294mm calibration frames? - Forums

Topic starter

Aug 12, 2023

Y’all (yes I know, I’m canadian but I like y’all) are such a helpful bunch! There’s a lot less “read the sticky”, or “did you do a google search”, or “buy this book, I did”.

anyway…

calibration frames for the 294mm pro.

I have built a library of darks with a set temp, exposures, and gains. I get that part.

For flats, I’m gathering that the 294 isn’t linear below 1s exposure time so I’m doing 8-10 seconds each.

Now the questions. What is the purpose of calibrating your flats with the 294? Is amp noise really going to be a problem on 10 second exposures? Is there another reason? Is the “up to 1 second non-linear” really a hinderance in the real world? Should I be taking flat darks/dark flats or is this a holdover from older technologies? Am I out in left field? Lol.

As always, thx in advance!

Respectful Engaging Supportive

Astro_Vino

Aug 12, 2023

I also have the 294MM Pro and would say the following:

- the darks library is more about your amp glow, exposing at 10s will be very minimal (if at all) but as you increase the exposure time it will become more evident, at 300s the amp glow is very present (a set of 300s, 180s, 60s dark frames at -10deg for example)

The light frames are usually much faster as they are aiming for a median ADU but this will change if using filters, LRGB will be much faster Vs. narrowband where the exposure time is longer, but the light frames are intended to remove the variance of dust motes and/or vignette from the OTA (single frame example of a Luminosity filter at 0.05s exposure) .... so once stacked you should not see these artifacts in the image

Astro_Vino

Aug 12, 2023

(The example light frame)

Sean Mc

Topic starter

Aug 12, 2023

Yes, i get that. The question is about dark flats. Do i NEED to calibrate my flats? I’m currently only using flats and darks and pixinsight gives me warnings. I’m wondering if they’re unfounded or needed.

Brian Meyerberg

Aug 12, 2023

I have always used Darks, Flats and Dark Flats with all my cameras and separate Flats for each of the LRGBHSO filters. No amp glow or dust bunnies visible on any subs. I use APT for stacking and the flats require dark flats or bias frame for proper calibration. APT likes it that way.
Brian

Arun H

Aug 13, 2023 · Edited: Aug 13, 2023

Now the questions. What is the purpose of calibrating your flats with the 294? Is amp noise really going to be a problem on 10 second exposures?

Even if you accept that amp glow at 10s is not a problem, you need to get rid of the offset to properly calibrate your lights. The offset is quite significant - at gain 120, mine is around ~1600 ADU, which is 10% or so of a 16,000 ADU target of your flat. I suppose you can try to calibrate that out with the inaccurate bias under the assumption that the inaccuracy is small. I don't know what would happen if you did. Given all the questions, for me it is worth it to calibrate with a flat dark and not bother with a bias. Darks are very easy to take.

Helpful Insightful Concise

Sean Mc

Topic starter

Aug 13, 2023 · Edited: Aug 13, 2023

Thank you for your time. I wonder if you can explain what the offset is and how it relates to calibration. Sorry if that’s a question with a long answer. I think I’m misunderstanding.

what I’m probably misinterpreting is you’re saying that the camera is adding up to 10% extraneous signal in a 10 second exposure? That seems like an awful lot.

basically what I’m getting at is if i dither every frame, that gets rid of hot pixels right? The only thing left is amp glow, which is removed by darks, and I can’t see the amp glow affecting 10s flats. The camera doesn’t have pattern noise, so random noise can’t be calibrated out by flats or darks…?

Arun H

Aug 13, 2023

what I’m probably misinterpreting is you’re saying that the camera is adding up to 10% extraneous signal in a 10 second exposure? That seems like an awful lot.

The offset is not extraneous signal. All cameras I know add a fixed offset when the signal is converted from analog to digital. This is done so that signals close to zero are not clipped - due to random noise, these signals can sometimes have a negative value during the conversion and without a bias, they'd be clipped to zero. When you calibrate with a flat, you are dividing your calibrated light by the calibrated flat:

(Light - Dark)/(Flat - Flat Dark)

or

(Light - Dark)/(Flat - Bias)

These equations both assume that the calibrated flat signal captures the non uniformities as a direct multiplicative relationship with the incident light signal - which does not hold true unless the offset is removed.

That is why it is important to subtract either the bias or flat dark from your flat.

Helpful

Sean Mc

Topic starter

Aug 13, 2023

Ahhhh I get it now! The “bias” or “offset” is contained in all images, flat, dark or light. The darks remove the “bias” from the lights, but if you don’t calibrate the flats with dark flats, the bias isn’t removed from them, so the bias will be removed a second time from the lights… I think… yes?

thanks again!

Arun H

Aug 13, 2023

Ahhhh I get it now! The “bias” or “offset” is contained in all images, flat, dark or light. The darks remove the “bias” from the lights, but if you don’t calibrate the flats with dark flats, the bias isn’t removed from them, so the bias will be removed a second time from the lights… I think… yes?

thanks again!

Mostly correct. Except because you are dividing by the flat, it isn't that the bias will be removed a second time from the light (which requires subtraction) but that you will be dividing my an incorrect number.

Sean Mc

Topic starter

Aug 13, 2023

Ok I’m in the process of re-doing my flats. I had previously let the asiair calculate exposure tines so my exposure times are all over the place, which will make flat-darks a pain. I’m assuming that I can simply tune my flat panel brightness to get the same exposure time for each filter (With approximately the same median adu) ?

That way i can use one set of flat darks for all the flats right?

Chaitanya Khoje

Aug 20, 2023 · Edited: Aug 20, 2023

TLDR; So I usually have 30 Darks, 90 Flats (30 for each filter), 90 Dark Flats (30 for each filter), and then lights.

I have excellent results with this kind of set of calibration frames for my 294mm;
Darks - 30 frames (any filter does the job, you don't need all filters) | gain 200 (same as lights) | 300s
Flats - 30 frames / filter
- Use NINA software's Flat Wizard to do this, also get a light panel (A3 size should be fine). Get some white papers (I use about 7-10) and put them on the scope and then the light panel. This will help diffuse the light. Note that I have tried using iPad and it didn't give me good results.
Dark Flats 30 frames / filter - Dark flats are basically Darks but the exposure matches your flats. You load them as Darks.

You can also use the asiair if you have it for generating flats and dark flats

I use the Custom button in WBPP to do all the loading for the frames, including lights.
Load your Dark Flats for each filter as Darks, WBPP in Pixinsight knows that it has to calibrate Flats with them.

In case there's something I am missing or doing wrong, please correct, I am in the learning phase as well.. just entered the world of monochrome & narrowband imaging a few months ago.

Helpful