How can I improve on this flat?

19 replies661 views
Jerry Gerber avatar
This is taken with the ASI2600MC and the Esprit 100 ED.  Histogram looks good, just left of center, 5 second exposure, image train, bias and temperature same as lights.

I'm attempting to make my first flats.  I see problems, I think with this master flat, not sure what's causing it or how to correct it.   Any comments?


Thank you for any help!
Jerry
Michael Finan avatar
What's your light source? Seems to be a color gradient.
Jerry Gerber avatar
Michael Finan:
What's your light source? Seems to be a color gradient.

Light source is a 10" round LED light.  But it's most likely too far away, which I am guessing is why the vignetting is only on the right side.  I am now doing it over with an additional white cloth over not only the scope but also the light source.  That way I can bring it a lot closer and still aim for around a 5" exposure, assuming that gets me a histogram that looks acceptable.

The light color is very even, not sure why that's happening either... Trial and error...
Jerry Gerber avatar
Here's my second attempt at flats.  I brought the light source within a few inches of the scope's dew shield, the light source is about 5 inches greater in diameter than the primary lens in the telescope.

But it still seems off to me.  Why is there a little more vignetting on the right side than the left?   Are there supposed to be colors?  The camera is an OSC, the ASI2600MC.  The image train also contains the Optolong L-enhance narrow-band filter, which may account for the colors...

Any critiques or comments are welcome..

Thanks,
Jerry
Helpful Concise Engaging
Joseph Biscoe IV avatar
@Jerry Gerber are your light frames showing the same vignette as your flats?

What I’ve found so far with respect to the coloring…I image with a Canon 6d(Ha) and the L-eNhance. The colors you’re getting are the result of the bandpasses on the filter. The blue pixels are registering the wavelength for Hb. That wavelength does not peak like Oiii does for green and Ha does for red. The Ha(red) and Oiii(green) have a higher peak in the bandpass. So you won’t have as much “blue” data to color balance the image. This is what makes processes like PCC and the new SPCC so important for filtered OSC cameras. But this won’t matter for flat calibration as flats are generally applied to a light frame without debayering either the flat or the light frame in the calibration process. Debayering is generally done after calibration and often time after cosmetic correction. The flat frame is simply attempting to achieve a balanced field illumination. So the software will subtract the bias and dark frames, and then divide the flat frames. Thus, removing vignetting and dust etc.

can you upload a light frame?
Helpful Insightful
Joseph Biscoe IV avatar
“Are there supposed to be colors? “

if you’re viewing the flat frame on your computer, it will show you the colorized version of the frame. Software like PI and DSS will apply the frame without the color information as before mentioned about debayering.
Marc-Antonio Fischer avatar
That is normal smile i was confusued at the beginning too with my color cam. Very important: use bias frames for calibration. I was always skeptical, but it was the reason why my flats overcorreced all the time. I did bias and they worked. I have the exact same color gradient with your camera. Even the bias image got it😅 very wiered but hey it works
Supportive
Marc-Antonio Fischer avatar
You using offaxis guider? That can cause a vignetting on one corner/side. Also small missallignment or tilt can cause that too. On the other hand, the l enhance most likely is not perfectly well made, so there is less light comming through the edges. 
Long story short: it does not matter that much. ;D flats will fix it anyways, but if you stil got problems, let me know. A very bad bortle can make perfekt flats impossible. Trust me i hat my b6😅😂 because the lights are most likely to bright or way to background limited. Thats why SHO narrowband images allways work with flats and L RGB most of them not. Especialy near the horizon.
Supportive
Jerry Gerber avatar
Joseph Biscoe IV:
@Jerry Gerber are your light frames showing the same vignette as your flats?

What I’ve found so far with respect to the coloring…I image with a Canon 6d(Ha) and the L-eNhance. The colors you’re getting are the result of the bandpasses on the filter. The blue pixels are registering the wavelength for Hb. That wavelength does not peak like Oiii does for green and Ha does for red. The Ha(red) and Oiii(green) have a higher peak in the bandpass. So you won’t have as much “blue” data to color balance the image. This is what makes processes like PCC and the new SPCC so important for filtered OSC cameras. But this won’t matter for flat calibration as flats are generally applied to a light frame without debayering either the flat or the light frame in the calibration process. Debayering is generally done after calibration and often time after cosmetic correction. The flat frame is simply attempting to achieve a balanced field illumination. So the software will subtract the bias and dark frames, and then divide the flat frames. Thus, removing vignetting and dust etc.

can you upload a light frame?

Thanks Joseph for the explanation.  I don't have any light frames yet.  I am taking small steps and learning everything I want to learn in order--first night focusing and polar alignment, 2nd night autoguiding, and today, since it's cloudy and/or the seeing is very poor, I spent the day learning how to take bias, darks, flats and dark flats.  I'm trying to be methodical so that I don't take any short-cuts.  There are no short cuts in life or art!
Well Written
Joseph Biscoe IV avatar
@Jerry Gerber of course you’re right, no short cuts. Ok, so since you haven’t shot lights before with this equipment. You will see the same vignetting in your light frames. I’m not sure why, obviously I can’t see your imaging train, but I wonder if maybe there is some tilt at the sensor? You wont be able to be desire of this until you take light frames and not the Star shape in the corners. But I’m guessing there may be some tilt on the right side and that’s the reason for the light fall off. 

