Resistance is a good solution, it also might be interesting to treat the coil like a resonator.

Try using a uC to give it a rising edge on the end clipped into and measure the rising edge on the other side. I could simulate this and see if it was feasible, but it would probably save hours to use a function generator and an Oscope, or just use a uC. A capture pin on a PIC and a PWM output would be my choice, pin goes high from PWM on a PIC at timer overflow, capture is relative to timers.

Just an Idea, if I think of any others I will let you know. If it could be done you could have two simple interrupts and just w check the variable they wrote to before doing any audio out. -Max

If you used a very thin gauge wire you could measure the resistance from one end of the coil to the connection point. However, even with 28 gauge wire you're looking at a total of 6 ohms for the entire coil, so reading that with a micro is going to require additional components.

Alternately you can measure the inductance of the coil from one end to the tap, which would vary based on which tap was chosen.

However, you're just building an interface. Does it have to be a real coil, or does it just have to look like a coil?

Were I in your shoes, I'd cut the back side of the coil around the taps and use one of many methods for determining which tap was connected. This could be done with one input per tap, or with an analog input if you added resistors to the coil so the resistance significantly increased between each tap.