With Flat vs Without Flat

[Earl]

For comparison

I admit to been lazy for ages with flats.... not any more.

[Physopto]

Yup, so am I. Still trying to get hardware working first. Must knuckle down

[swag72]

I was lazy for some time too - Thought that PI would sort it! Nothing sorts it out like proper flats

[Earl]

Calibrated with Flats vs Non calibrated RGB combine.

These have only been combined and a screen stretch applied.

[ollypenrice]

Many long years ago Paul Jenkins (a member here when I was a beginner) told me that flats were essential. I should have listened sooner rather than later. Flats are essential. And note that all my imaging scopes have flat fields pretty much as big as it gets. Even working with an 88mm circle on the FSQ106 I need flats.

The good news is that I don't need lots of them. I use a lum flat for all filters. Heresy? If it works for you, do it it. If it doesn't, don't.

Olly

By the way, on my courses I use this demo:

Flat.

Not nice. Flat divided by flat:

Very nice indeed! How much room for debate does that leave?

Olly

[Physopto]

Many long years ago Paul Jenkins (a member here when I was a beginner) told me that flats were essential. I should have listened sooner rather than later. Flats are essential. And note that all my imaging scopes have flat fields pretty much as big as it gets. Even working with an 88mm circle on the FSQ106 I need flats.

The good news is that I don't need lots of them. I use a lum flat for all filters. Heresy? If it works for you, do it it. If it doesn't, don't.

Olly

By the way, on my courses I use this demo:

Flat.

Not nice. Flat divided by flat:

Very nice indeed! How much room for debate does that leave?

Olly

Hi Olly,

I easily understand the need for flats, but am intrigued by the idea of using a luminance for all flat fields? To some extent makes sense if all filters are the same diameter, but dosen't  the filter wavelength change the pass band characteristics as the angle of incidence of the light alters, or is the change so small that it is unnoticeable on the whole?

Derek

[ollypenrice]

Hi Olly,

I easily understand the need for flats, but am intrigued by the idea of using a luminance for all flat fields? To some extent makes sense if all filters are the same diameter, but dosen't  the filter wavelength change the pass band characteristics as the angle of incidence of the light alters, or is the change so small that it is unnoticeable on the whole?

Derek

In theory, yes, you should use a different flat for each filter and a different flat each time you touch the focus during a run.

Enter reality: rather than stop your data gathering to perfect your flats at each focus tweak you would be far better off getting more data.

What about the difference between a red flat and a green flat etc? This will depend on the parfocality of your filters and the level of colour correction in your optics etc. Unless your filters are exceptionally close to your chip (eg QSI integral wheel) then it is unlikely that bunnies will derive from dust on your filters, though they might. The thing is simply to check and see if you need to go through the palaver of separate flats per filter. At the moment I find (to my delight!) that I don't have to.

Flats are fickle things. The theory is nice. Image an evenly illuminated surface after eliminating extraneous light sources and you have a perfect flat. What a nice idea! But reality is such a persistent pain in the the rear end - or in mine, at least. There is something I would call 'a good flat' and that is simply a flat that works really nicely. When I have one of these I milk it to death and might use it for six months or more. (My kit is observatory based and is never disturbed.)

I've just had to shoot a new set for one of the rigs. The R, G and B flats are not identical. The bunnies are identical but the vignetting isn't. Now I simply don't believe this. I don't believe that this is a genuine result. I might be wrong but I reckon that one of the three will be 'a good flat', a tad better than the others. I haven't played around with this yet but I'll have a session comparing them. Ever the pragmatist!

Olly

[Physopto]

In theory, yes, you should use a different flat for each filter and a different flat each time you touch the focus during a run.

Enter reality: rather than stop your data gathering to perfect your flats at each focus tweak you would be far better off getting more data.

What about the difference between a red flat and a green flat etc? This will depend on the parfocality of your filters and the level of colour correction in your optics etc. Unless your filters are exceptionally close to your chip (eg QSI integral wheel) then it is unlikely that bunnies will derive from dust on your filters, though they might. The thing is simply to check and see if you need to go through the palaver of separate flats per filter. At the moment I find (to my delight!) that I don't have to.

Flats are fickle things. The theory is nice. Image an evenly illuminated surface after eliminating extraneous light sources and you have a perfect flat. What a nice idea! But reality is such a persistent pain in the the rear end - or in mine, at least. There is something I would call 'a good flat' and that is simply a flat that works really nicely. When I have one of these I milk it to death and might use it for six months or more. (My kit is observatory based and is never disturbed.)

I've just had to shoot a new set for one of the rigs. The R, G and B flats are not identical. The bunnies are identical but the vignetting isn't. Now I simply don't believe this. I don't believe that this is a genuine result. I might be wrong but I reckon that one of the three will be 'a good flat', a tad better than the others. I haven't played around with this yet but I'll have a session comparing them. Ever the pragmatist!

Olly

Thanks Olly,

Wasn't sure that I had read you meaning correctly at first. But makes perfect sense. I do have the 683 and 583 ccd's. Love them both!  Have been doing flats when time permitted. But as new bits get added,find it makes sense to get them working before running, if you get the meaning. I'll give it a try at the first opportunity. I'll do the whole set as usual but try the reduced set as well to see what happens. I've only ever posted one image as such, as never been happy with any others.  I like to take my time!

Regards,

 Derek

[laser_jock99]

The 'halfway house' is to use library flats. These are not ideal but better than nothing. They don't account for the vagaries of dust in the optical train- but if you blow the dust of your CCD & filters and keep every generally clean it's better. You must always keep the camera in the same orientation to the scope to use library flats.

Example stretch of library master flat for an ED80 + field flattener.

By comparison a flat from a 12" F4 Newtonian + coma corrector looks a lot smother

I think refractors are more prone to the effects of dust in the optical train than reflectors. Counter intuitively all my reflectors give a more even flat field than any of my refractors or camera lenses?

However they all illustrate the need to use flats wherever possible.