Return to the Cloudy Nights Telescope Reviews home page


Astrophotography and Sketching >> CCD Imaging & Processing

Pages: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | (show all)
Peter in Reno
Postmaster
*****

Reged: 07/15/08

Loc: Reno, NV
Re: 5 or 30 minute exposures for narrowband? new [Re: bilgebay]
      #5661608 - 02/04/13 11:51 AM

Quote:

Quote:

Hi Frank,

Try these numbers I found from another person's Atik 460EX camera:

Gain = 0.26
Readout Noise = 4.79e-
Total System Noise = 5.10e-

Temperature was at -15C.

Thanks,
Peter




Hi Peter,

Do these figures belong to my 460EX?




Hi Sedat,

I am not sure. I got it from "Measure the performance of your CCD camera!" thread started by Konihlav. Are they yours?

Peter


Post Extras: Print Post   Remind Me!   Notify Moderator  
freestar8n
Post Laureate
*****

Reged: 10/12/07

Re: 5 or 30 minute exposures for narrowband? new [Re: Inverted]
      #5661670 - 02/04/13 12:30 PM

I don't think there is any way to explain this stuff that will work for everyone - but clearly the existing methods aren't working perfectly. I think a focus on the noise itself makes perfect sense and will be understandable - and this background noise is completely distinct from whatever choice of signal is made.

It sounds like the main source of confusion you are alluding to is that subs look cleaner in light pollution simply because they are a uniform gray. But anyone with a little experience in this stuff is familiar with stretching an image to bring out the details - and that process will remove the "bias" you are referring to (sky glow signal) and reveal the true noise in the exposure due to sky glow shot noise and read noise - and there shouldn't be any doubt the one with light pollution is inherently more noisy even though the raw image looks more uniform gray.

In fact - many viewers will automatically stretch the sub - so the noise would be apparent.

Either way - I'm trying to indicate the actual amount of background noise in various scenarios - and I think a rising sqrt plot with bumps caused by read noise is a good way to do it.

Frank


Post Extras: Print Post   Remind Me!   Notify Moderator  
bilgebayModerator
Post Laureate
*****

Reged: 11/06/08

Loc: Turkiye - Istanbul and Marmari...
Re: 5 or 30 minute exposures for narrowband? new [Re: Peter in Reno]
      #5661733 - 02/04/13 01:00 PM

Quote:



Hi Sedat,

I am not sure. I got it from "Measure the performance of your CCD camera!" thread started by Konihlav. Are they yours?

Peter




My bad! I thought I have sent you my subs and you calculated those values from those whereas all I have sent to you was combined fits files.

I am getting old

Meanwhile, I owe these files to Pavel. As soon as get to my obs I will collect those files and send tehm to Pavel so that he has better values with his spreadsheet.

Cheers

Sedat


Post Extras: Print Post   Remind Me!   Notify Moderator  
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: 5 or 30 minute exposures for narrowband? new [Re: freestar8n]
      #5661747 - 02/04/13 01:09 PM

Right, the dynamic range of the signal is effectively clipped in the LP scenario. Normalizing and rescaling to show relative signal proportions of the various scenarios, will show that the maximum signal strength is limited by sky-glow and the contrast is reduced i.e. the standard deviation of the pixels is lower. It's fine to show the total background noise though, as long as people understand that. You could reduce the background noise from the dark-site, by using a histogram transfer function to clip the range and achieve a similar result. I don't think it is anything that need be particularly mystical, but if we ignore the obvious then it's easy to misinterpret the results....

I think your method is great though and perfectly valid in full context.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Alph
Carpal Tunnel


Reged: 11/23/06

Loc: Melmac
Re: 5 or 30 minute exposures for narrowband? new [Re: Inverted]
      #5661872 - 02/04/13 02:11 PM

Quote:

Right, the dynamic range of the signal is effectively clipped in the LP scenario. Normalizing and rescaling to show relative signal proportions of the various scenarios, will show that the maximum signal strength is limited by sky-glow and the contrast is reduced i.e. the standard deviation of the pixels is lower. It's fine to show the total background noise though, as long as people understand that. You could reduce the background noise from the dark-site, by using a histogram transfer function to clip the range and achieve a similar result. I don't think it is anything that need be particularly mystical, but if we ignore the obvious then it's easy to misinterpret the results....





