Datasets
Standard Dataset
Power Efficient Signal Generation for Capacitive Sensors
- Citation Author(s):
- Submitted by:
- Mehdi Moradian
- Last updated:
- Fri, 04/15/2022 - 03:29
- DOI:
- 10.21227/bvvv-9d30
- License:
- Categories:
Abstract
For the interaction of humans with machines but also for the interaction of machines with the environment, e.g. in robotic manipulation tasks, large area sensors like sensor skins are of high interest. Capacitive sensor have become widely used for touch sensor and proximity sensors and are well suited for such large area sensing. However, for the integration on a sensor skin, a small outline of the sensor electronics is important and since many applications use wireless connectivity for sensors, minimizing the power consumption is also very relevant to minimize the size of energy storage or energy harvesting systems. For capacitive sensors a major part of the power consumption relates to the generation of the excitation fields. This is in particular relevant when carrier frequency and broad band approaches are used, e.g. spread spectrum approaches to ensure robustness to external disturbers and enabling coexistence of multiple sensors in close vicinity, as exploitation of narrow band resonators cannot be utilized.
To improve the performance of capacitive sensors in terms of power consumption and size, this paper proposes a of a closed-loop class D amplifier fully on-chip driver circuit that does not require any off-chip components such as inductors. Using flip chip bonding the chip can be integrated e.g. in sensor skins with minimum space requirements. It is suitable to drive large parasitic capacitances (e.g. for active guards) of up to 1 nF over a frequency range between 10 kHz and 10 MHz. Experiments with an initial test chip showed a current consumption of 1.92 mA for a sinusoidal signal with a frequency of 1 MHz and a signal amplitude of 1 V applied to a load capacitance of 1 nF, which is significantly lower than for linear rail to rail capable amplifiers used in capacitive sensing
just need a video player to watch the video recorded during experiments