Sensitivity improvement in progressive-scan FT-CCDs for digital still camera applications

Abstract

Sensitivity improvements in a 3.2M-pixel CCD image sensor developed for digital still camera applications are presented. The introduction of gap-less microlenses increases the sensitivity with 25-30% while the high angular response is maintained. With the binning possibility at the image-storage transition, the sensitivity in monitor mode can be increased. Finally the sensor output amplifier now combines low noise and excellent linearity with a much higher conversion factor. This improvement is obtained by reduced paracitic capacitances around the Floating Diffusion area. CCDS FOR DIGITAL STILL CAMERASThe pixel race in the field of digital still cameras (DSC) still continues [1,2]. To maintain acceptable consumer price levels, the image diagonal of the CCDs should not exceed 2/3" (11").Thus pixel sizes have to decrease to meet the demands of increasing resolution. Shrinking pixels also implies a reduction of sensitivity. A 3.2M-pixel FT-CCD imager is presented in which three important contributions for maintaining a high CCD sensitivity with decreasing pixel size are combined: new type of microlenses optimised for frame-transfer CCD pixels, charge binning in monitor mode while maintaining color reproduction, and an improved output amplifier combining a high conversion factor and low noise with excellent linearity.

Frame transfer CCDs for digital still cameras: concept, design, and evaluation

“…However, in monitor mode, an image capture rate between 15 to 30 images/s is required for comfortable viewing on an LCD monitor. This of course limits the exposure time to 1/15 or 1/30 s. To increase the sensitivity in monitor mode for low exposure scenes, vertical charge binning is possible [9]. Fig.…”

“…To maintain a sufficiently high sensitivity, microlenses were applied on pixels smaller than 5 5 m . Their primary purpose is to focus the light onto the center, most sensitive part, of the CCD image pixel, [9]. Fig.…”

“…The output amplifier is a three-stage source follower [9], [12]. The capacitance of the floating diffusion detection node was minimized and the gain was maximized to achieve the highest possible conversion from charge to voltage.…”