The "Blind Squirrel"
Pulsed Induction Metal Detector

This is my version of a Pulsed Induction type of Metal Detector.  I named it "Blind Squirrel" because even a blind squirrel finds an acorn every now and then.

This project came about when it was suggested by our club for the 2005-2006 winter contest.

After studying the various types of metal detectors and circuits that were available on the Web, I decided to design my own using ideas from previous designs as well as my own ideas.

The Pulsed Induction (PI) detectors seem to have the advantage of more depth, at least that's what most folks were saying, so I set out to design another PI detector for the contest.

Most of the PI detectors work the same way:

1.  A high current, short duration pulse is repeatedly applied to the search coil at a fixed frequency which is provided by a timing circuit.

2.  Either another coil is used as the receiver or, most commonly, the same coil is used as both transmitter and receiver.

3.  Immediately after the pulse is applied, the receiver coil detects a faint signal which is produced when the magnetic field in the coil collapses.  This collapse or decay is lengthen in duration when the search coil is near metal objects due to the eddy currents they produce.

4.  This weak received signal must be amplified many times in order to detect the very small changes in the decay length.  Since there is also a very large kick-back pulse present, the weak decay signal must be separated with a sampling circuit which is gated by the same timing circuit.

5.  Once the received signal is isolated, it is further amplified and converted to a relative voltage which is in proportion to the size and/or proximity of the metal being detected.

6.  This voltage is then used to drive an audio and/or visual indicator such as a meter and speaker.

7.  A source of regulated power is necessary because of the high gains and offsets that are involved.

One of the hardest parts of this design is amplifying the weak decay signal when the kick-back pulse is present.  In just about all designs I have seen, an op amp with a high slew rate and wide bandwidth is used.  Because of the big offset at the input, two power supplies, plus and minus, are needed on the op amp IC.  This is usually accomplished by using a voltage inverter IC to produce the negative voltage or sometimes a booster is used with a floating ground reference in order to bias the IC correctly.

One of my goals was to eliminate the extra supply.  I wanted my detector to operate on a single supply with a common ground connection.

Another goal was to simplify the timing circuit.  Here, you want to have a stable circuit as far as frequency and pulse width and it must be simple.  Some of the published circuits simply use a R/C oscillator with a trim pot to set the frequency.  Some use a microprocessor which makes it simple and stable, but hard to change frequencies.  Being able to change frequencies makes it easy to experiment with different search coils and timing schemes.

I believe I have met the goals with the following design, however, I want to emphasize the fact that this is still experimental and I can't guarantee the results you may have.  I am presenting it here as an alternative to the circuits that are currently available.  I also encourage further refinement.  Please play with it and see if you can suggest improvements or refinements.  Please share them with me so I can post them here.  I would really like to see pictures of your detector too.

(Click on picture for more information)

Schematic and Parts List                                                                                      

Circuit Board Layout

Parts Placement

Waveforms and Adjustments

Pictures of the Completed Detector

Click here to return to the MODERN HOMEBREW page

© 2006 ~ Michael R. Starcher ~ All Rights Reserved