What a great idea, I've never done this, but it looks like a simple test...

[kirkster501]

I do flats per filter.  I don't like the "use luminance for all filters" flats model.  However, I take flast at the last  focus position that night since, with autofocus, the focus changes throughout the night.  The flats do not have to be at the exact focus point of the lights.  That would be impossible, especially so with autofocus.