Let me use the statistician's lingo with you. Do you know why stars are not visible with naked eye in broad daylight? The reason is that the variance or the standard deviation of photon count from the sky background becomes extremely large (Poisson distribution). That makes the difference in the light levels between a star and the sky statistically insignificant. As a statistician you should know how to remedy the problem. You need to increase a sample size which in our case means increasing the total exposure time. A much more pragmatic solution though is to make a trip to a dark site to cut out the sky background from the equation.


Post Extras: Print Post   Remind Me!   Notify Moderator  
cn register 5
scholastic sledgehammer


Reged: 12/26/12

Re: 5 or 30 minute exposures for narrowband? new [Re: Alph]
      #5662148 - 02/04/13 05:18 PM

Or just - wait until it's dark

It's an interesting idea that daylight is a rather extreme form of light pollution.
We can't all go to a dark sky site, or even send our kit there. Working out our best - or least worst - options for coping with where we are and the equipment we have seems useful to me.

From what I can see rules of thumb are:
Use narrow band filters, especially if the site isn't dark because of light pollution, moonlight or twilight.
Longer exposures will help, especially at a dark site, but less so at a light sight. Exposures of 5 to 10 minutes will be good enough that other noise sources will dominate at a light site but at a dark site you will be significantly better off using longer exposures - 20 minutes or more. Even with shorter exposures you will be better off than at a light site.

The thing I find hard to believe is that light pollution will reduce the amount of light received from the deep sky objects. I don't see how that can happen and I think this is a misinterpretation of what the experts are saying. I think that what happens is that the light from the deep sky object stays the same but that the light pollution is added to this.
While the average value of the light pollution can be removed it has it's own noise and this noise is added to the image (in quadrature of course) and this makes the image more noisy. This reduces the signal to noise ratio because the noise has increased, not because the signal has decreased.

Hope this makes sense,

Chris


Post Extras: Print Post   Remind Me!   Notify Moderator  
Mike7Mak
Pooh-Bah
*****

Reged: 12/07/11

Loc: New York
Re: 5 or 30 minute exposures for narrowband? new [Re: cn register 5]
      #5662193 - 02/04/13 05:43 PM

Quote:

...and this noise is added to the image (in quadrature of course)...



Is there a quick explanation for what 'quadrature' means in this context? The only other time I've seen that word it involved feed systems for multi-element antenna arrays. That was long enough ago that I don't remember if I actually understood it then either.


Post Extras: Print Post   Remind Me!   Notify Moderator  
freestar8n
Post Laureate
*****

Reged: 10/12/07

Re: 5 or 30 minute exposures for narrowband? new [Re: Mike7Mak]
      #5662211 - 02/04/13 05:55 PM

The sum of the noise goes as the square root of the sum of the squares of all the independent noise terms.

It's just how independent errors accumulate. If you had aluminum rods that were 1m long with an error of 1mm, if you stacked two together the length would be 2m with an error of 1.4mm - rather than an error of 2mm.

So it isn't just an esoteric Poisson noise thing - it applies to stacked aluminum rods also.

Frank


Post Extras: Print Post   Remind Me!   Notify Moderator  
freestar8n
Post Laureate
*****

Reged: 10/12/07

Re: 5 or 30 minute exposures for narrowband? new [Re: Peter in Reno]
      #5662221 - 02/04/13 06:02 PM

Hi Peter-

I looked at your two darks at 0c, 15m and 30m, and I see a difference of about 14 adu or 5.6e in 15 minutes. That amounts to 0.006e/s. Although that is tiny, it is only about 1/3 of your total "sky" signal - i.e. your actual sky signal may be only 0.014e/s but it is added to a dark current signal of 0.006e/s. In terms of the noise model there is no difference between sky and dark current signal.