What is your total rig btw?
Jerry Gerber avatar
Joseph Biscoe IV:
@Jerry Gerber of course you’re right, no short cuts. Ok, so since you haven’t shot lights before with this equipment. You will see the same vignetting in your light frames. I’m not sure why, obviously I can’t see your imaging train, but I wonder if maybe there is some tilt at the sensor? You wont be able to be desire of this until you take light frames and not the Star shape in the corners. But I’m guessing there may be some tilt on the right side and that’s the reason for the light fall off.

Esprit 100 , ASI2600MC, AM5 mount, Pixinsight, Star Tools, ASI studio, ACD Ultimate (better than Photoshop and much less expensive)..

I think the vignetting is better in my 2nd attempt as I moved light source closer to telescope.
andrea tasselli avatar
The light panel should rest on the dew-shied, assuming it is of the rigid kind. The rest is par for course and you shan't be worried by the apparent vignetting, it is just normal. You should make your master flat with raw files, not debayered.
Joseph Biscoe IV avatar
@Jerry Gerber ah ok, nice rig to start with! Yes the second flat’s vignette looks more normal. So you probably are good to go! ! 

calibration frames=trial and error until you zero in on your specific equipment. 

Can’t wait to see the end results you get
Jerry Gerber avatar
Mark, Andrea and Joseph, 

Thank you for the comments! I've looked at each of your images and you're all very inspiring to me! 

​​​​​​I'm going to try redoing the flats later today outside with a t-shirt over the scope. I think I bought an LED light source that isn't really the best for flats. 

Jerry 
www.jerrygerber.com
Well Written Respectful Supportive
Arun H avatar
LED light - assuming they are like this one are completely fine for taking flats. I have used one for years. If the light panel cannot be dimmed enough, a good practice is to add a couple of sheets of white paper on top. This has the added benefit of acting as a diffuser and improving uniformity. You need to make sure that the panel, with diffuser if any, completely covers the field of view of the scope. The way I use is to have the scope pointed vertically up, then place the panel on top of the dew shield making sure atleast some lit part is seen outside the dew  shield on all sides. The flat used for calibration must have the flat dark subtracted from it (or the bias of you are confident enough that you can get a proper bias from your camera). And yes, calibration should be done before debayering.
Helpful Engaging
Jerry Gerber avatar
Arun H:
LED light - assuming they are like this one are completely fine for taking flats. I have used one for years. If the light panel cannot be dimmed enough, a good practice is to add a couple of sheets of white paper on top. This has the added benefit of acting as a diffuser and improving uniformity. You need to make sure that the panel, with diffuser if any, completely covers the field of view of the scope. The way I use is to have the scope pointed vertically up, then place the panel on top of the dew shield making sure atleast some lit part is seen outside the dew  shield on all sides. The flat used for calibration must have the flat dark subtracted from it (or the bias of you are confident enough that you can get a proper bias from your camera). And yes, calibration should be done before debayering.

Yep, I bought the wrong one. It's not completely flat. I'll try outside and maybe get a flat LED screen.

Thanks Arun.
Joseph Biscoe IV avatar
@Jerry Gerber sky flats are great and pretty easy to acquire. 👍🏽
andrea tasselli avatar
Joseph Biscoe IV:
@Jerry Gerber sky flats are great and pretty easy to acquire. 👍🏽

Except in winter when is always dull and rainy and crappy all around, ly. Otherwise I'd wholeheartedly agree.
Joseph Biscoe IV avatar
andrea tasselli:
Joseph Biscoe IV:
@Jerry Gerber sky flats are great and pretty easy to acquire. 👍🏽

Except in winter when is always dull and rainy and crappy all around, ly. Otherwise I'd wholeheartedly agree.

Of course you’re right haha!
Jerry Gerber avatar
andrea tasselli:
The light panel should rest on the dew-shied, assuming it is of the rigid kind. The rest is par for course and you shan't be worried by the apparent vignetting, it is just normal. You should make your master flat with raw files, not debayered.

Yep, I realized that and made new flats, I think this one looks better:

Related discussions
TOPIC #1 BIAS VS DARKFLAT (ZWO 6200 MM PRO)
HI Here is my first post about the camera ZWO 6200 MM PRO First of all , carefully and without contaminating both Masters, apply the same value to both. TOPIC#1 ( BIAS VS DARKFLAT) TESTED 09/23/2004 MASTERS DATA INFO : MASTERS LIGHT HA 35 X 300S 780M...
Discusses bias and darkflat calibration frames for ZWO cameras.
Sep 24, 2024
PixInsight WBPP and calibration frames
I wanted to keep this on the PI forums but my account still waiting for admin approval so here it goes. I use PI and WBPP. as for calibration frames I use masters created from the ASAIR plus. Is this okay to do? might not be ideal I know, but is ther...
Covers calibration frames including flats in PixInsight WBPP workflow.
Aug 21, 2024