"L" flats problem..

Hello All,

My L flats at 3910 mm never seem to apply properly in WBPP; or at least not as well as the RGB flats. SHO always seem to work well. All else being the same.

Values are consistent, calibration method in WBPP is consistent (with the other filters and in general when ever I take flats for my other rigs etc. In other words, I'm pretty sure that I'm not making any glaring errors). I use bias and darks all at the same gain, offset and temperature (and sub time in the case of darks).

I have tried several config differences in WBPP as well with no obvious differences.

Furthermore, the other night, in order to eliminate possible filter wheel positioning errors, and/or any chance of the flats being "stale"; I immediately took new flats right after about four hours of L data collection and they still presented the same problem. Some motes simply do not get cleared and a general blotchiness (is that a word?!) exists. I can cheat a bit with Clone Stamp but would rather address it properly. Oh, and sky or LED flats present the same issue.

I haven't disassembled yet to look but is it possible that some humidity based "slime" could have built up on the filter? But if so, why not the others?

Any ideas would be appreciated!

Do you have an example you can share of this problem?

Do you have an example you can share of this problem?

You beat me to it.

Would a master "L" file help?

What is an L flat?

Little Flat, Lacerta Flat, Large Flat, Live Flat, Luminance Flat, Lizard Flat????

Not having seen the data yet. What camera? CCD or CMOS? If using flat-darks on the flat frames with a CMOS sensor, do not use bias frames. Also my flat issues were solved by dimming the light source and increasing exposure time. My RGB flat frames typically are 3 to 5 secs and NB flat frames are 10 to 15 sec. A short version is that a CMOS sensor behaves differently at short exposures vs long exposures. Increasing my exposure on the flat frames solved all of my flat frame issues.

Looking forward to looking at your data samples and hope you find your answers to your Flat Frame issues. Some really smart people are on this site, you are in good hands.

CS

Tim

Would a master "L" file help?

Anything beats what we have to go on now. 🙂

Tim Ray:
My RGB flat frames typically are 3 to 5 secs and NB flat frames are 10 to 15 sec

I'll try to post some data but you have provided the first step that I should try - My L flats have only been about 0.25 sec whereas my SHO flats are all much longer and don't suffer from this problem. I'll dim the panel and try again !

Anything beats what we have to go on now. 🙂

*Ok I'm working on it; seems that I can only attach 10 MB files here; I'll have to do it another way or just send a screenshot showing the effect..

Anything beats what we have to go on now. 🙂

*Ok I'm working on it; seems that I can only attach 10 MB files here; I'll have to do it another way or just send a screenshot showing the effect..

Either way. A service like OneDrive or Dropbox works too. The suggestion above to increase the flats duration makes sense to try, as luminance would always have the shortest duration flats and seem to be the only portion of the data impacted.

Seeing the problem would help in case that doesn't solve it.

3.81
messierman3000#29 minutes ago· 44 views
What is an L flat?

Little Flat, Lacerta Flat, Large Flat, Live Flat, Luminance Flat, Lizard Flat????

Was really hoping it was a Lizard Flat. On second thoughts, may be not. That's probably a lizard that's been run over by a bus.

I see you are using the 6200 & 2600. FWIW, I’ve used the 294, 071 & 2600 and always have shot 3 second flats & dark/flats and they have calibrated out just fine. I am shooting OSC however.

Dale

Furthermore, the other night, in order to eliminate possible filter wheel positioning errors, and/or any chance of the flats being "stale"; I immediately took new flats right after about four hours of L data collection and they still presented the same problem. Some motes simply do not get cleared and a general blotchiness (is that a word?!) exists. I can cheat a bit with Clone Stamp but would rather address it properly. Oh, and sky or LED flats present the same issue.

I presume your flats are for the ASI2600 MM Pro. I have the color version but had flat issues too. My problem was that the default camera parameters in NINA for gain and offset were really bad. I changed the gain from 700 to 200 and the offset from 0 to 50. This is important for the short exposures for the L flat. An offset of zero produces a significant number of zero values in short exposures, making calibrations poor. Check What you've been using and see if you have a similar problem.

I never shoot flats longer than 0.5s and I'm none the worse.

The duration on the Flat frames and how the CMOS sensor behaves were pearls dropped by Adam Block on one of his YouTube videos on Flats, Bias and Image Calibration I watched a few years back… In my opinion, Adam Block has forgotten more about astrophotography than I will ever know in 3 life times. A true Maven in Astrophotography!

