Clarification of Flats

[JonCarleton]

I recognize the need to darks and perhaps bias shots each time you photograph, as they are noise-dependent and subject to ambient conditions.  My understanding of flats (or lack thereof) leads me to believe they are not so much about temperatures or transient issues.  This leaves me with a few questions.

First, to the main procedural question:  Must flats be taken at each and every photo shoot or are they something you do once, or perhaps periodically?  

Now the operational questions: I'm thinking about using a tablet I have (LCD) and going to "about:blank" in a browser to create a white screen, then set the camera (SVBony305) right on top of the LCD screen (pointing at, and in contact with the screen).  Is this a reasonable procedure and do I need to space it off the screen some (and if so, how much)?  Or is that a really bad idea for reasons that would become obvious after the fact?

[Adam1234]

Hi, I have recently started taking flats myself, so I'm no expert, but it seems best practice to take flats for each imaging session at the end of the night as you can then virtually guarantee that the orientation of the camera and location of any dust on the sensor is the same between the lights and flats. They don't actually take long to take, couple minutes at most.

Tablet should work, again I'm no expert, but my images have come out ok by using a blank PowerPoint presentation in full screen and putting my laptop screen on top of my telescope. Might not be the best way, but the difference it has made to my images is incredible.

If you're using your camera when teleacope, make sure to keep the camera attached and at the same focus as the lights.

Hope that helps.

Adam

Edited April 8, 2020 by Adam1234

[JonCarleton]

Ah, I was hoping to do the lights "sans telescope."  I have a 10" Dob and that would take one big tablet to cover.  I would have to use a 22" or larger laptop monitor to cover the aperture. I suppose that eliminating the scope from the mix would negate the value of removing defects that originated in the scope itself.  Rats!

Thank's though.  I ask stupid questions to become less so.

[Adam1234]

3 minutes ago, JonCarleton said:

Ah, I was hoping to do the lights "sans telescope."  I have a 10" Dob and that would take one big tablet to cover.  I would have to use a 22" or larger laptop monitor to cover the aperture. I suppose that eliminating the scope from the mix would negate the value of removing defects that originated in the scope itself.  Rats!

Thank's though.  I ask stupid questions to become less so.

Ah yes, that could be problematic with a 10inch dob and a tablet! I've  heard alot of people use the dawn sky

[JonCarleton]

A sky shot is certainly an option.  I wonder if I need to wash out the blue or shoot in monochrome mode or will a blue flat work?

[Pompey Monkey]

A (plain) t-shirt over the aperture of the scope at dusk or dawn works fine.

[alacant]

19 hours ago, JonCarleton said:

bias shots each time

Hi

Bias and dark frames are not dependent on optics. Take say 50 bias frames -minimum exposure in the dark- stack them and use the resultant master bias. For ever...

Flat frames: necessary only if you rotate the camera or -in my experience seldom- find that there are new dust particles on the sensor.

19 hours ago, JonCarleton said:

thinking about using a tablet

You'll need an A3 light pad to cover a 10". Just go for the t-shirt for now? Any colour is fine so long as the colour is even.

Unless you use a program with a decent optimisation algorithm, I think dark frames with your camera will introduce more noise. Try it with and without?

HTH

Edited April 9, 2020 by alacant

[JonCarleton]

OK, got it (I think).  So, no using the 1960's tie-dyed T-shirt or the white one with the big coffee stain on the front.  I actually do own a plain, white T-shirt.  If I can't find it, I can always put on a HAZMAT suit and pick one up at Dollar General.

Thanks all!

[ollypenrice]

In a permanent setup in which you leave the camera in place there is no need to do flats every time. I find mine usually continue to work well for several months.

Darks and bias do evolve over time but very slowly from my experience.

I'm not sure I agree with Alacant on the use of a coloured T shirt with an OSC camera. If its filtering effect is blocking a certain colour the OSC pixels filtered to pass that colour will get little light and risk being noisy as a consequence. In a mono camera, even with colour filters in place, you could simply increase the flat exposure time for the light-starved colour.

Olly.

Edited April 9, 2020 by ollypenrice
Typo

[JonCarleton]

Thank's Olly. 

My setup is not permanent and my SVbony305 is like an eyepiece.  Excepting the brand stamp on the back there is no real way to tell exact alignment, it may not go in exactly the same way each time.  I suppose I could scribe alignment marks on the camera and focus tube. I typically swap it out for a variety of eyepieces as I am selecting my target and trying to find it.  Once I find a target and center it, I put the camera in and take my lights.

I would prefer to leave the camera in place and look about that way, but it has a very narrow FOV and I have neither the experience nor the GOTO accuracy to find objects without a bit more sky to view.  I have ordered a .5x reducer to help with that issue, but I don't expect great improvements from that sort of adaptation.

[ollypenrice]

1 hour ago, JonCarleton said:

Thank's Olly. 

My setup is not permanent and my SVbony305 is like an eyepiece.  Excepting the brand stamp on the back there is no real way to tell exact alignment, it may not go in exactly the same way each time.  I suppose I could scribe alignment marks on the camera and focus tube. I typically swap it out for a variety of eyepieces as I am selecting my target and trying to find it.  Once I find a target and center it, I put the camera in and take my lights.

I would prefer to leave the camera in place and look about that way, but it has a very narrow FOV and I have neither the experience nor the GOTO accuracy to find objects without a bit more sky to view.  I have ordered a .5x reducer to help with that issue, but I don't expect great improvements from that sort of adaptation.

A small sensor will not vignette as much as a large so that's a plus.

You can mark the camera for orientation but the best trick is to fit it a closely as you can to orthogonal (aligned along RA and Dec) and then take a roughly 5 second sub while slewing slowly in just one axis. This will produce star trails which show your present camera angle. Adjust and repeat till your trails are horizontal or vertical. It doesn't take long and is entirely repeatable, which is a bonus.

Olly

[JonCarleton]

Thanks again, Olly.

My GOTO mount is a Dobsonian ALT-AZ.  But I believe I get the idea.  Results may be a bit different, but I should still be able to use star trails for a camera alignment.

[rickwayne]

Bias and darks tend to be "Meh, good enough is good enough". The thing about flats is that they capture position-dependent information* and so if you're not pretty scrupulous about how you shoot them, the stretching process can amplify small errors into visible discontinuities.

"Can", I say. YMMV, as Alacant's apparently does. 

Vignetting and the position of dust bunny shadows anywhere other than right on the sensor are going to vary with focus position as well as camera position. I shoot a set of flats for every filter with my mono camera, because experience shows that the filters, as clean as I try to keep them, are just not identical.

One issue to keep in mind with twilight flats is that your 10" mirror is a lot better at seeing stars than you are! So you might wind up with little dark holes in your final images where the computer subtracted a star's image in the flat from your lights. Hence the advice to use a diffuser even then.

----

*Of course, so do bias and dark frames, but amp glow, hot pixels, differential response, and the like are tied directly to the sensor and so show up consistently in the same spot .

[JonCarleton]

Thanks, Rick.  I did consider the sky star thing.  I frequently do alignments as soon as  I can make out Sirius, and when it skews to the next far less brighter star, I can't see it at all in the sky, but I can use it to align as I see it clearly in the scope.

The heads-up for the flats makes sense.  I'm going to have to find a way to key my camera to the focuser.  I can see that.