On 2/18/08, Jinx wrote: > > I'm sure your circuit works as advertised. > > > > But for a 6" distance, the coupling between a hand and a sensor > > that is roughly 6"X6" is on the order of a picofarad. And the ability > > to reliably resolve a detection distance on the order of 6 inches > > requires a circuit that can resolve much smaller than a pF. I think > > your circuit can achieve this if the potentiometer is adjusted > > correctly and the noise is low enough > > Hi Scott, yes, believe it not, the circuit does work quite well at > any scale of distance vs sensor size. You'll notice that the opamp > o/p is filtered. What you'd see as a hand approaches the sensor > (from memory, it's been a while since I had a scope on this) is a > decreasing amount of the 75kHz getting through, and the DC > component gradually increasing. Filtering adds that AC to the > DC for a cleaner signal. At a certain distance there is little AC > to smooth and a more rapid change of DC level > > I have found that the capacitive qualities of people has quite a > range. I'm quite good at activating these but others aren't. Even > those with big hands. It mightn't sound right that physiology is > so different between people of the same sex and approximate > size, but that seems to be the case. It may even come down to > what they had for breakfast - eg electrolyte levels, blood > pressure, sweat gland activity, who knows what else. I've never > been troubled by static discharges, unlike those who pick up > very annoying charges in offices or cars. There's probably a > mechanism like that for capacitance The circuit is kind of similar to my first project as a real electronics engineer back in 9 years ago. It is a capacitive level sensor (actually it can detect something like cement as well). It is also using the simple oscillator (based on CD40106) and (quasi-)peak detector. And yes it is good enough to detect pico-farad changes. The calibration of the detection threshold is done with a poti. The major problem with the circuit at the time was the RF immunity of the peak detector. After fixing the other problems with 3 PCB layouts, only with the 4th PCB layout I figured out how to solve the RFI problem. I added a small resistor in series with the diode (after the comparator) and finally it passed the tests. Xiaofan -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist