Project help: arduino mobile metal detector

I am attempting to interface this metal detector circuit to the arduino where it says 'to M.W. Radio' using PWM for my mechatronics course. I am using a 32 turn 7 inch detector coil in place of the 70 turn 120mm coil originally used. This may be calculated incorrectly and I would appreciate it if someone could give me applicable correct equations to resize the coil. At our last open lab we hooked everything up and connected it to the 6V power supply, then looked at the output at the above mentioned point using a oscilloscope. When waving metal or a magnet over the coil, there is no signal response in the output. I would really appreciate it if someone could look over the schematic and what the expected output should be, then advise me on how to proceed. Thanks very much, T. Warcup

I also wanted to mention that even when isolated from any nearby source of signal, the oscilloscope trace shows random spikes unrelated to any nearby metal or magnet.

really? not even a hint? =/

"Although 230kHz is out of range of the Medium Wave band, an MW radio will clearly pick up harmonics of this frequency."

What frequency were you attempting with your inductor? At 230 KHz, will be hard to do anything with the arduino. You'll have to do something like buffer the output, make an envelope follower circuit or something to get the output down into the audio range where you can work with it.

Hmm, I was hoping to keep it really simple, is there an easier way to do metal detection with the arduino? The only other ways I have seen through research are people taking a pre existing metal detector and doing it that way..

Then you need to make a "local oscillator" near the frequency of what your circuit puts out, mix the two together so one cancels the other, and the difference that is left is what you want to digitally sample with the arduino and act upon.

230 kHz is no big deal depending on how you tackle it. Have a look at my crystal deviations experiments.

That is: you could generate something as close as possible to 230 kHz with one of the hardware timers. E.g. 16 MHz / 64 = 250kHz. Then you could feed this into one pin of an optocoupler and your signal into the other. Similar to my experiment you will then see something at the difference frequency ~ 20kHz. This is much easier to handle.

Also 230 kHz is below 4 MHz thus you could feed it directly into one of the hardware counters. E.g. digital 5 == PD5 == T1 for feeding into counter 1. For details you may want to consult the datasheet.

I am using the duemilanove so I will look at the datasheet I have, and start learning how to use the timer. That sounds like it might be promising. I am an ME student so I only have very rudimentary electronics knowledge .. oddly I am great at the programming aspect of the mechatronics course, but not so great at the electronics. Hopefully this will pan out. Thanks, -T