For work I need to interface a Crossbow three-axis accelerometer to our usual data collection instrument which, unfortuantely, only has a single pulse counter channel available. So I'm designing a circuit that takes three AC voltages with a DC offset and turns them into a single frequency modulated square wave. Obviously I'm losing some information; I have to translate the voltages to being zero-centered, then rectify them, sum them, amplify the result, then offset again. The amplitude of the resulting signal is basically the amplitude of vibration in all three axes summed together. This signal becomes the frequency modulation input on the Maxim MAX038 function generator IC. The output from the MAX038 is a frequency modulated square wave that varies by +/- 70% from its center frequency, which I haven't yet chosen.

Anyway the point of this story is that all of this signal conditioning was done in analog electronics, with op-amps-a-plenty. And I'm really not an analog guy. I had a schematic all done and was pretty happy with it, then I showed it to Kiril (KEER-eel)--our Bulgarian analog electronics guru down in the lab. He was very nice about it, but was able to make a few simplifications. For instance, within moments of looking at the schematic, he was able to draw a circle around an opamp and four resistors and point out that the functionality of this block could be replaced by a single capacitor and resistor.

Somehow I had decided it would be a good idea to use a full difference amplifier and a reference voltage source to remove the DC offset. RC circuit. Yeah.