White T-shirt or sheet only for Flats? - Forums

Topic starter

Jul 19, 2021

Hi gang,
Interesting question at a gathering the other night: A new imager, who uses a light panel for Flats in a home observatory was asking if it was possible to use a darker-colored T-shirt or sheet with a mono CCD camera to acquire her Flats. Apparently, the problem is that she's not able to dim the panel enough to get the right amount of signal. Her telescope/mount and light panel are in fixed positions. This was a stumper to me - never thought about it. I suggested using two or more layers of T-shirts to lower the intensity, but thought I'd throw the question out to the Astrobin assembled. She's not a member of Astrobin (yet!), so I'll relay any suggestions.

- - Steve Solon

Well Written Respectful Concise Engaging Supportive

dkamen

Jul 19, 2021

The way I see it any shade of gray would work just as well. As well as most colors which do not have an extreme imbalance between R G and B.

John Hayes

Frequent contributor

Jul 19, 2021

I think that the best solution is to try a t-shirt over the scope aperture first. If it's still too bright, try two t-shirt layers. That almost always works.

John

Well Written Concise

andrea tasselli

Frequent contributor

Jul 19, 2021

None.

Douwe79

Jul 19, 2021

I use the screen of my iMac which is very well dimmable, works perfectly

Steve Solon

Topic starter

Jul 19, 2021

Some additional info she gave - she already uses a white pillow case over her 9.25" SCT. This light panel is dimmable, just not enough, she feels. Her flats are on the order of .2 to .4 seconds.

Alan Brunelle

Jul 19, 2021

I cannot add to a solution, but I have a question. I am someone who uses sky flats and I certainly have no capability of dimming the sky. And no, I have no reason to go out and layer fabric either. I just pick an appropriate exposure.

So my question is: Why can't your friend just adjust the exposure time so that she is on the right part of the curve? Can she not take exposures less than .2 seconds? Or is there a reason she cannot use shorter exposures?

Well Written Concise

Claudio Tenreiro

Jul 19, 2021

Yesterday I was taking Flats and Dark Flats using the NINA wizard, and a GN flat panel, for some reason NINA (Wizard) did not allow me to play with exposure, so I used a sheet of white paper between the tube and the panel, and it worked out fine. Of course I could go to the normal way of taking exposures and then just change the exposure time at will. The other possibility is SharpCap where you can select any exposure time you wish, looking at the histogram at the same time. With SharpCap you can do experiments with T-Shirts, or anything of different colour, and see the result, or look for the best result.
Actually I just heard that for taking flats with these dual NB filters, like L-Enhance, it can help a layer of red in front of the panel, and I have to try that as well..

Alan Brunelle

Jul 19, 2021

Claudio Tenreiro:
Yesterday I was taking Flats and Dark Flats using the NINA wizard, and a GN flat panel, for some reason NINA (Wizard) did not allow me to play with exposure, so I used a sheet of white paper between the tube and the panel, and it worked out fine

I use NINA as well but have not yet tried the flats Wizard. I should look into it.

I do like the freedom of choosing my exposure though. And you can choose "flats" as a sub category, so it's at least in the meta data.

I should have said that I am an OSC photographer, so that makes it easy to do a single set of flats in the morning. My flats are very short because I need the sky to be bright to remove any star images. This is often just on or after sunrise. I am pointing at the sky maybe 15-20 degrees past zenith, opposite the position of the sun, to minimize gradients. So that means exposures of approx. 0.03.

But I understand that for narrow band drivers how it becomes more urgent, when flipping through filters during a session. I believe that doing so implies that one is dealing with specs on the optics/sensor mostly. Otherwise vignetting and stuff like that should be consistent filter to filter. I may be wrong, but it seems that if one keeps their optical train clean the other issues need only be done with simple flats from any one filter, assuming all filters are dimensionally identical. I assume these are normally bought as sets, right?

Helpful Insightful Respectful Engaging

Andy 01

IOTD Staff

Jul 20, 2021

I used to have the same problem. Solved it by using my SpaceX Falcon 9 t-shirt, which is charcoal grey in colour.

Nowadays though, I do them indoors using an old LCD TV connected to my laptop.
Just boot up the google homepage and point the scope to the white area beside the logo.

Viola, perfect flats every time, same exposure times all year round too!

Helpful

Bruce Donzanti

Jul 20, 2021

I have used both the morning sky (facing away from the sun) with a t-shirt and a light panel in my observatory with no t-shirt. Both give very consistent results whether for BB or NB filters. My light panel is adjustable but I keep it on the highest setting and simply regulate the exposure time using SharpCap to keep the histogram peak within the 40-60% range to get ideal flats. I have done this using both MM and MC cameras with no issues. So, I think your best bet is to use SharpCap if you want to use your light panel and just the exposure time like Alan suggests. I have not used NINA but SharpCap makes this such a simple process that it only takes a few minutes. Using the morning blue sky with a t-shirt is also ideal at dusk while you are worried about picking up stars. Anyway, this is my 2 cents.