A must watch!

CS

Tim

andrea tasselli:
I never shoot flats longer than 0.5s and I'm none the worse.

If you do the research on the 294 & 071 you will see that 3-5 second flats are recommended. Once I did that it fixed all the problems I ever had and followed thru with my 2600.

Dale Penkala:
If you do the research on the 294 & 071 you will see that 3-5 second flats are recommended. Once I did that it fixed all the problems I ever had and followed thru with my 2600.

Recommended by whom? Not by me. I have the ASI294MC, HC533C, AA26C (IMX571) and ASI294MM and I can't find a single reason for this assumption.

One other issue is that you are imaging, it seems, from Bortle 8. Luminance images are by far the most sensitive to any issues, light leakage, etc. Anything that doesn't perfectly mimic the light coming in from the sky when you are imaging will contribute to a miscorrection.

andrea tasselli:
Dale Penkala:
If you do the research on the 294 & 071 you will see that 3-5 second flats are recommended. Once I did that it fixed all the problems I ever had and followed thru with my 2600.

Recommended by whom? Not by me. I have the ASI294MC, HC533C, AA26C (IMX571) and ASI294MM and I can't find a single reason for this assumption.

Couldn't ever get the 294C to produce images without a massive red blotchy mess in the background. This was years ago though so maybe those problems are over these days.

Like others have said, it would help to see pictures. If you need to resize them there's a resize tool in MSPaint, also exporting from Pixinsight as a jpeg and lowering the slider will make the file size smaller too.

In the meantime, one of the things that helped my flats with the 2600mm pro, is setting the target histgram to 25% with a tolerance of I think 15%. Originally I tried using Ninas flatswizard at 50%, and it would over correct the vignetting, but miss dust spots. Length of my flats with that camera hasn't mattered much. I exclusively use skyflats, so the exposure time is pretty quick.

Mine is from "years ago".

andrea tasselli:
Dale Penkala:
If you do the research on the 294 & 071 you will see that 3-5 second flats are recommended. Once I did that it fixed all the problems I ever had and followed thru with my 2600.

Recommended by whom? Not by me. I have the ASI294MC, HC533C, AA26C (IMX571) and ASI294MM and I can't find a single reason for this assumption.

No not at all Andrea, I used to shoot 1 second flats and I always had something that didn’t calibrate out. If you search “flat issues 294” there is many threads over on CN (cloudy nights) that address this problem. When I got my 071 I had similar problems but not as bad so I went and shot 3, 4 & 5 second flats and to my surprize the 3 second flats fixed all my problems.

With my issues in the past I just followed thru with 3 second flats with my 2600 and whatever I shoot my flats with I shoot dark flats at the same settings.

Dale

Dale Penkala:
If you search “flat issues 294” there is many threads over on CN (cloudy nights) that address this problem.

The issues with the 294 sensor seem to relate to how it handles timing for short exposures. Timing for very short exposures is handled by on chip circuitry which heats the sensor and messes the temperature. The reason to force exposures of 4 seconds or more is to avoid this, and also to ensure that the exposure is long enough that timing errors introduced by driver controlled timing for longer exposures does not become a factor. With chips that don't have amp glow type issues, these small errors do not contribute materially to correction, but with the 294, they seem to. I think you can afford to be much less careful with the 6200 and 2600 sensors.

Dale Penkala:
No not at all Andrea, I used to shoot 1 second flats and I always had something that didn’t calibrate out. If you search “flat issues 294” there is many threads over on CN (cloudy nights) that address this problem. When I got my 071 I had similar problems but not as bad so I went and shot 3, 4 & 5 second flats and to my surprize the 3 second flats fixed all my problems.

With my issues in the past I just followed thru with 3 second flats with my 2600 and whatever I shoot my flats with I shoot dark flats at the same settings.

Dale

I'm not disputing that those settings work for you, Dale. I'm just saying what works for me. And I never use the 1/2 of the histogram rule (if that is a rule). Maybe 1/10 to 1/5 and not beyond.

Arun H:
The issues with the 294 sensor seem to relate to how it handles timing for short exposures. Timing for very short exposures is handled by on chip circuitry which heats the sensor and messes the temperature. The reason to force exposures of 4 seconds is to avoid this, and also to ensure that the exposure is long enough that timing errors introduced by driver controlled timing for longer exposures does not become a factor.

Never had an issue with that and I don't put any stock in the effect of temperature on exposures less than a second.

Related discussions