The History of the Metal Detector: From Invention to Today

Disclosure: Some links in this article are affiliate links. We may earn a commission on qualifying purchases at no extra cost to you.

Ask most hobbyists when the metal detector was invented and you'll usually get a single name: Alexander Graham Bell. It's an understandable shortcut, but it misses the older, stranger, and more useful story behind the discrimination and target ID features on the machine in your trunk. The actual timeline stretches from a Parisian engineer probing for a bullet to a Polish lieutenant clearing minefields in the desert, and from vacuum-tube physics to the wireless, multi-frequency detectors sold today.

That history matters because nearly every feature on a modern metal detector is a direct answer to a failure someone encountered along the way. Ground balance exists because an induction balance couldn't see a bullet through bedsprings. Discrimination exists because early beach and park machines couldn't tell a nail from a coin. The story is not just a list of names and dates — it's a map of why your detector works the way it does.

This guide traces that corrected timeline from 1874 to the present, with primary patent numbers and the specific technical lineage that most short histories skip. You'll see the real inventors, the myths that refuse to die, and how each breakthrough changed the user experience on the ground.

It's written for hobbyists who want more than the usual "Bell invented it" paragraph: you'll get the version that helps you understand frequency choice, target ID, pulse induction, and why a single modern machine can now handle saltwater, parks, and gold country.

Quick Reference Timeline

The Real Invention Story: Trouvé, Not Bell

The earliest documented use of electromagnetic induction to find metal inside a living person belongs to Gustave Trouvé, a Parisian electrical engineer. In 1874, he built a hand-held probe intended to locate and assist in extracting bullets from patients. It was a medical device rather than a hobby machine, but the operating principle — a coil generating an electromagnetic field and responding to conductive metal — is the same one used by every VLF detector today.

Trouvé's device gets less attention because it never became a commercial product and because the dramatic Bell-Garfield story quickly overshadowed it. Still, his 1874 work is the earliest credible starting point when you ask who first turned induction into a practical metal locator.

The more common answer, Alexander Graham Bell, actually came seven years later. What Bell attempted was real, historically documented, and technically important — but calling him the inventor is wrong on three counts: Trouvé preceded him, Bell adapted an existing four-coil induction balance design, and his device never found the bullet it was built to locate.

Bell, Garfield, and the Induction Balance Failure

On July 2, 1881, President James A. Garfield was shot at the Baltimore & Potomac Railroad Station in Washington by Charles Guiteau. The bullet lodged somewhere in his torso, and his doctors could not locate it. Bell, already famous for the telephone, volunteered to build a device based on the induction balance principle originally developed by David E. Hughes.

The instrument was a four-coil induction balance: two transmitter coils and two receiver coils arranged to produce a null, with the null disturbed when a conductive object entered the field. Bell attempted to locate the bullet on July 26 and August 1, 1881. Both attempts failed.

The failure was not simply a technical weakness. The bed contained metal bedsprings, which overwhelmed the instrument, and the doctors restricted the search to Garfield's right side while the bullet was most likely on his left. Garfield died on September 19, 1881, of infection and septicemia — not from the bullet extraction. The correct death date matters because many casual sources still cite September 12.

Bell published his account in 1882 before the American Association for the Advancement of Science in Montreal under the title "Upon the Electrical Experiments to Determine the Location of the Bullet… and upon a Successful Form of Induction Balance." That paper is a primary source, and it clearly describes an adapted device, not an original invention. The bedspring failure, in particular, became a foundational lesson for later detector design: any practical machine had to cope with metal and mineral interference in the surrounding environment.

The Patent Era: Herr, Fisher, and the First Commercial Detectors

The first US patent for a hand-held hidden-metal detector was filed by Shirl Herr of Crawfordsville, Indiana, on February 4, 1924, under Serial No. 690,360. The "Hidden-Metal Detector" patent, US 1,679,339, was granted July 31, 1928. This distinction between filing and grant dates is important because short articles often compress the two, but the primary record is unambiguous.

Herr's device was reported as effective to a depth of eight feet and was later used, according to historical accounts, on Admiral Richard Byrd's Second Antarctic Expedition in 1933. He also assisted in recovering artifacts from Caligula's galleys at Lake Nemi in Italy in late August 1929. Those applications show the early hand-held detector as a tool for engineering, archaeology, and exploration long before it became a hobby product.

Gerhard R. Fisher, an electrical engineer who had studied in Dresden and worked on aircraft radio direction finding, filed his "Metalloscope" patent on January 16, 1933, under Serial No. 651,974, after a renewal filed August 6, 1936. The patent, US 2,066,561, was granted January 5, 1937. Fisher founded Fisher Research Laboratory in 1931 in a home garage in Palo Alto, California, and his device became known commercially as the M-Scope.

The M-Scope is widely regarded by collectors and hobby historians as the true birth of the commercial metal detector. It was used by geologists, utilities, lumber mills, law enforcement, and treasure hunters. Fisher served as company president until retiring in 1967, and the firm he founded remains part of the modern hobby's lineage.

