240 replies to this topic

[Shiraz]

Sorry i missed that. If you do the flat darks at the same settings as the flats, the dark subtraction process should get rid of the bias and any related glows in the flats - regardless of the flat/flatdark exposure time. However, I will do a quick test tonight just to make sure this is what actually happens with very short exposures.

 

Cheers Ray

[Shiraz]

 I did a flat cal process at 0.02s (flat and flat dark) and at my normal 3 seconds. Comparing the 2 different flats shows that there is a slight difference  - not extreme, but possibly enough to mess things up on some targets by adding slight gradients. Agree that very short flats are not ideal - maybe try them out for your setup if you have a very bright cal source, but be aware that you may end up with some reduction in calibration quality. I didn't do enough tests to determine how long "long enough" is, but  Jon's suggestion of >0.2 seconds would probably be a good starting point and I definitely have good results at 3 seconds.

 

Cheers Ray

Edited by Shiraz, 29 April 2020 - 04:55 PM.

[Gymbow]

Thanks Ray for providing that info.  Much appreciated.  

[Shiraz]

Thanks Ray for providing that info.  Much appreciated.  

as it turned out, I need to thank you. The flats showed up a condensation area that I hadn't previously noticed - would've been annoyed if it had ruined an imaging run, so it was definitely worth doing the test. Recharging the dessicant and all good, so thanks again. Cheers Ray

Edited by Shiraz, 29 April 2020 - 10:03 PM.

[Gymbow]

Glad you found & fixed the condensation issue.  Out of curiosity, how often does the 1600 require a desiccant recharge?  I’ve only had mine for < 2 years, but haven’t been out at all this season due to various issues, so I doubt it needs a recharge yet. 
 

FYI, I redid the flats wizard setup in SGPro with my light panel set to minimum, and the exposure came out to 1.48s at a mean of 14K ADU, so that should keep me in the clear.  Thanks again for your assistance. 
 

PS:  I noticed you’ve got a whole week more time on here than I do.  

[cups]

First of all, sorry my english; not is my native language.

 

I have been reading the whole thread very interested to see if the doubts I have about the "flats" and the "dark flats" were cleared up but it hasn't been. I understand that they seem to serve both very short flats, as Mr. Rista does, and longer ones, of several seconds. But that information would help me more if I had the information of how many ADUS you get such exposures. Also, same gain? gain 0?
So,

1º I deduce that there are many people who achieve good results (not overcorrection, elimination of doughnuts) with the ASI 1600 (I am especially interested in the OSC model, although I do not know if the data from the MM would also be useful)

2º Would any of you be so kind as to share what time of exposure and how many ADUS in the flat is working for you without over correcting and eliminating the doughnuts? I usually take the lights with 75/15

I've tried everything and I don't know what else I can do to get the lights calibrated properly.
Thanks in advance.

A desperate, frustrated, tired ASI 1600MC user. But I'm not giving up. 

 

[james474789]

Revisiting this thread with something I couldnt see mentioned in the previous posts:

 

The short exposure time is to minimise star clipping, but why wouldnt you take a bunch of long subs (with clipped stars) and then a bunch of short subs (without) before combining them with something like HDRComposition? What is the drawback of this approach compared to avoiding clipping in the first place with a larger bunch of shorter subs?

[james474789]

First of all, sorry my english; not is my native language.

 

I have been reading the whole thread very interested to see if the doubts I have about the "flats" and the "dark flats" were cleared up but it hasn't been. I understand that they seem to serve both very short flats, as Mr. Rista does, and longer ones, of several seconds. But that information would help me more if I had the information of how many ADUS you get such exposures. Also, same gain? gain 0?
So,

1º I deduce that there are many people who achieve good results (not overcorrection, elimination of doughnuts) with the ASI 1600 (I am especially interested in the OSC model, although I do not know if the data from the MM would also be useful)

2º Would any of you be so kind as to share what time of exposure and how many ADUS in the flat is working for you without over correcting and eliminating the doughnuts? I usually take the lights with 75/15

I've tried everything and I don't know what else I can do to get the lights calibrated properly.
Thanks in advance.

A desperate, frustrated, tired ASI 1600MC user. But I'm not giving up. 

Nina has a very good automated flat procedure and will collect flats and matching flat darks with minimal interaction.

Gain and offset should be matched to your lights.

Exposure duration will depend on the brightness of your flat source, gain and filter, but something like nina will aim for approx 50% histogram.

 

Dont bother with bias frames. Just collect darks, flats and flat darks for calibrating your lights.

[heno]

@shiraz

Ray, I have been using your spreadsheet quite a bit and I have a question: 
There is an input field for aperture of the telescope. But what if you have a telescope with a central obstruction? Should that be subtracted so that we a looking at a net value of glass/mirror?
If so, should there have been a separate input line for the central obstruction, for simplicity?
Not sure if I have the latest version of your spreadsheet. Could you provide a link to the latest version if there is more than one?
 

Cheers

Helge

[Odyseus10]

I’ve been searching for a thread like this???
So I have been happily using my ASI294MC Pro with my Redcat51 (f/4.9) and my GT81 IV (f/4.7 with reducer) and shooting 3 minute subs with a gain of 139 (medium gain according to my ASIAir Pro). 
 

I get very good results..

