How to verify a gold bar without destroying it

The cheapest verification is buying inside the chain of integrity: a Good Delivery refiner, an LBMA-associated dealer, and metal that has never left professional custody. Metal with unbroken vault history needs no testing, which is precisely what you pay a custody fee for.

Provenance fails at the moment a bar goes into a drawer at home and comes out again. From that point the market treats it as untested, and re-entry usually costs an assay fee or a wider bid. Factor that into any decision to take delivery.

Gold's density of 19.32 g/cm³ is exceptional, and the practical counterfeits — tungsten cores, gold-plated brass, gilded lead — either miss it badly or require precision machining that destroys the economics of the fake at retail sizes.

Measure mass on a scale reading to 0.01 g and volume by water displacement or calibrated calipers. A result within about 0.5% of 19.32 clears the crude fakes. Tungsten at 19.25 g/cm³ is the one metal that survives this test, which is why density is a filter and not a verdict.

• Scale to 0.01 g and calipers to 0.01 mm — total outlay under €60
• Target 19.32 g/cm³ for fine gold; 17.7–17.9 typical for 22ct alloys
• Tungsten passes density; it fails conductivity and ultrasound

An ultrasonic thickness gauge measures sound velocity through the bar — roughly 3,240 m/s in fine gold — and a tungsten or lead core changes it unmistakably. This is the standard non-destructive test for larger bars and the one professional counters rely on.

Eddy-current conductivity testers work on the same principle electrically and are quick on coins. Both are affordable second-hand and both are defeated only by a fake that matches gold on multiple physical properties at once, at which point the counterfeit costs more than the metal.

Handheld X-ray fluorescence is the test buyers trust most and understand least. It reads composition to a depth measured in microns. A 25-micron gold plating over tungsten returns a clean 999.9 result.

XRF is excellent for detecting under-karated alloy and surface adulteration, and useless against core substitution. Pair it with ultrasound or density; never accept it alone on a bar large enough to be worth coring.

Match the serial on the bar to the assay certificate and, where the refiner offers it, to the online verification portal. Modern minted bars ship in tamper-evident packaging with a matching number; a broken seal moves the bar into the untested category even if genuine.

Photograph the bar, the serial and the sealed card on receipt. That five-minute record is what makes a later sale or insurance claim straightforward.

• Verify serial against certificate and refiner portal where available
• Keep assay cards sealed — resale value is measurably lower without them
• Photograph on arrival, before storage

Buy inside custody and skip the ladder. If you take delivery: check seal and serial, run dimensional density, add ultrasound above 100 g, and reserve a professional assay for anything without provenance or for a bar you intend to sell into a wholesale bid.

The failure pattern in retail fraud is almost never a sophisticated bar. It is a plausible seller, a discount to spot, and a buyer who ran no test at all.

The counterfeit risk in practice is not a sophisticated tungsten-cored bar sold by a major refiner’s distributor. It is a private-sale bar with a plausible card, a mis-stamped fractional coin, or a genuine bar bought outside any chain of custody. The overwhelming majority of retail losses trace to the purchase channel, not to a failure of testing.

Which means the highest-value verification step happens before you own anything: buy from the refiner’s authorised distributor or a vault-integrated platform where the bar never leaves the professional chain. Testing is what you do when you could not do that.

A bar that has remained inside LBMA-recognised vaulting from the refinery onward carries the market’s highest liquidity precisely because it has never needed re-testing. Take delivery, and the bar leaves that chain permanently. Selling it back later may require an assay at your cost, and the discount for a bar of unknown history is real.

This is the strongest practical argument for allocated vault storage over home possession for anything above a small holding: not security, but salability. The bar you never touched is the bar you can sell fastest.

• In-chain bar: sells at bid, no assay required
• Delivered bar returning to market: may need assay, may trade at a discount
• Private-sale bar with no documentation: test it, and expect a discount regardless

If dimensional density is off by more than a couple of percent, or a conductivity reading disagrees with the stamp, stop. Do not re-sell the bar onward, and do not attempt to drill or cut it — destroying the bar destroys the evidence and your recourse. Photograph everything, keep the packaging and the assay card, and go back to the seller in writing.

A legitimate dealer will take it back. A refusal, combined with a failed test, is the point at which the matter belongs with your payment provider and, depending on the amount, with the police. Document the sequence as if you will need to show it to someone, because you may.