One persistent error deserves correction: there was no Fisher metal-detector patent in 1925. Fisher's 1920s patents covered aircraft radio direction finding, not metal detection. His metal-detector patent was filed in 1933 and granted in 1937. Whenever you see a "1925 Fisher patent" reference, the source is repeating a myth.

WWII and the Polish Mine Detector

The next major leap came not from a laboratory but from a battlefield. In 1941, Józef Kosacki, a Polish Army lieutenant stationed in Scotland, designed the Polish Mine Detector Mark I. The device used two coils on a bamboo pole — an oscillator coil and a receiver coil feeding an amplifier and headphones — and weighed about 14 kg (31 lb). A single operator could carry and use it.

The design was simple, robust, and immediately effective. Five hundred units were sent to the Second Battle of El Alamein in 1942, where they doubled the speed at which heavily mined sand could be cleared, from roughly 100 to 200 metres per hour. More than 100,000 units of Marks I through IV were produced, and British forces used variants into the 1990s.

Kosacki did not patent his mine detector. He gave the design to the British Army without charge and received a letter of thanks from the King. That detail matters because it corrects another common error: the Polish Mine Detector was a gifted military instrument, not a commercial patent, and it established the basic transmit-and-receive coil architecture still used by most detectors today.

The Transistor, Garrett, and the Zero Drift Revolution

Early detectors depended on vacuum tubes and were bulky, fragile, and power-hungry. The invention of the transistor by John Bardeen, Walter Brattain, and William Shockley in 1947 changed that. By the 1950s and 1960s, smaller battery-powered detectors became practical for hobbyists and security personnel.

Charles Garrett founded Garrett Metal Detectors in Garland, Texas, in 1964. His company's first major contribution was solving a problem that plagued early machines: oscillator drift. A detector that could not hold a stable frequency would produce false signals as the circuit warmed up. Garrett's late-1960s coaxial search-coil winding eliminated much of this drift, a period the company calls the "Zero Drift Revolution."

That stability was the foundation for reliable discrimination. Without a stable base frequency, it was nearly impossible to compare phase shifts consistently enough to distinguish one metal from another. The zero-drift coil made the next era possible.

VLF and TR: Discrimination Becomes Usable

Garrett introduced its first TR (transmitter-receiver) detector in 1973 and the GroundHog VLF/TR in 1977. Garrett also coined the term "VLF," for very low frequency, and the label stuck across the entire industry.

VLF works by transmitting a continuous signal from one coil and receiving it in a second coil. The phase shift between the transmitted and received signals varies with the conductivity of the target. Lower frequencies favor high conductors such as silver and copper, while higher frequencies respond more strongly to lower conductors such as gold and lead. That phase-shift discrimination is why your detector can assign a target ID number and why frequency choice still matters for coins versus small gold.

TR discrimination had existed earlier, but early BFO and TR machines suffered from depth loss and were unpopular with practical hunters. VLF improved stability and allowed discrimination to become a mainstream feature rather than a novelty. The trade-off between accepting more targets and losing depth, still familiar to anyone who runs aggressive discrimination settings, traces directly to this era.

Pulse Induction and the Gold Boom

A fundamentally different approach, pulse induction, emerged from military and industrial work long before it reached hobbyists. PI transmits brief high-current pulses into the ground and measures the decay time of the resulting field. The method is largely immune to ground mineralization and saltwater, which makes it ideal for mineralized goldfields and wet salt beaches.

But PI gives up something in return: it offers weak or no discrimination. That trade-off — depth and stability in bad ground, but little target information — remains the core reason PI machines are specialized tools rather than general-purpose detectors.

The breakthrough that brought PI to the mass market was the Minelab SD2000, released in 1995. It is widely credited as the first true consumer pulse-induction prospecting detector, and it transformed Australian goldfields during the 1990s. Minelab's later GPX series descends directly from that line. For gold prospectors working heavily mineralized soil, the SD2000 and its successors made previously unmaskable ground huntable.

Digital Target ID and the Microprocessor Era

The late 1980s brought microprocessors into detectors, and with them came visual target identification. White's Electronics introduced the Eagle in 1987, featuring VDI — Visual Display Indicator — that displayed a numeric target ID. The Eagle Spectrum followed in 1991 and became a benchmark for coin and relic hunters. Collectors still run these machines today, and early VDI units remain sought after for their build quality and repairability.

Garrett's parallel development arrived with the Grand Master Hunter in 1988, the company's first microprocessor detector. The GTA in 1990 added graphic target analysis and notch discrimination, allowing users to accept or reject specific conductivity bands rather than relying on a blanket discrimination threshold.

Digital target ID did not eliminate the old lessons. Soil mineralization, depth, target shape, and nearby trash still shift ID numbers and tones. The practical skill of reading a repeatable ID across multiple sweep angles, rather than chasing a single high number, became more important with the new displays. But the microprocessor era made target information visible in a way that fundamentally changed how hobbyists hunted.

