Build Your Own Metal Detector

[Dzl] and his rather serious looking son are metal detector enthusiasts. But when they couldn’t find their store-bought metal detector earlier this summer they just went ahead and built their own. [Dzl] starts his write up with an explanation of how most oscillator based metal detectors work. This one differs by using an Arduino to read from the metal detecting coil.

The circuit starts with an oscillator that produces a signal of about 160 kHz which is constantly measured by the Arduino. When metal enters the coil it alters the frequency, which is immediately picked up the Arduino. Instead of that characteristic rising tone this rig uses a Piezo buzzer, issuing the type of clicks you’d normally associate with a Geiger counter.

The last part of the build was to find the best coil orientation. They settled on thirty turns around a metal bucket. An old Ikea lamp is the perfect form factor to host their hardware which seems to work like a charm.

35 thoughts on “Build Your Own Metal Detector

  1. Cool beans actually makes a project I’m working on that much easier :)
    Actually two projects didn’t think I would ever get around to the second but now I just might assuming its not too hard to adapt it to an ATtiny.

    1. Seriously, nobody’s falling for your bullshit.

      Congratulations, you get a rise out of people by being a belligerent piece of shit when you post as Eatith Mee, and then you pat yourself on the back with alts. How fucking original.

      Get a life.

  2. One of my new obsessions is building metal detectors. BFO’s are great for “probes” (ie close range detector for searching the larger area scanned by your bigger detector).

    Lots of projects, a great forum and a book are available at:

    Schematics and whole metric tonnes of knowledge are free there! (Check the forum)
    There IS a bit of well, snobiness/touchiness there, but I find that in a lot of niche forums. (ie read lots of articles, search the forum and lurk for awhile before you post)

    1. That was gonna be a project last winter over here. I dove into BFOs, printed out a bunch of stuff bread boarded one that seemed to work and for whatever crackheaded reason, shelved it. I may have to give both this and the BFO another shot. I enjoy my pro model but metal detectors are fun projects and getting things tuned right can be tricky but ya do learn a lot. Thanks for mentioning them :)

  3. Anyone who has some experience with this, perhaps you can tell me if this is feasible?

    my idea would be to use multiple coils ex. 2-8 coils of varying placement and or different sizes/ turns/ shielding then maybe you could ‘image’ an object to varying degrees, for a start I would think using half the turns on a second coil would allow you to get a range for depth

    1. Unless you can map the magnetic fields coming off the object, I don’t think it’s possible. I think you’re thinking of ground penetrating radar, which still cannot image. It’s more like the old fish finders in image quality.

  4. Beat Frequency Oscillator (BFO) designs were used in the early years of portable metal detectors. The problem was those early designs detected all metals, no discrimination to ignore types of metals the operator didn’t want to find.

    Solid state technology made possible VLF (Very Low Frequency), discrimination, zero-motion detection, notch discrimination and signal processing to provide pretty good accuracy at determining what sort of metal(s) and even what kinds of objects were detected.

    BFO technology was left behind in the late 1970’s, even though transistorized models were made for a while. Just imagine lugging around those early portable detectors with vacuum tubes and huge batteries that didn’t run very long due to the power hungry heater filaments.

    What would be interesting is applying current signal processing techniques to a BFO type detector. BFO does have some advantages over VLF, IIRC it can penetrate farther into the ground.

    1. I think ya may have answered your own question Galane :) VLF is a pretty good trade off in terms of depth and discrimination. The problem with super depth comes retrieval of the object. I had a Master Hunter 7 with a Bloodhound dual coil deep seeker. You could hit targets veryyyy deep but then again I had no backhoe to go 12 feet into the ground to dig up what was probably at the end of the day an empty milk jug with axe holes in it where the revenuers had hacked up a moonshine run from a still.
      I think you bring up some very good points and we should take a moment to inform readers that Molecular Frequency Discriminators and Electronic Dowsers are bunkum pseudo science and a waste of money. IIRC I think Lectrasearch had a Turbo Rat Distortion pedal circuit board and the only wired part was a voltmeter for the battery. I include MFDs for their insanely prohibitive cost and because of real-world tests that yield mixed results.
      A funny true story: I got my first basic pro level detector from the classifieds of the paper because the guy had bought it the week before, spent a week digging up trash (only way to get to the good stuff) and peaked in embarassment when he hit a squarish shaped signature on the beach. Crowd gathered around as he dug down nearly 6′ in the sand and was faced with a porcelain container. Opening it he found it to be a toilet tank, stuffed with random parts. (I think both the cavitation and parts let him hit so deep). Everyone but him had a good laugh and I want to say there was even a blurb in the local paper about it lol. Same reason my dad used to call it a “trashfinder” until the day I found a derringer pistol with it. He still called it a trashfinder, but was more hesitant lol.
      Best of luck out there and hope we all get rich :)

  5. hello ^_^ I’ve tried this circuit & it realy works that you can noticed the changing in frequency when there is a metal near to the coil it is sensetive to large metals… but its too noisy ..

  6. Is it possible to differentiate between ferrous and non-ferrous metals or even different elements? I am looking into building a digital prod-type detector for when you are on your knees digging for something you found with your big one. I’m not seeing much evidence that going digital is the best solution.

  7. I’m not a metal detector expert but: a.) the people at are very helpful; b.) An electrical shield as mentioned earlier (make sure not to form a shorted turn) is a common technique.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.