CCD-world: About EEV4220 and binning

[MYPIXEL at aol.com]

The following was posted to CCD-world:

************************jj
Below. . 
************************jj

In a message dated 10/27/99 9:28:17 AM Pacific Daylight Time, 
roger at ctios1.ctio.noao.edu writes:

<< 
 The following was posted to CCD-world:
 
 Christian,
 
 I think you have just rediscovered that
 
   1) binning adds the dark signal of many pixels together, and
 
   2) dark signal for zero exposure time is finite.  
 
 The profile is a little odd (high in the center, rolling off to zero at
 the left and right edges) but otherwise this looks like ordinary
 spurious charge to me. It is only 0.25 e- per physical line which is
 consistent with many CCDs, though I can't speak for the EEV4220.
 
 If indeed it is spurious charge, then the effect should also be present in 
the
 unbinned frame.  Do you get the same profile if you add those 100 lines
 together offchip instead of binning?  Usually the spurious charge will
 be proportional to the number of lines binned. If not, look for slope
 in the column average (after overscan subtraction) of an unbinned
 image, since spurious charge is not always constant from top to bottom
 (eg when different parallel clocking is used for erase and for
 readout.)

**********************jj
m m m m m ..  . spurious charge is very sensitive to clock duty cycle (just 
not waveshaping). .  . things may not be as proportional as you think. 
************************jj
 
 Spurious charge can be mitigated by tri-level clocking (slow exit from
 inversion), slowing the rising edges of the parallel clocks, and/or
 reducing the positive rails.

***********************jj
Last resort . . . I haven't heard of any group that couldn't eliminate 
spurious via waveshaping. But there is always the first time.. ..  especially 
with binning which amplifies the problem. Tri-level may become necessary. Do 
you have to invert during readout?
************************jj
 
 
- -- 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/