Door access/Howard Community College/Fall2012/p2-502-cash/microphone

Microphone
Included in the project bucket are a variety of microphones. One, consisting of a cloth strap with two round black objects and two wires, is a carbon button throat microphone from a Swiss army tank. Another, consisting of a white plastic speaker like object with a rubber gasket, is the speaker from a telephone handset. Both have been found to be fairly sensitive and should be fairly easy to mount. When mounting, some sort of elastic device should be used that holds the microphone against the door with some pressure, so that if the mounting comes a bit loose, the microphone is still well connected to the door. For the Swiss army tank microphone, this could be as simple as bending the spring in the middle of the two buttons a bit flat (be careful not to bend it beyond it's elastic limit), and taping securely to the door with the aluminum foil tape (Aluminum foil tape was found to be the stickiest of the options available in the engineering lab). For the telephone speaker, this could be a bit more complicated, requiring a bracket taped to the door and the speaker attached with springs or rubber bands, both available in the lab.

Swiss Army tank Microphone
Unlike the piezoelectric loudspeaker, this microphone does not make any voltage by itself. It changes the resistance in a circuit by having grains of carbon pack together or loosen up in response to sound waves. To turn this change of resistance into a voltage to feed to an amplifier, an external circuit must be used. Two such circuits are shown in this schematic. The first uses potentiometer R1 as a current to voltage converter. Current flows from the positive terminal of the power supply through the series combination of the potentiometer and microphone to ground. As the microphone's resistance changes with sound pressure variations, the total resistance of the circuit changes with it, and therefore the total current draw changes. However, R1 remains a fixed value over any knock signal period, so current changes in the circuit produce voltage changes across R1. These voltage changes are taken off at the junction of R1 and the microphone, and fed through DC blocking capacitor C1 to the amplifier's input terminal. R1 is a 10k pot, and should initially be set to its full resistance in the circuit, then adjusted downward for maximum sensitivity. With a power supply of 12 volts and an adjustment of 5k-8k ohms, the sensor was showing almost 2 volts (!) on the oscilloscope screen. The second circuit, at right in schematic, uses a step up transformer to simultaneously increase the voltage produced by the microphone and block DC from getting on the arduino. If a suitable transformer can be found, it may be possible to use a much simpler, lower gain amplifier, or even to omit the amplifier all together. You might try coupling the microphone to a tv flyback transformer; one should be available in the lab, or in old style computer monitors. Some resistance, about 3 to 4 thousand ohms,should be kept in series with the microphone in either circuit to limit current to 3 or 4 mA (this value would change if a different power supply voltage were used). For additional information on circuit operation and characteristics, read the notes in the schematic.



ITT telephone speaker
This is the white object with the black gasket. It makes its own voltage by having sound waves move a coil of wire through a magnetic field, so it does not require the circuitry described above in connection with the tank microphone. It is fairly sensitive, though not as much as the tank microphone above. There is no DC component, so it may be connected directly to an amplifier's input, no blocking capacitor needed. When mounting, the end with the holes in it should be placed against the door, and the rubber gasket placed between the speaker and the door (actually, that configuration was found to work well, but experiment; you may find a configuration that works better).

Piezoelectric Loudspeaker
This is the small black and white plastic object with two wires sticking out of it and a hole in one side. This was found to be very insensitive, difficult to connect reliably to a surface, and generally unpredictable. Though it does make a signal, focusing efforts on it would probably not be the most productive use of time.



Return to Team Page