The metal detector shown here has, in concept, been widely recognised as a new genre. The general concept, of which I have developed three embodiments, is capable in principle of matching the performance of an Induction Balance (IB) metal detector. This is the first embodiment to be released on the Internet (the other two were published in Everyday Practical Electronics and Elektor magazines). When this circuit is correctly set up, an old Victorian penny (30mm diameter) should induce a shift in frequency of at least one tone on the Medium Wave band at 140mm (5½").
Apart from using two overlapping coils, the concept is fundamentally different to IB. Unlike IB, its Rx section is an integral part of the oscillator. Further, unlike IB, the design does not require the critical placement of the coils, which should have significant advantages for manufacture. A special characteristic is that sensitivity covers a wide area of the coils, thus making the design well suited to sweeping. It also provides discrimination. Further, while the design uses a beat frequency oscillator (in this case a MW radio), it differs fundamentally from a BFO metal detector. Its performance far outstrips that of BFO — and further, unlike BFO, it is dependent on the mutual inductance of two coils (BFO, of course, uses only one).
The circuit should be instantly recognisable as a transformer coupled oscillator (TCO) — a well known oscillator type. This essentially consists of an amplifier which, by means of a transformer, feeds the output back to the input, thus sustaining oscillation. In the circuit, the TCO transformer is replaced with two search coils, L1 and L2. These have the same action as the transformer in a TCO, L2 being the "transmitter", and L1 the "receiver". On the basis of its similarity with a TCO, I named this metal detector a Coil Coupled Operation (CCO) Metal Detector. The presence of metal induces changes both in the inductance and the coupling of the two coils, thereby inducing a shift in the oscillator frequency. A single stage common emitter amplifier provides 180 degrees phase shift, and the "transformer" provides a further 180 degrees. Base bias is provided by R1, and C1 provides decoupling. Depending on the placement of the coils, the oscillator frequency is around 200kHz. In the absence of a 2N3904 for TR1, a BF494 or BC109C may be pressed into service.
The two coils are each made of 50 turns 30swg (0.315mm, or 22awg) enamelled copper wire, wound on a 120mm (4¾") diameter former. Each coil has a Faraday shield, which is connected to 0V as shown. This is essentially a tin or aluminium foil screen, which does not quite make the full circuit of the coil — a gap of 10mm or so is left open. The coils are positioned side by side on the search head, with their beginning (B) wires to the left, and end (E) wires to the right. They are wired to the circuit as shown. The circuit will sustain oscillation with wide variations of coil overlap, and the best degree of overlap may be found through trial and error. The circuit is connected to a Medium Wave radio aerial by means of a screened cable as shown, and a suitable heterodyne is tuned in.
I present the circuit here merely as a bare bones or experimental idea, and look forward to seeing its further development in the future. To give an indication of what the concept is capable of, the Elektor design obtained nearly one-third better performance. I would welcome comments at my e-mail address email@example.com. However, while I would like to reply to all mail, I cannot guarantee that I shall be able. Happy hunting!
Copyright [url=http://thomasscarborough.blogspot.com/]Rev. Thomas Scarborough[/url]