The most reliable checks are density, ultrasound and X-ray fluorescence — everything before that is a first sift. Modern fakes with a tungsten core pass weight and dimensions, but fail on the speed of sound through the material. Buying from accredited dealers moves that check upstream.
The classic fake was a gold-plated brass bar — easy to spot, because it is too light or too large. The problem is the new generation: tungsten has a density of 19.25 g/cm³, deceptively close to gold's 19.32. A bar with a tungsten core and a gold shell has the correct weight, the correct dimensions, and passes the magnet test too.
That is why the order matters: dimensions and weight sort out the crude fakes. If you want certainty, you then need a method that looks inside.
It works on Archimedes' principle: weigh the bar dry, then submerged in water. The difference gives you the volume, and mass divided by volume gives the density. For fine gold the result must come out close to 19.32 g/cm³.
The test needs only precision scales and a glass of water, and it rules out practically every material except tungsten and platinum. With a tungsten core the deviation is a few tenths — measurable only with very accurate scales and careful technique.
On coins, density measurement is less precise, because the strike and the edge structure distort the volume. Here, dimensions and the ping test together tell you more.
A probe sends a sound pulse through the material and measures the time of flight. Sound travels through gold at around 3,240 m/s and through tungsten at about 5,180 m/s. A core of foreign material shows up immediately, however well it is clad. This is the most reliable non-destructive method for bars.
The instrument excites the surface and reads the elemental composition from the emission. It gives exact fineness values — but only for the outermost micrometres. A thickly gold-plated tungsten bar can pass an XRF check. Only the combination of XRF and ultrasound is effectively counterfeit-proof.
Every method above costs time, equipment and practice. The pragmatic approach is to move the check to where it happens anyway: to the goods-in desk of a dealer who sources exclusively from LBMA-accredited refineries.
Look for three things when buying: a stamp showing the fineness and the maker's mark, a serial number, and — on units from 100 grams upwards — a sealed assay card. Never buy loose bars without documentation from private sellers; that is where practically all the counterfeit risk sits.
Check the dimensions and weight first, then the density: fine gold sits at 19.32 g/cm³. Only an ultrasound test, which measures the time of flight of sound through the material, gives certainty against fakes with a tungsten core. X-ray fluorescence determines fineness, but only at the surface.
No. Pure gold is not ferromagnetic. A bar that sticks to a neodymium magnet contains iron or nickel and is not fine gold. Conversely, a negative magnet test proves nothing, because tungsten is not magnetic either.
A sealed pack with a certificate of authenticity, in which the bar is encased along with its serial number, fineness and the refinery's assay mark. It makes resale considerably easier and should not be opened.
With a dealer who sources exclusively from LBMA-accredited refineries and state mints, counterfeit risk is effectively ruled out. The risk lies in buying loose, undocumented stock from private sellers.
Physical values follow standard reference literature. The methods described do not replace professional materials testing. If you suspect a problem, contact a refinery or an expert assessor.