Helpful

Dale Penkala

Jul 20, 2021 · Edited: Jul 20, 2021

Like many here I use SharpCap and its an outstanding software package. In my case I take 3 second flats with my 294mc pro and I use a large flat screen that is dimmable. I use 3 white t-shirts and multiple layers of white paper between the screen and t-shirts. I always try to get my histogram right about the 50% range.

I used to do the sky illumination but found this flat panel to be way more controllable and it was quite inexpensive as I got it from a friend of mine.

I have Nina myself and have never used the program for this. I’d like to try it at some point.

Just my process, is all.

Dale

Helpful Engaging

Bogdan Borz

Jul 20, 2021

Well, the solution for me is definitely more T-shirts ! I only had problems for flats with my 250mm F4 Newtonian, mainly due to some internal reflections that I haven't been able to perfectly treat so far.

The T-shirts are needed only for RGB filters in order to obtain a flat duration > 1-2s with my CMOS ASI cameras. The flat panel was too strong at its minimum position even with 2 white T-shirts so I added another one : ). Sheets were too thin. A very important aspect of successful astrophotography is having an appropriate worerobe.

Clear skies !

Bogdan

Helpful Engaging

Steven

Jul 20, 2021 · Edited: Jul 20, 2021

Alan Brunelle:
I cannot add to a solution, but I have a question. I am someone who uses sky flats and I certainly have no capability of dimming the sky. And no, I have no reason to go out and layer fabric either. I just pick an appropriate exposure.

So my question is: Why can't your friend just adjust the exposure time so that she is on the right part of the curve? Can she not take exposures less than .2 seconds? Or is there a reason she cannot use shorter exposures?

exactly my thoughts. I used to just change my exposure to get the histogram where I wanted it to be for flats.
Why "bother" with dimming and setting brightnesses, when changing the exposure is a lot easier? (as long as you don't go crazy short with the exposure times)

So far I've just used a t-shirt. usually double layered. stuck on with an elastic band..
Point it somewhere at an evenly lit part of the sky, it does the trick.

Now I control everything through a ZWO ASIAIR pro. Which provides assistance with with the flats, and calculates its own exposure. exactly like Alan said. It adjusts the exposure to get the right kind of flat.

So far at least, I haven't had any issues with my flats. My images come out pretty well, without vignetting or other issues.
So, I guess it's fine?

But....

I do understand that the flats might be better with a more "controlled" method and a controlled environment.
Maybe I'll use my iPad or something to create a more "controlled" flat-library in the future.

Alan Brunelle

Jul 20, 2021 · Edited: Jul 20, 2021

(as long as you don't go crazy short with the exposure times)

My exposures are quite short. Others here may know when short is too short. However I believe to short only applies to those older style CCD cameras that have an actual physical shutter. With those, a too short exposure would actually capture the shutter sweep and yield a very imbalanced sub. But most now have electronic shutters, can do millisecond exposures etc...

I actually think what many people do now for flats came about because of what was required from the older cameras.

In doing research on the topic, I have learned that professional observatories will often do straight up sky flats. Pretty tough to find a 300 inch T-shirt, right?! Or otherwise they have a painted white area on the inside of the dome which they shine a light on, point the scope there and take their flats.

I think the flats panels are really only needed for the many who do multiple filter changes during single sessions. Each change will want it's own flat, especially if the filter change interjects it's own artifacts. But, as I said above, if you keep your optics clean (especially your camera's window and sensor), the only reason for flats should be vignetting of some sort. If your filters are matched, and your filter wheel positions accurately, then that vignetting should be identical for all your filters. Because of this, only one set of flats should work for all. Please correct me if I am wrong.

I know some who first make sure their optics are clean, second, do not fiddle all the time with their system (thereby ensuring it stays clean), and use one set of flats for weeks or more. Ask yourself what it is you are solving with your flats. We are not lemmings. If you don't have dust spots and you have no vignetting or some other artifact, you don't even need flats.

Helpful

Claudio Tenreiro

Jul 20, 2021

It is difficult to disagree with most of what has been said, particularly because a lot is based on real experience, so it reflects what works better for everyone, however there is one point which is a bit complicated (at least for me), it is when using a ZWO294 mc Pro together with L-Extreme or L-Enhance dual NB filters. Actually you can see that Gerd Neumann sells an extra film to take flats for these filters.
Using SharpCap or any other program you notice that is impossible to get a decent flat with the normal rule (40% to 60% histogram) because the red channel is well out to the left, to less than 10% to 12%, and you get a complicated pattern. Maybe I am doing something wrong in this case but in Cloudy Nights it is plenty of discussions about this very same topic.

