KAF−0261

http://onsemi.com

4

Image Acquisition

An image is acquired when incident light, in the form of

photons, falls on the array of pixels in the vertical CCD

register and creates electron−hole pairs (or simply electrons)

within the silicon substrate. This charge is collected locally

by the formation of potential wells created at each pixel site

by induced voltages on the vertical register clock lines

(

fV1, fV2). These same clock lines are used to implement

the transport mechanism as well. The amount of charge

collected at each pixel is linearly dependent on light level

and exposure time and non−linearly dependent on

wavelength until the potential well capacity is exceeded. At

this point charge will ‘bloom’ into vertically adjacent pixels.

Charge Transport

Integrated charge is transported to the output in a two−step

process. Rows of charge are first shifted line by line into the

horizontal CCD. ‘Lines’ of charge are then shifted to the

output pixel by pixel. The timing diagram illustrated in

Figure 7 illustrates how the integration of charge is

performed with

fV1 and fV2 held low. Transfer to the

horizontal CCD begins when

fV1 is brought high, causing

charge from the

fV1 and fV2 gates to combine under the

fV1 gate. The fV1 and fV2 gates are now reversed in

polarity, causing the charge packets to ‘spill’ forward under

the

fV2 gate of the next pixel. The falling edge of fV2 also

transfers the first line of charge into the horizontal CCD. A

second phase transition places the charge packets under the

fV1 electrode of the next pixel. The sequence completes

when

fV1 is brought low. Clocking of the vertical register

in this way is known as accumulation mode clocking. Next,

the horizontal CCD reads out the first line of charge using

traditional complementary clocking (using

fH1 and fH2

pins) as shown. The falling edge of

fH2 forces a charge

packet over the output gate (OG) onto one of the output

nodes (floating diffusion) which is buffered by the output

amplifier. The cycle repeats until all lines are read.