 

So then I decided to buy a ASI1600MM Pro with LRGMHOS filters and took some shots of SH2-101 to photo the CygnusX1 gas bubble.

 

I shot 45 x 3 minutes for R,G,B BUT when I stacked and combined in Pixsight, the resulting photo was very noisy.

 

Reading through this thread, it seems that I should be taking 30 minute subs not 3. Btw I use a guided EQ6-R Pro.

 

Or am I missing something, I sort of wish I had gone for the expensive ASI2600MC Pro and saved myself the hassle of Mono astrophography with narrow band filters : ( 

[Noobulosity]

I’ve been searching for a thread like this???
So I have been happily using my ASI294MC Pro with my Redcat51 (f/4.9) and my GT81 IV (f/4.7 with reducer) and shooting 3 minute subs with a gain of 139 (medium gain according to my ASIAir Pro). 
 

I get very good results..

 

So then I decided to buy a ASI1600MM Pro with LRGMHOS filters and took some shots of SH2-101 to photo the CygnusX1 gas bubble.

 

I shot 45 x 3 minutes for R,G,B BUT when I stacked and combined in Pixsight, the resulting photo was very noisy.

 

Reading through this thread, it seems that I should be taking 30 minute subs not 3. Btw I use a guided EQ6-R Pro.

 

Or am I missing something, I sort of wish I had gone for the expensive ASI2600MC Pro and saved myself the hassle of Mono astrophography with narrow band filters : ( 

Did you mean 30 second subs?  The values for around f/4-f/5 @ gain 139 show roughly 30-50s times in the tables on page 1.

[Odyseus10]

Did you mean 30 second subs?  The values for around f/4-f/5 @ gain 139 show roughly 30-50s times in the tables on page 1.

Yes you are correct , I was getting mixed up between LRGBHOS filters and narrowband such as the L- Extreme.

 

So, it looks like I should be taking 30sec subs with the colour camera and 30sec subs with the mono+LRGBHOS filters for my setups in Bortle 5?.

 

I can’t see how taking 30sec subs instead of 180sec subs will greatly reduce the noise??

Edited by Odyseus10, 18 September 2021 - 11:24 AM.

[TareqPhoto]

Yes you are correct , I was getting mixed up between LRGBHOS filters and narrowband such as the L- Extreme.

 

So, it looks like I should be taking 30sec subs with the colour camera and 30sec subs with the mono+LRGBHOS filters for my setups in Bortle 5?.

 

I can’t see how taking 30sec subs instead of 180sec subs will greatly reduce the noise??

With thousands of frames been taken maybe?

[Odyseus10]

With thousands of frames been taken maybe?

Indeed

Well I will give M31 a go tonight.

Gain set to 200 in ASIair Pro

L = 120 x 60sec

R, G & B each = 80 x 30 sec

Total time 2 hours L and 2 hours RGB = 4 hours in total .

Fingers crossed it stays clear and I get decent data this time ??

[ecorm]

Yes you are correct , I was getting mixed up between LRGBHOS filters and narrowband such as the L- Extreme.

 

So, it looks like I should be taking 30sec subs with the colour camera and 30sec subs with the mono+LRGBHOS filters for my setups in Bortle 5?.

 

I can’t see how taking 30sec subs instead of 180sec subs will greatly reduce the noise??

Unless you're badly underexposing, it's total integration time and background sky brightness that determines the overall signal-to-noise ratio once stacked, not the exposure time of a single frame. You only need to go to a shorter exposure if you have poor guiding or there's too such star clipping. Otherwise, it's better to use longer exposures to keep the number of frames reasonable (at least enough frames for rejection algorithms to do their thing).

 

With the ASI1600, don't be afraid to use lower gain instead of shorter exposure if there's too much star clipping. With newer sensors featuring High Conversion Gain, it's a bit more complicated: you want to avoid going between zero and unity gain where there's that sudden drop of dynamic range (check the dynamic range vs gain curves).

When stacking with the ASI1600, you don't really "lose" photons if your gain is below 1e-/ADU, so don't be afraid of use the lower gain settings. When stacking, you recover bit resolution due to what I call "temporal oversampling" (to avoid confusion with spacial oversampling).

Edited by ecorm, 18 September 2021 - 05:11 PM.

[ecorm]

Indeed

Well I will give M31 a go tonight.

Gain set to 200 in ASIair Pro

L = 120 x 60sec

R, G & B each = 80 x 30 sec

Total time 2 hours L and 2 hours RGB = 4 hours in total .

Fingers crossed it stays clear and I get decent data this time ??

There are differing opinions on this, but in general with LRGB for pretty pictures, you only need to gather enough RGB frames for the rejection algorithm to work. The rest of the time is more efficiently spent gathering luminance (where 100% of the light is captured). What you then do is heavily denoise (or blur) the RGB masters so that they don't contribute to the SNR (signal-to-noise) after combining with the luminance master. Our eyes are more sensitive to changes in the luminance, not the color, when evaluating an image.

The RGB masters should look blurry (but noiseless), but after you combine them with a sharp-looking luminance master, you obtain what looks like a sharp-looking color image! I'm always amazed by this after performing the LRGB combination, even when I have a relatively low number of RGB frames.

Edited by ecorm, 18 September 2021 - 05:26 PM.