Is My Ring Real Gold? 8 Tests That Work and 4 That Don’t

Gold is easy to imitate across a room and surprisingly hard to imitate in the hand. Brass, plated steel and gold-coloured tungsten all look the part, but each one fails at least one test you can run at a kitchen table: it sticks to a magnet, it leaves the wrong colour on a tile, it rubs through where the band meets the next finger, or it simply weighs the wrong amount for its size. Run the harmless checks first, in order, and most rings sort themselves out before you need a jeweler at all.

Start with what’s stamped inside the band

Clean the inside of the shank, zoom in with a phone camera or a 10× loupe, and read the marks before you test anything. A plain fineness number — 375, 417, 585, 750, 916 — or a karat mark such as 10K, 14K or 18K is a claim of solid gold. Letters attached to the number change the meaning completely: GP, GEP, HGE, RGP, GF or a fraction like 1/20 12K mean a layer of gold over something cheaper. Type what you see into our ring stamp decoder if you’re unsure which kind you have.

A stamp is evidence, not proof. Karat marks are punched by whoever made the ring, and counterfeit “14K” marks are common on plated pieces. Independent hallmarks — struck by an assay office after testing, as in the UK — are much harder to fake. And a ring with no stamp isn’t automatically fake: resizing cuts marks away, decades of polishing wear them flat, and plenty of real gold was made where nothing was stamped. That’s what the tests below are for.

The tests, from harmless to definitive

1. The magnet test

Gold is not magnetic. Hold a strong magnet — a neodymium fridge magnet or the kind sold for finding studs — against the band. If the ring jumps to it, there is iron, steel or nickel inside and it isn’t solid gold. The catch runs the other way: no pull proves very little, because brass, copper, aluminium, silver and tungsten aren’t magnetic either. A clasp, a spring or a steel reinforcement inside a hollow ring can also pull while the gold around it is real. Treat the magnet as a fast way to rule things out, never to rule them in.

2. Look where it wears

Plating fails first at the base of the shank, where the band rubs against the neighbouring fingers, and along sharp edges. Under a loupe, look for a second colour showing through — a brassy, coppery or grey line, or patches where the surface looks dull while the rest shines. Solid gold wears too, but it wears to more gold: the scratches are the same colour as the surface.

3. The skin test

A green line on the finger means copper is reacting with sweat — either a plated ring whose base metal is showing, or a low-karat gold with a lot of copper in it, since 9K and 10K are more than half alloy. A black smudge is less damning than it looks: makeup and sunscreen often contain hard mineral powders that abrade a little real gold off the band and leave a dark mark. Use it as a clue that points you at another test, not as a verdict.

4. The ceramic streak test

Drag the ring firmly across a piece of unglazed ceramic — the rough ring on the underside of a mug or a plate works. Real gold leaves a gold-coloured streak; a black or grey streak means the metal you just abraded isn’t gold. It works because the streak is powdered metal, so surface colour can’t disguise it. Two warnings: it removes a little metal, so use an inconspicuous spot and skip it on anything antique, and a thick plating can pass if you only rub lightly — the streak has to go through the surface to mean anything.

5. Weigh it in air and in water

Gold is extraordinarily dense: pure gold is 19.3 times heavier than the same volume of water, while brass is about 8.5. That gap is why a plated brass ring feels like a toy once you’ve held the real thing — and it can be measured with a kitchen version of the jeweler’s method. Weigh the ring dry. Then stand a cup of water on the scale, zero it, and lower the ring on a thread until it hangs fully under water without touching the sides or the bottom. The scale now shows the weight of the water the ring pushes aside, which in grams equals its volume in cubic centimetres. Density is the dry weight divided by that reading.

Ranges are typical published values; white golds with palladium sit at the heavy end of their band, nickel white golds at the light end. A ring set with stones reads low, because diamonds are only 3.5 g/cm³ — test plain bands, or expect the number to be dragged down.

The weakness is precision. A 5-gram 14K ring displaces under 0.4 g of water, so a kitchen scale that reads to 0.1 g can’t tell 14K from brass; you need a jewelry scale that reads to 0.01 g, and the heavier the ring, the better. The calculator shows the range your scale can actually support. Note also where tungsten carbide sits: right on top of 18K. Tungsten is the metal used to fake gold bars for exactly that reason, so a “perfect” density on a suspiciously hard, unscratchable ring isn’t the end of the story.

6. The acid test

The jeweler’s classic. A small scratch on a black touchstone, then a drop of acid matched to a karat: if the streak survives the 14K acid but dissolves in the 18K acid, the gold sits between the two. Kits cost little and are accurate in practised hands, but the test needs a scratch and uses nitric acid. Fine on scrap, wrong on anything you mean to keep — and a jeweler will do it in front of you for nothing if you’re thinking of selling.