Multi-Frequency, Wireless, and the Modern Machine

The next two revolutions happened on parallel tracks. The first was multi-frequency technology. Minelab's BBS (Broad Band Spectrum) appeared in the Sovereign in the early 1990s, transmitting 17 frequencies from 1.5 to 25.5 kHz sequentially. FBS (Full Band Spectrum) expanded that to 28 frequencies from 1.5 to 100 kHz. Then in 2017, the Equinox series introduced Multi-IQ, which transmits multiple frequencies simultaneously rather than sequentially.

The distinction matters: simultaneous multi-frequency lets a single machine process the responses from several frequencies at once, which stabilizes target IDs across different soil and target types. That is why one modern machine can now do reasonably well in saltwater, parks, and gold fields — a range that previously required two or three dedicated detectors.

The second revolution was wireless. XP released the Deus in 2008 as the world's first fully wireless metal detector, linking the coil, remote, and headphones over a digital radio protocol. The Deus II continued that architecture with Fast Multi-Frequency, combining simultaneous multi-frequency with the same cable-free design. Check the XP Deus on Amazon. For a deeper look at the wireless platform, see our XP Deus review.

The combination of simultaneous multi-frequency and wireless ergonomics defines the current flagship class. It also explains why single-frequency VLF remains relevant: mono modes can be more stable in high-EMI environments, and some gold prospectors still prefer manual-ground-balance PI machines over newer digital GPX models. The old tools did not disappear — they found their place inside a broader ecosystem.

Why This History Explains Today's Features

The best reason to learn this timeline is that every major feature on a modern detector is a direct answer to an earlier failure.

Bell's bedspring failure is the ancestor of ground balance and iron discrimination. When your detector compensates for mineralized soil or lets you reject ferrous targets, you are using the solution to a problem that defeated an induction balance in 1881.

Kosacki's bamboo-pole mine detector established the transmit-and-receive coil architecture. Modern coils are smaller and lighter, but the basic two-coil configuration on a shaft is the same idea that cleared El Alamein.

The 1970s VLF phase-shift discrimination is why your detector shows a target ID number and why you choose a higher frequency for small gold or a lower frequency for silver and copper. It is the direct result of Garrett's zero-drift stability work and the TR-to-VLF transition.

Pulse induction's immunity to mineralization is why gold prospectors and wet-salt beach hunters still buy PI machines even though they give up discrimination. The SD2000 lineage explains the modern GPX series.

Multi-frequency, first sequential and now simultaneous, is why one machine can handle saltwater, parks, and gold country. That capability is not magic; it is the culmination of more than a century of solving interference, stability, and discrimination problems in sequence. For a practical look at how target ID and discrimination settings operate on current machines, see Metal Detector Discrimination & Target ID Explained, and for the broader Garrett lineup that came out of the zero-drift era, see Garrett Metal Detectors: Complete Brand Guide.

Myths Debunked

Four errors show up so often in metal detector histories that they are worth correcting directly.

1. Alexander Graham Bell invented the metal detector.
He did not. Bell adapted David E. Hughes's four-coil induction balance, and his attempt came seven years after Gustave Trouvé's 1874 bullet-probe. Bell's device also never located Garfield's bullet.

2. Gerhard Fisher received a metal-detector patent in 1925.
There was no Fisher metal-detector patent in 1925. Fisher's 1920s patents covered aircraft radio direction finding. His metal-detector patent, US 2,066,561, was filed January 16, 1933, and granted January 5, 1937.

3. Józef Kosacki patented the Polish Mine Detector.
He did not. Kosacki gave the design to the British Army without charge and received a letter of thanks from the King. More than 100,000 units were produced, but no patent protected the design.

4. President Garfield died on September 12, 1881.
The correct date is September 19, 1881. He was shot on July 2 and died 79 days later of infection and septicemia. The September 12 date appears only in low-quality secondary sources.

Frequently Asked Questions

Conclusion

The real history of the metal detector is a chain of specific problems solved by specific people: Trouvé's medical probe, Bell's failed induction balance on Garfield, Herr's first patent, Fisher's commercial M-Scope, Kosacki's mine detector, Garrett's stable VLF discrimination, Minelab's SD2000 pulse induction, White's digital target ID, and the modern combination of simultaneous multi-frequency and wireless ergonomics.

Every modern feature you rely on — ground balance, discrimination, target ID, multi-frequency, wireless audio — is the answer to a challenge that defeated someone earlier in this timeline. Once you see that lineage, the machine in your hand stops looking like a black box and starts looking like a century of accumulated fieldwork.

If you want to apply that history to a purchasing decision, start with our discrimination and target ID guide to understand how modern machines classify targets, then look at the Garrett brand guide for one mainstream lineage or the XP Deus review for the wireless side of the story. The right detector for you will depend on your ground, your targets, and the specific problems you want that long history to solve.