Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/hst/wfpc2/documents/isr/9503_3.html
Дата изменения: Unknown
Дата индексирования: Mon Apr 11 10:26:55 2016
Кодировка:

Поисковые слова: п п п п п п п п п п п п п п п п п п п
WFPC2 ISR 95-03
STScI Logo

Hubble Space Telescope
WFPC2 ISR 95-03

1. Introduction

What makes CCDs possible is the fact that the charge deposited in a given pixel during the exposure can be efficiently transferred to adjoining pixels along the same column when the CCD is read out at the end of the exposure. The ideal CCD would have pixels with 100% charge transfer efficiency. While this is very nearly true for the vast majority of the 2,560,000 pixels on the WFPC2, there are about 30 pixels which do not transfer charge as efficiently. These "traps", or "defects", may transfer as little as 20% of the electrons during each time step of the readout. Only about 0.001% of the pixels are actually defective but about 0.4% of the image is affected since all the pixels in rows above the trap (i.e., higher value of Y) are clocked out across the bad pixel during the readout. The worst traps leave long tails, and are often known as "bad columns".

Figure 1 shows a portion of an image from WF2 with the location of five of the seven known charge transfer traps on this chip. The strongest trap is at (337,201) which transfers only 19% of the charge during each time step. The rows below Y = 201 are unaffected since they are transferred down the column during the readout and hence never pass over the bad pixel. Figure 2 shows an idealized model of how the bright tail is created, assuming a uniform background of 8 counts in every pixel during the exposure.

Idealized Model of a Charge Transfer Trap Compared to Real Data
Parameters:
           Trap Location (WF2) = (637,389)
           Cosmic Ray Location = (637,645)
           Transfer Efficiency = 34 %
           Background          = 8 counts
 
       Charge         Leftover Charge from   Total        Charge   Background  C(J-1) Observed  C(J-1)
Pixel  Deposited            Previous Pixel  Charge   Transferred   Subtracted  / C(J)   Value   / C(J)
-----  ---------  ------------------------  ------  ------------   ----------  ------ -------  -------
 
                           (equilibrium producing a background of 8 counts)
 
(637,644)      8                      15.53   23.53      * 0.34 = 8.00
 
                                 (hit with a 989 count cosmic ray)
 
(637,645)    989   23.53 -   8.00 =  15.53 1004.53  * 0.34 = 341.53    333.53           333.48
(637,646)      8 1004.53 - 341.53 = 662.97  670.97           228.13    220.13    0.66   217.91  0.65
(637,647)      8                    442.84  450.84           153.29    145.29    0.66   146.06  0.67
(637,648)      8                    297.55  305.55           103.89     95.89    0.66   100.33  0.69

Abstract
1. - Introduction
2. - Characteristics of Traps
3. - Removing the Effects of the Charge Transfer Traps
4. - How Charge Transfer Traps Affect Photometric and Astrometric Results