In trying to calibrate the tach I tried to find a way to do so on the bench out of the car. This means that I would have to come up with a reliable way to reproduce the currents and pulse signals etc that the tach would normally encounter in the car.
The tachometer measures frequency. For every rotation of the engine the tachometer gets two pulses. This means that at 1000 rpm the signal is at a frequency of 33 1/3 hz (one hertz is one cycle per second). In theory, if one could put the right frequency in to the unit, one should be able to easily bench calibrate it at any rpm. With that thinking in mind, I used my computer to generate a set of signals at set frequencies.
I tried a number of waveforms. A square wave, a sawtooth, a sine wave, and I even generated one that was based upon an oscilliscope reading of a car. I then borrowed my fathers audio test equipment. After hooking it up to a power supply and a powerful amp, we discovered that the unit is rather sensitive to both the shape of the waveform and the voltage of the signal. The amp was quite powerful and yet using that method we could not create a signal that would be useful for calibration. On to plan 2. My dad and I borrowed an oscilliscope and reconnected the tach to the car. The oscilliscope would be used to measure the frequency of the signal from the low tension lead of the coil and thus the speed of the engine. If you know the frequency, just multiply it by 30 to get the rpm. Once you know the rpm, simply adjust the callibration pot with a small flathead screwdriver and you're done.
Now I know that most of you wont have access to an osciliscope, but what also can work well instead is a timing light with tachometer. There actually are some timing lights that have this feature and it may well be a worthwhile investment. It certainly would have made the whole thing a lot easier.