Camera-on-a-chip

Motorola has announced a small, low power VGA image sensor – the MCM20114 – engineered to interface directly with it’s DragonBall MX1 microprocessor.

Motorola has announced a small, low power VGA image sensor – the MCM20114.

Engineered to interface directly with Motorola’s DragonBall MX1 microprocessor, the new sensor can support a variety of electronic shuttering modes. A global shutter mode is designed to allow manufacturers to eliminate mechanical shutters, thus reducing camera cost. The continuous rolling shutter mode supports video applications in excess of 30 frames per second. The sensor will also support sub-sampling for lower resolution applications and window-of-interest for panning and digital zoom.

Integrated on chip is the required circuitry for timing, I2C control and row and column decoding. The device also features an integrated analog image processing chain including programmable gain amplifier, programmable white balance control, and a 10-bit analog-to-digital converter.

The sensor features 5.6 micron square pixels, three shutter modes, frame rates in excess of 30 frames per second, a Bayer RGB colour filter array, and micro lenses for greater sensitivity. A redesigned image processing chain is designed to reduce the device power requirements by up to 50% over Motorola’s first generation VGA products. With a pixel area reduced by nearly 30% from first generation sensors, the optical format shrinks from one-third to one-quarter inch. This not only reduces the die size, but also helps to reduce the cost of the imaging optics.

With a maximum power consumption of only 100mW, the MCM20114 is suitable for battery powered devices.

General sampling of the MCM20114 VGA sensor is expected to begin in March of 2002, with volume shipments expected in the third quarter of 2002. The suggested resale price for the MCM20114 is $6.00 in quantities of 100,000.

An MCM20114 based imaging module with integrated optics, for small form factor camera applications, is expected to be available for sampling in the second quarter of 2002.

On the web