Note that your darks don't appear to have pattern noise - so this isn't about needing to subtract the darks. But if you can work at a lower temperature it might help. It would help because your noise is dominated by this sky/current term at your current f/ratio and filter width.

It's just another example where it's good to know the values in your system so you can determine what the dominant noise terms are, so you can address them with priority. If you can go cooler without too much difficulty, you might see a bigger win than with longer exposures.

Frank


Post Extras: Print Post   Remind Me!   Notify Moderator  
Mike7Mak
Pooh-Bah
*****

Reged: 12/07/11

Loc: New York
Re: 5 or 30 minute exposures for narrowband? new [Re: freestar8n]
      #5662231 - 02/04/13 06:06 PM

Thanks Frank, I'm glad you included the aluminum rods.

Post Extras: Print Post   Remind Me!   Notify Moderator  
Peter in Reno
Postmaster
*****

Reged: 07/15/08

Loc: Reno, NV
Re: 5 or 30 minute exposures for narrowband? new [Re: Mike7Mak]
      #5662314 - 02/04/13 07:04 PM

Hi Frank,

Thanks for analyzing my data. It looks like I have a pretty good imaging equipment.

I can get the camera running colder if I can image under clear sky in the Winter. This Winter has been miserable with cloudy nights.

Peter


Post Extras: Print Post   Remind Me!   Notify Moderator  
korborh
scholastic sledgehammer
*****

Reged: 01/29/11

Loc: Arizona
Re: 5 or 30 minute exposures for narrowband? new [Re: Peter in Reno]
      #5662512 - 02/04/13 09:24 PM

Hi Peter, do you run your camera at 100% power unregulated? You may want to try that - maximize cooling and no darks.

Post Extras: Print Post   Remind Me!   Notify Moderator  
Peter in Reno
Postmaster
*****

Reged: 07/15/08

Loc: Reno, NV
Re: 5 or 30 minute exposures for narrowband? new [Re: korborh]
      #5662552 - 02/04/13 09:50 PM

Well, Sony CCDs don't need to run at super cool temperature like Kodak CCDs run better at colder temperature. Plus I don't want the cooler to run too hard so that it would last longer.

The reason for 0C is that when I was imaging Bubble Nebula in late August and through September, the ambient temperature was quite high and it was pushing pretty hard at -5C, so I set it at 0C.

How can you tell if the cooler is running unregulated? I thought the cooler isn't running until the camera is connected to the PC and the capture software is running. I thought in order to cool the camera, you have to enable TEC and set the temperature set point otherwise the cooler may not be running. I just tested it with Nebulosity. When I turn on TEC, there is a temperature set point box and the cooler cools down to the set point temperature. I don't see a way to run it unregulated.

Peter


Post Extras: Print Post   Remind Me!   Notify Moderator  
Inverted
sage
*****

Reged: 01/19/13

Loc: LP Land
Re: 5 or 30 minute exposures for narrowband? new [Re: Alph]
      #5662594 - 02/04/13 10:14 PM

Yes, that makes sense.

Edited by Inverted (02/04/13 10:45 PM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
shams42
Pooh-Bah
*****

Reged: 01/05/09

Loc: Kingsport, TN
Re: 5 or 30 minute exposures for narrowband? new [Re: Peter in Reno]
      #5662603 - 02/04/13 10:19 PM

Quote:

I don't see a way to run it unregulated.





The TEC cooler has no moving parts so I don't think it can wear out.

To run it unregulated, just specify the set point lower than it can achieve. It will run at 100% to try to get there.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Mike7Mak
Pooh-Bah
*****

Reged: 12/07/11

Loc: New York
Re: 5 or 30 minute exposures for narrowband? new [Re: Peter in Reno]
      #5662606 - 02/04/13 10:20 PM

Quote:

I don't see a way to run it unregulated.



