CCD-world: Fwd: From Jim Janesick

[Tim Abbott]

The following was posted to CCD-world:


Jim's still not able to connect to CCD-world and I still don't know
why not.  In the meantime, the following is forwarded from him and we'll
keep working on getting Jim back in the flow:

MYPIXEL at aol.com wrote:
> *****************jj
> Below .. . 
> *******************jj
> 
> 
> In a message dated 10/24/99 4:12:33 PM Pacific Daylight Time, 
> Tim.Hardy at hia.nrc.ca writes:
> 
> << 
>  We did some measurements of the dark signal noise in some radiation
>  damaged CCDs. I don't know if the results apply to un-damaged CCDs, but
>  what we found was that if we subtracted two dark frames (to eliminate
>  spatial variations), and divided by root 2 to account for the noise
>  increase from the subtraction, the noise in non-MPP mode followed
>  Poisson statistics (i.e. noise = SQRT(mean) ). However, the
>  dark signal noise in MPP mode appeared to follow a linear relationship
>  to the mean signal level with a slope of about 0.008. The noise was
>  always greater in MPP mode than in non-MPP mode, for a given mean dark
>  signal level. I don't have an explanation for this, but perhaps there
>  is a difference between the statistics of the surface dark current
>  generation which dominates the non-MPP mode, and that of the bulk dark
>  current generation which makes up most of the signal in MPP mode. >>
> 
> ***********************jj
> Maybe.  ..  . . 
> 
> For good MPP CCDs  (say  <10 pA/cm^2)  dark current can also come in the form 
> of bulk dark spikes. Spikes are promoted by impurties or lattice damage in 
> the presence of high electric fields near the surface of the CCD (i.e., the 
> leading edge of the buried channel). Field assisted emission occurs producing 
> hot pixels or dark spikes at these defect sites. As dark (or light) charge 
> fills up the well . . . the fields collapse . . . and the dark current rate 
> for the spikes usually decrease depending on initial field strength (this 
> makes despiking by simple frame differencing difficult). 
> 
> Therefore, statistics for dark noise can be very complex. . .  . anything is 
> possible. However, for normal non MPP CCDs, surface dark current generated at 
> the Si-SiO2 interface hides and dominates these spikes. . . dark noise then 
> obeys the shot noise equation (i.e., square root of average signal). 
> 
> Possibly you are seeing some spike influence. Radiation induced spikes are 
> very common. .   especially for MPP CCDs where bulk dark current is dominant. 
> This is a major problem with Hubble. . .  new dark spikes appear all the time 
> . . .
> *************************************jj


-- 
Tim Abbott                                                    tabbott at not.iac.es
Support Astronomer                               http://www.not.iac.es/~tabbott/
                           Nordic Optical Telescope
       Apartado 474 E-38700, Santa Cruz de La Palma, Canary Islands, Spain
- -- CCD-world -- --
CCD-world is fully moderated. Send posts to CCD-world at astro.ku.dk
Standard replies will go to the list; address personal replies manually.
For more information, please go to:  http://www.not.iac.es/CCD-world/