Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

What is Gain(e-\adu)? Can someone explain?

  • Please log in to reply
4 replies to this topic

#1 SkyHunter1

SkyHunter1

    Vostok 1

  • *****
  • topic starter
  • Posts: 194
  • Joined: 09 Nov 2017
  • Loc: Long Island NY

Posted 26 March 2020 - 04:54 PM

Hello all,

What is Gain(e-\adu)? Can someone explain? I understand DR, readnoise, full well, QE, etc. 

 

Can someone tell me why it is important and what it means in terms of imaging?

 

Regards,

Skyhunter1



#2 scopewizard

scopewizard

    Vostok 1

  • -----
  • Posts: 163
  • Joined: 04 Oct 2010
  • Loc: Alberta, Canada

Posted 26 March 2020 - 05:04 PM

This may explain it.

https://www.cloudyni...adu-well-depth/



#3 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5,735
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 26 March 2020 - 05:06 PM

Number of electrons that count as one digital number (DN) also known as analog-to-digital unit (ADU).

 

For example, if a camera has a gain of 1, then a digital output of 1000 (above bias level) means the pixel captured 1000 photoelectrons.

 

This is an unimportant camera specification.  It's an arbitrarily chosen number which you can vary with the gain setting (with astrocameras) or ISO setting (DSLR).

 

Gain is also given as ADU/e-, which is the reciprocal.  ADU/e- goes up with higher ISO setting; e-/ADU goes down.


Edited by Michael Covington, 26 March 2020 - 05:06 PM.


#4 jfrech14

jfrech14

    Viking 1

  • *****
  • Posts: 923
  • Joined: 17 Jan 2012
  • Loc: Virginia, USA

Posted 26 March 2020 - 06:40 PM

Personally, I would argue that gain is an essential camera parameter because it directly changes your maximum possible dynamic range because it changes where you saturate, changes your read noise, and while you can change the gain on the newer CMOS sensors on the fly, you don't necessarily want to because calibration becomes a headache. CCDs usually have a factory set gain that maximizes the dynamic range or makes it linear across the entire ADC readout range. There is a good reason it is a primary parameter in sensor manuals.

But to the question, the other answers are right. It is the conversion from electrons before the analog-digital converter to digital units after the conversion. So, it is photoelectrons, readout electrons, and dark electrons converted into a digital value through this multiplicative gain. Because, in a CCD for instance, there is a pre-amplifier that amplifies the electrons in the pixel well to some value that gives you a better dynamic range. Sensor pixels can hold, say 10,000 electrons and a 16-bit ADC can hold 65,535 values. So you can change the gain so that when the pixel saturates at 10,000 electrons, it fills the ADC up to 65,535. Expensive sensors have pixels that can hold 150,000 electrons, so the gain would have to be less than 1 ADU/e- to have those 150,000 electrons fill a 16-bit ADC.

But, those are the old days of CCDs. It is very common for people to set CMOS gains higher because they aren't going to saturate the sensor. So, with the higher gain, you can take shorter exposures and your signal will fill more of the ADC. So, if you sensor pixels can each hold 10,000 electrons, but your exposures are only a few seconds long and your maximum value is 1,000, you can give it a much higher gain (ADU/e-) so that the 1,000 electrons fills 65,535 rather than having 10,000 fill 65,535 and wasting 90% of your ADC range. It really depends on what sort of imaging you want to do.



#5 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5,735
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 27 March 2020 - 07:54 AM

When I said it's unimportant I mainly had in mind that with DSLRs, instead of calculating gain, we measure dynamic range against ISO setting directly.  Earlier DSLRs (up through two or three years ago, e.g., Canon 60D) have to be set to about ISO 800 to overcome read noise.  Newer ones (Sony, Nikon D5500, Canon 80D) have less read noise and work well at ISO 200, with greater dynamic range.

It happens that the gain in ADU/e- is about 3.3 on a Canon 60D at 800, and about 1.0 on a Canon 80D at 200.  In e-/ADU, these are respectively 0.33 and 1.0.

 

So unity gain is useful for finding the lowest ISO setting you can use -- but read noise is likely to require you to use a higher ISO setting anyway.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics