Sensors for a world where smaller is better

What started off as the development of a micro digital sunsensor within frame of the MicroNed program has led to a highly optimized single chip detector with a 5*5*0,5 mm size and 25mW power consumption. The target accuracy for a sensor based on this chip is 0,01 degree of arc 1o which would give an unprecedented level of performance. The actual sensor could be the same size as the chip although a slight increase in height would be needed (0,8mm) in case no connections would be required. An actual application however would require both power to be supplied to the chip as well as data connections to and from the chip. In order to optimize the total system size and mass, TNO has been investigating several approaches among which autonomous powering through a dedicated solarcell and wiretess data communication. In all cases investigated the peripheral circuitry by far exceeds the actual sensor core. In a number of cases the connector is even the largest part of the system. Since alignment problems have to be taken into account also (10 µm error on a 5mm sensor constitutes a 2mrad or approximately 0,01 degree alignment error) our tendency is to look at micro sunsensors as part of integrated sensor systems. Further miniaturization of sunsensors (from micro to nano) is currently seen as not profitable as the size of the overall system will not be significantly Influenced by a smaller sensor. Since there is no clear definition for a nano sensor (the above mentioned chip uses a 180nm CMOS process so could be considered a nano sensor with a lot of fantasy) the discussion will be blurred, but our preliminary conclusion is that no further mintaturization of the sensor is ret quired. A similar effort to minimize the peripheral circuitry will have a much larger effect on the overall system size. The presentation will focus on the state of the art with respect to sunsensor miniaturization and will address issues like autonomous and integrated sensor systems.