Log in

View Full Version here: : Dark Flats - Exposure time


DIYman
14-07-2021, 11:28 PM
Hi members. I know the reasoning behind the need to take Dark Flats. However, what I am unsure about is when taking Flats where the exposure lengths differ slightly, between LRGB filters, do you need to take a set of Dark Flats (for each filter ) using the same exposure as for each respective Flat? If this is the case, then there will be a set of Flats as well as a set of Dark Flats for each filter.

rustigsmed
15-07-2021, 12:13 AM
that's it unfortunately

The_bluester
15-07-2021, 06:54 AM
Depending on what camera you are using you may be able to calibrate flats with a master bias instead. My older camera (ZWO ASI294MC Pro) you could not do that, but my current pair (A ZZWO ASI2600MC Pro and 2600MM Pro) produce useful bias frames so I calibrafe the flats with a master bias. It hugely simplifies creating good master flats.

multiweb
15-07-2021, 08:31 AM
Same temp, same duration for each filter.

AnakChan
15-07-2021, 10:23 AM
I had a brief chat with Adam Block about that and it seems it may not be so simple. i.e. it’s not a matter of which model of sensors have a stable bias and which don’t. Even within the same sensor model, one unit may have a stable bias whilst another unit from the same model may not.

According to my understanding the bias of one sensor unit should be assessed over time & if that particular sensor unit demonstrates to have a stable bias over that time period, then one could use the master bias approach.

However if one is uncertain a particular sensor unit has a stable bias or not, then the safer approach is just to use flat darks.

That’s my understanding anyway.

The_bluester
15-07-2021, 11:25 AM
To date, I found my old ASI294 not to produce useful bias frames (Which was a known issue with them) and my ASI2600MC produced good bias frames, I have been using the same master bias for six months. The ASI2600MM that I received very recently also appears to produce good bias frames, but I can't comment on anything longer term there, I have only had it a couple of weeks.

For a master bias, I generate that out of at least 100 bias frames, given they are fast to shoot there is no reason not to use lots of them to average out any random noise aspects really nicely.

gregbradley
16-07-2021, 07:48 AM
Boy you guys go at it hard. Unnecessarily in my opinion.

I use 6 subs for bias.
I use 3 for flats
I don't use flat darks
I use darks bias and flats. The master flat does not have a bias subtracted from it. I do that when calibrating.

I always test my calibration frames on a sub to make sure they correct as this is a bit of an art and they don't always work well and you have to check why.

I do one flat for all RGB usually the red filter. I keep my filters clean so not relying on flats to remove dust donuts but if yours are dirty then yeah you need a flat for each of RGB.

I get good results. So these 100 bias, 60 flats etc I see people doing I would conclude are minimal improvements to an image and people are copying each other needlessly.

Far more important is exposure length, cleanliness, enough subs, dithering.
Whether you use 6 or 100 bias, darks etc I have yet to see makes hardly any difference unless someone can show me I am wrong there.

Especially with these clean CMOS and most of the late model CCDs that were pretty clean.

Lots of short exposures for CMOS similarly is not needed with these 16 bit cameras. Just treat em like sensitive CCDs. So 300 seconds subs makes sense instead of 600 sec subs. Although if the object is not bright I use 600 sec subs too.


Greg.

The_bluester
16-07-2021, 08:29 AM
My calibration routine is

50 darks per exposure time I plan to use to generate master flats (Time consuming but I only do it once or twice a year)
100 bias as they don't take long to shoot and I re use the resulting master bias for months.
25 flats per filter per session. I shoot dawn sky flats and it happens so it is no bother to me. But with dawn flats you really need a good number so that dithering between shots allows for any visible stars in each flat to be rejected.

The flats I calibrate with the master bias to generate a master flat per filter per session that I can store with the appropriate lights.

The lights I calibrate with a bad pixel map, the master flats and either the master bias or the correct length dark. With the ASI294 bias frames were no good and it has a predictable glow so darks were essential but the ASI2600 (Both MC and MM) have essentially zero glow and good bias frames so I am using the master bias more and more now and darks less and less.

I shoot lots of bias and dark frames as it is a relatively small time investment in smooth masters that I can re use for months and months.

gregbradley
16-07-2021, 09:40 AM
My calibration routine is
50 darks per exposure time I plan to use to generate master flats (Time consuming but I only do it once or twice a year)
100 bias as they don't take long to shoot and I re use the resulting master bias for months.
25 flats per filter per session. I shoot dawn sky flats and it happens so it is no bother to me. But with dawn flats you really need a good number so that dithering between shots allows for any visible stars in each flat to be rejected.

The flats I calibrate with the master bias to generate a master flat per filter per session that I can store with the appropriate lights.

The lights I calibrate with a bad pixel map, the master flats and either the master bias or the correct length dark. With the ASI294 bias frames were no good and it has a predictable glow so darks were essential but the ASI2600 (Both MC and MM) have essentially zero glow and good bias frames so I am using the master bias more and more now and darks less and less.

I shoot lots of bias and dark frames as it is a relatively small time investment in smooth masters that I can re use for months and months.[/QUOTE]


I take dusk flats with a white even cloth over the end of the scope. You have to be fast and efficient as light fades quicker than you think.
With 7 filters to flat and some 2x2 as well as 1x1 its quite a few. But I get it done with no real trouble unless I start too late.

Or when I had an Apogee U16M it took 35 minutes to cooldown so if you weren't thinking ahead you'd miss the opportunity to take flats.

I'd like to see the difference between a calibrated sub using both approaches to see if there is any visible difference.

I am pretty sure not as my calibrated frames are pretty perfect.


Greg.

The_bluester
16-07-2021, 11:52 AM
It is a good thing in Voyager, letting it sort out the flat exposure times for itself, shot by shot, I also allow it to shoot anything from 30 seonds to 0.2 seconds. I do have to remember though to have it shoot LRGB flats before NB so that the flats requiring most light per unit of time are done with the brightest sky.

I do rotate my cam from night to night and the rotator is after the flattener so there is added value in session by session flats there.