Ruediger

Jul 20, 2021

Alan Brunelle:
My exposures are quite short. Others here may know when short is too short. However I believe to short only applies to those older style CCD cameras that have an actual physical shutter. With those, a too short exposure would actually capture the shutter sweep and yield a very imbalanced sub. But most now have electronic shutters, can do millisecond exposures etc...

Hi all,

Please allow, but I have a have a different experience. Such short exposures did not work for me. Especially with CMOS, since at these very fast exposures they operate in a range, where the chip is not working perfectly linear anymore. These nonlinear variations are not visible in single high signal sub like flats, but when used for calibration they clearly become noticeable.

Also the read noise becomes more significant. Your matching flat darks are more or less bias frames and as you know, using bias for CMOS leads to unpredictable results and should be omitted. Therefore the best practice advice not to use bias, but flat darks, where the (still linear) bias noise is contained, but short enough that thermal noise is not a topic, became the golden rule for CMOS calibration.

Of course, there might be cameras, which will work linear down to very fast speeds, but quite sure not in consumer range we usually use.

I agree, if you have a mechanical shutter (even some expensive CMOS do have) then short exposures are not possible as mentioned.

Personally I have experienced serious issues with too fast flats which resulted in unpredictable calibration results, since bias noise gets volatile. Therefore my personal conclusion is too stay between 2 and 5 seconds, take precisely matching dark flats, and for me many calibration problems vanished.

CS
Rüdiger

Helpful Insightful Respectful

Alan Brunelle

Jul 20, 2021

Ruediger:
Also the read noise becomes more significant. Your matching flat darks are more or less bias frames and as you know, using bias for CMOS leads to unpredictable results and should be omitted. Therefore the best practice advice not to use bias, but flat darks, where the (still linear) bias noise is contained, but short enough that thermal noise is not a topic, became the golden rule for CMOS calibration.

Of course, there might be cameras, which will work linear down to very fast speeds, but quite sure not in consumer range we usually use.

I agree, if you have a mechanical shutter (even some expensive CMOS do have) then short exposures are not possible as mentioned.

Personally I have experienced serious issues with too fast flats which resulted in unpredictable calibration results, since bias noise gets volatile. Therefore my personal conclusion is too stay between 2 and 5 seconds, take precisely matching dark flats, and for me many calibration problems vanished.

I can't speak to all cameras and all circumstances. That is why I gave the mechanical shutter example. My two cameras are the ZWO 071mc pro and the QHY268c. I'm curious as to what cameras gave you such issues?

I do think that your point about camera-specific results is very important. But as I said, for me, flat exposures of even 0.03 s work just fine.

I do not see my short flats resembling or acting like what you see. And I have done this with two consumer grade CMOS cameras. What is so very different is that the sensor is receiving a lot of photons in my flats, generally many more than a typical light. Probably like most peoples' using screens. So shot noise should be especially irrelevant than for my lights. Same for read noise.

I don't see your point regarding read noise at all. It is the same for all subs, whether light, dark or flat. I did not want to suggest that my flats are very low adu. Again, they are much greater than that of my lights and I always shoot for about 30% on the histogram. So even if the camera generated such weird fluctuations because of extremely short exposures it would have to be extremely high level to be reflected in the flat correction algorithms. Maybe your camera is reacting badly for overly short exposures. But I'm not sure 0.03 sec counts as extreme for my cameras. They image just fine at 0.00X sec...

I have successfully gotten good results doing my stacks in both DSS and PI. So my flats do not seem to be an issue. Clearly each algorithm is capable of correctly scaling my master flat.

If for some reason someone uses extremely low adu for their flats, then I can see that leading to issues caused by noise and USB timing artifacts. But I am not sure why one would want to do that. I have not felt the need to do flat darks since sky flats take me less than a minute to do (40 subs). The only issues I have is when I wake up and the clouds rolled in! Trust me cloud flats don't work! But I keep my system clean, so I only need to correct for vignette type issues and can often use flats from another day. Which saves me even more time!

Dale Penkala

Jul 20, 2021

Ruediger:
Personally I have experienced serious issues with too fast flats which resulted in unpredictable calibration results, since bias noise gets volatile. Therefore my personal conclusion is too stay between 2 and 5 seconds, take precisely matching dark flats, and for me many calibration problems vanished.

CS
Rüdiger

I totally agree with you Rudiger. The ZWO ASI294MC Pro has a history of “more then normal” amp glow and I’ve found that 3 second exposures for my flats calibrate out really well. I used to do the shorter exposures and could never get rid of the amp glow along with other artifacts in the image. Once I did that I ended up with a much better working image so now ALL flats with my 294mc pro are shot with 3 sec exposure times.