7. An electronic gold tester

Many jewelers and gold buyers use a handheld tester that reads the electrical conductivity of the surface through a gel. It’s quick, it doesn’t mark the ring, and it gives a karat reading. Because it only reads the surface, a heavy plating can fool it, which is why a buyer who doubts the result will still file a notch and test underneath.

8. XRF — the answer, not a guess

An X-ray fluorescence analyser reads the elements in the metal in a few seconds without touching or marking it, and prints the alloy: 58.4% gold, so much silver, so much copper. Many gold buyers and a growing number of jewelers have one, and it’s the test to ask for before you sell, insure or buy anything expensive. Its one blind spot is depth: it reads the outer layer, so thick gold-filled metal can occasionally pass. The only fully conclusive method is a fire assay — which melts a sample and is used by refiners, not on rings anyone wants back.

Four “tests” that prove nothing

• The bite test. Gold is soft, but so are lead and many plated alloys, and modern 14K is too hard to dent with your teeth anyway. The only thing it reliably tests is your dentist.
• The float test. Every metal ring sinks. Floating says nothing about gold.
• The vinegar test. Gold doesn’t react to vinegar — but neither do lacquered brass, stainless steel or a good plating. A ring that darkens is probably not gold; a ring that doesn’t is not proven to be.
• The flame test. Gold doesn’t blacken in a lighter flame, but heat can crack or cloud stones, damage plating and loosen solder. You learn little and risk a lot.

Putting the results together

• 14K or 750 stamp, no magnetic pull, gold streak, no wear-through — very likely solid gold. Confirm with XRF before a sale or insurance.
• Karat number with GP, HGE, RGP or GF beside it — plated or filled. The number describes only the gold layer.
• Strong magnetic pull — not solid gold, whatever the stamp says.
• No stamp, no pull, heavy for its size — promising and inconclusive. Get an electronic or XRF test before you form an opinion.
• 925 on a gold-coloured ring — gold-plated sterling silver (vermeil). Real silver, plated gold.

Do first

Read the stamp, then the magnet

Rules it out

Magnetic pull, black streak, base metal showing through

Rules it in

XRF, acid or an electronic tester — not a home test

Gold density

14K ≈ 13, 18K ≈ 15.5, 24K = 19.3 g/cm³

Watch out for

Tungsten: as dense as 18K, and very hard

Skip

Biting, floating, vinegar and flame

The takeaway

Home tests are good at one thing: catching fakes. A magnet, a tile, a loupe and a 0.01 g scale will expose most plated and base-metal rings in minutes. What none of them can do is prove purity — that takes an acid, an electronic tester or, best of all, an XRF reading, all of which a jeweler or gold buyer can run in front of you. Once you know the karat, the gold value calculator turns grams into money, and what is my ring worth explains why the offer you get won’t be the number on the screen.

At a glance

Gold tests compared

Test	What it tells you	Risk to the ring
Stamp	The claimed metal and karat; plating marks	None
Magnet	A strong pull rules gold out; no pull proves little	None
Wear check	Plating rubbing through on edges and the base of the band	None
Ceramic streak	Black or grey streak rules gold out	Removes a trace of metal
Density (dry weight ÷ water displaced)	Narrows the karat; tungsten can mimic 18K	None
Acid test	Karat range	Small scratch, acid
Electronic tester	Karat of the surface	None
XRF	Exact alloy of the surface layer	None

Frequently asked questions

Does a magnet stick to real gold?

No. Gold is not magnetic, so a strong pull means steel, iron or nickel inside. But no pull doesn’t prove gold — brass, copper, silver and tungsten aren’t magnetic either.

Is the vinegar test reliable?

No. Gold doesn’t react to vinegar, but neither do lacquered brass, stainless steel or a good plating. It can rule some fakes out, never gold in.

Can a ring be real gold without a stamp?

Yes. Resizing, polishing and some countries and periods leave real gold unmarked. A jeweler can test it with acid, an electronic tester or XRF.

What does 14K HGE mean?

Heavy gold electroplate: a thin layer of 14 karat gold over base metal. It is not solid gold and has almost no melt value.

How accurate is the water density test?

Only as accurate as your scale. With a 0.01 g jewelry scale and a ring of five grams or more it separates brass from gold and narrows the karat; a 0.1 g kitchen scale can’t.

Sources and further reading

1. 16 CFR Part 23 — Guides for the Jewelry, Precious Metals, and Pewter Industries — U.S. Electronic Code of Federal Regulations
2. Karat (purity) — Wikipedia
3. Gold-filled jewelry — Wikipedia
4. Streak (mineralogy) — Wikipedia
5. X-ray fluorescence — Wikipedia
6. Fire assay — Wikipedia

Links checked October 4, 2026.