I think he means pick a set point lower than the camera can possibly achieve so the cooler runs continuously at 100%.

Although I would hope the Atik cooling system can survive being run maxed out continuously, I kinda agree doing so seems a bit like abuse.


Post Extras: Print Post   Remind Me!   Notify Moderator  
Peter in Reno
Postmaster
*****

Reged: 07/15/08

Loc: Reno, NV
Re: 5 or 30 minute exposures for narrowband? new [Re: shams42]
      #5662610 - 02/04/13 10:21 PM

I was referring to heat. The more you want TEC to cool, the hotter TEC runs, therefore can shorten the life. My SXVR-M25C camera body runs pretty hot when exceeding 100% so I had to lower the set point temperature so the TEC circuit does not run too hot.

I prefer to run at less than 100% and Sony CCDs do not need to run too cold thanks to their low noise CCD design at warmer temperature. Take a look at my sample darks and bias I uploaded earlier.

Peter

Edited by Peter in Reno (02/04/13 10:25 PM)


Post Extras: Print Post   Remind Me!   Notify Moderator  
shams42
Pooh-Bah
*****

Reged: 01/05/09

Loc: Kingsport, TN
Re: 5 or 30 minute exposures for narrowband? new [Re: Inverted]
      #5662651 - 02/04/13 10:40 PM

Hi Frank,

Could you run the numbers for me? I wonder if going from 20min to 30min subs would be helpful.

Here's my data:

TOA-130 @ f/5.76 ~2.3" FWHM
QSI-583 with Astrodon 5nm Ha filter
suburban site

TEC cooler set to -20 deg C

Camera gain: 0.46 e-/ADU
Camera read noise: 9.6e-
Sky flux: 0.153 e- / s


Post Extras: Print Post   Remind Me!   Notify Moderator  
korborh
scholastic sledgehammer
*****

Reged: 01/29/11

Loc: Arizona
Re: 5 or 30 minute exposures for narrowband? new [Re: shams42]
      #5662686 - 02/04/13 11:16 PM

Hi Peter - unregulated as in setting a very low temperature so power is always 100% and temperature chip varies as per ambient. It will run at maximum cooling but not above 100%. Are you saying your camera ran above 100% and hot ? That is strange - I never saw that with Maxim on my Atik.

Post Extras: Print Post   Remind Me!   Notify Moderator  
Peter in Reno
Postmaster
*****

Reged: 07/15/08

Loc: Reno, NV
Re: 5 or 30 minute exposures for narrowband? new [Re: korborh]
      #5662757 - 02/05/13 12:31 AM

Nebulosity does not show temp percentage capacity for my SX camera but it does for my Atik camera. For example if I set the SX camera temp set point to -20C and the CCD temp reading stays at -15C or warmer then it tells me I set at maximum cooling but the camera's body gets very warm. So I would set to -5C to -10C to make sure TEC is not abused.

Atik may have a better TEC design than SX since Atik's camera's body never gets warm. Atik cools a lot slower than SX but that's okay because once Atik's temperature hits the set point, it stays there very consistently.

Nebulosity displays over 100% (even over 200%) capacity when I select Atik native driver. It may be a bug in Nebulosity. I usually ignore this and pay attention to outside temperature and make sure delta temperature between outside and CCD does not exceed at least Atik spec of 25C.

Peter


Post Extras: Print Post   Remind Me!   Notify Moderator  
Pages: 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | (show all)


Extra information
11 registered and 14 anonymous users are browsing this forum.

Moderator:  Oldfield, droid, bilgebay, WOBentley, dr.who 

Print Thread

Forum Permissions
      You cannot start new topics
      You cannot reply to topics
      HTML is disabled
      UBBCode is enabled


Thread views: 7608

Jump to

CN Forums Home




Cloudy Nights Sponsor: Astronomics