“Using SharpCap or any other program you notice that is impossible to get a decent flat with the normal rule (40% to 60% histogram) because the red channel is well out to the left, to less than 10% to 12%, and you get a complicated pattern. Maybe I am doing something wrong in this case but in Cloudy Nights it is plenty of discussions about this very same topic.”

I know what your talking about here and yes my histogram was out of the 40%-60% range however I still used them with very good results with my 071mc pro camera. I wasn’t aware there was a film that could be used to counter act that red shift in the histogram.

Dale

Helpful

Richard Witzke

Jul 20, 2021

I own a ZWO 183C - Pro and I also find that you must take flats between 2 to 5 seconds (which I learned from the Lazy Geek). My LCD flat panel light is made for illumination in a photography studio, which doesn't dim enough, so am also forced to use a couple layers of white tee shirt and sometimes add a sheet of paper to get those long exposures. Short exposure flats just don't work as well.

Well Written Helpful Concise

Alan Brunelle

Jul 20, 2021

Dale Penkala:
I know what your talking about here and yes my histogram was out of the 40%-60% range however I still used them with very good results with my 071mc pro camera. I wasn’t aware there was a film that could be used to counter act that red shift in the histogram.

I have to consider myself lucky, having picked cameras that are relatively easy! I can assure you that this issue was not anything I was planning for when deciding which camera to buy, even though the adverts certainly pushed the fact that the cameras had zero amp glow. Dale, I am curious as to what amp glow does within your darks and when is the correction made in processing.

The color imbalance issue is also interesting. I can say that sky flats also have a color imbalance. My sky is most certainly not white. As you might expect. It must not be too great since the algorithms seem to handle them pretty well. But in the interest of full disclosure...

Alan

Engaging

Claudio Tenreiro

Jul 20, 2021

Dale Penkala:
I wasn’t aware there was a film that could be used to counter act that red shift in the histogram

Here is the link for those: https://www.gerdneumann.net/english/astrofotografie-parts-astrophotography/aurora-flatfield-panels/aurora-accessories-aurora-zubehoer/lpr-konversionsfilter-fur-100mm-und-160mm-aurora.html
Yes, I saw also from the Lazy Geek the trick of longer flats, I am around 3 sec for the ZWO294...but is a complicated camera. However the QHY183C also shows some effects with this dual NB filters.
But, as you can follow in the thread, people try to get a solution but is hard to establish them as a general rule.

Ruediger

Jul 20, 2021

Alan Brunelle:
I can't speak to all cameras and all circumstances. That is why I gave the mechanical shutter example. My two cameras are the ZWO 071mc pro and the QHY268c. I'm curious as to what cameras gave you such issues?

Hi Alan,

I use a an ASI6200.

The problem with (too) short flats was discussed over at CN and PI forum (multiple times). The problem with linearity means, that not twice the amount of photons will lead to double ADU when operating in non linear range. Especially when doing flats this is crucial since we want to measure light differences. The non linearity produces additional gradients, which are not visible in the flat, but in the corrected stack.

That is the reason why to stay in the 40% to 60% full well thumb rule and decent exposure times.

Of course this is sensor specific. Therefore I advice everyone to test with his own setup. But I would absolutely refrain from giving a general statement that such short exposures will work. Actually this is rather the exception then the norm. Especially you are getting close to bias frames instead of dark flats, which is definitely fatal.

Please, don’t get me wrong: If it is working for you, it is absolutely fine, but I think it is a very dangerous general advice.

CS
Rüdiger

Helpful Insightful Respectful Engaging

Bob Lockwood

Jul 20, 2021 · Edited: Jul 20, 2021

I use this to dim up laptops and to adjust my flat box brightenss. Works great, you just add as many layers as you need.

They have this in many different stops, so you can really fine tune what is needed.

https://www.bhphotovideo.com/c/product/166163-REG/Gam_GC1517_GC1517_Cine_Neutral_Density.html?sts=pi&pim=Y

Alan Brunelle

Jul 20, 2021

Ruediger:
I use a an ASI6200.

Thanks! In fact the 6200 is from the similarly manufactured 2600 sensor, which is in the same sensor in my QHY. So I will be very wary of this.

I didn't expect that my words were necessarily anything other than offering what I do (and presumably what I get away with!). I know that my experience is not uncommon, nor yours and the others who posted here. Given the original question in the post, I thought the idea of reducing exposure time as a simple first step to try, not to say the other ideas here are not valid or even better. Shorter exposure times, along with some of the other ideas posted here can all be studied in a single session. Given they all work, I would choose the simplest and the one that supports the equipment they already have.

I suffer from a rather fast largish aperature system, and the T shirt method (sky illuminated) actually failed me. Got all sorts of gradients. So I just pointed skyward and had success. Come to find many others do so as well. Given that, I have no choice but short exposures. And I'm not complaining!

Alan