Before you trust a chain — or a buyer — run the free checks first. Here are 7 at-home tests for real gold, ranked by honesty, with each one's real limits.
You’re turning it over in your hand — a chain from a flea market, a coin a relative swore was solid, a ring that looks a little too shiny for its price. The stamp says one thing; your gut says maybe. So how do you tell if gold is real without handing it to a stranger and hoping they’re honest?
Here’s the reassuring part: you can rule out most fakes at your kitchen table for free, and get close to certain for the price of a cheap test kit. The important part is knowing what each test actually proves — because several popular “tests” are far weaker than the internet claims, and one of them can quietly damage a piece you’d rather not scratch. We’ll rank the seven at-home tests from free and harmless to paid and mildly destructive, tell you the limits of each, and end with when it’s worth paying a professional.
Start with the free checks
Before you buy anything, spend five minutes with what you already have.
Test 1 — Read the hallmark stamp
Almost all real gold jewelry carries a tiny purity mark, usually on a ring’s inner band, a chain’s clasp, or an earring post. Solid gold reads as a karat or fineness number:
| Stamp | Meaning | Solid gold? |
|---|---|---|
| 24K / 999 | 99.9% pure | Yes |
| 18K / 750 | 75% pure | Yes |
| 14K / 585 | 58.5% pure | Yes |
| 10K / 417 | 41.7% pure | Yes |
| GF / GP / HGE / RGP | Gold-filled or gold-plated | No — thin layer over base metal |
That second group is the trap. GF (gold-filled), GP (gold-plated), HGE (heavy gold electroplate) and RGP (rolled gold plate) all mean a whisper-thin layer of gold bonded over cheap metal — with almost no melt value, no matter how convincing the shine. A “14K GF” stamp is gold-filled, not solid 14K. If you want the full decoder for every mark, including European and international ones, read what gold hallmarks really mean.
The stamp is often worn smooth or hidden. A little magnification earns its keep here — and this is where I’ll be upfront that the product links below are affiliate links.
A 30× jeweler’s loupe with a built-in light turns a faint, worn 585 into something readable in seconds, and costs less than a takeaway lunch.
Test 2 — The magnet test
This one is free, fast, and genuinely useful — as long as you know what it can and can’t say. Solid gold is not magnetic. Hold a strong magnet near the piece: if it jumps or pulls firmly toward the magnet, there’s magnetic base metal (usually iron or steel) inside, and it is not solid gold.
The catch is the magnet. A fridge magnet is far too weak to reveal anything. You want a proper rare-earth (neodymium) magnet — the small, startlingly strong kind — held right against the piece.
Test 3 — Look for skin and wear discoloration
Turn the piece to the light and look at the high-wear points — the underside of a ring, the edges of a clasp. On plated jewelry, the gold layer rubs through over time and a different colored metal peeks out underneath. And if the wearer’s skin went green or black where the piece sat, that’s often a cheap alloy or plating reacting with sweat.
Be honest with yourself about this one, though: it’s a hint, not a test. Solid higher-karat gold usually won’t discolor skin, but some people’s skin chemistry reacts even with the real thing, and a well-made plated piece may show no wear at all. Use discoloration as a reason to test properly — never as your final answer.
The paid tests: kits and testers
If the free checks leave you unsure, a small investment gets you much closer to certainty. Free advice first, always — but these are the tools that actually earn their place in a drawer.
Test 4 — The acid (scratch) test
For decades this was the jeweler’s standard, and it’s still one of the most reliable things you can do at home. A gold acid test kit comes with a black testing stone and small bottles of acid calibrated to different karats (10K, 14K, 18K, 22K).
Here’s how it works: you rub the piece firmly on the stone to leave a short streak of metal, then place a drop of the karat-matched acid on that streak. Real gold at or above that karat holds its color; lower-karat or fake metal fades, dissolves, or turns green or brown. Start with the acid matching the stamp — or with 14K if there’s no stamp — and work up or down from the reaction. Done carefully, acid testing is highly reliable at separating real gold from fakes and pinning down the karat.
Test 5 — Electronic (conductivity) testers
An electronic gold tester reads how the metal conducts electricity and shows an estimated karat on a small display. It’s quick, needs no acid, and — used within its limits — doesn’t mark the piece.
The limit is important: these testers only read the surface. A base-metal piece with a thick gold plating, or a gold shell over a tungsten core, can read as solid gold because the probe never “sees” past the outer layer. They’re reasonably dependable on solid, unplated jewelry and useless against a well-made plated fake. Think of an electronic tester as a fast second opinion alongside the magnet and the stamp — not a verdict on its own.
Test 6 — The ping (ring) test, for coins
If you’re checking a coin or a bullion round rather than jewelry, sound is a surprisingly good tell — and it costs nothing. Balance the coin on a fingertip and tap it gently with another coin. Real gold gives a warm, sustained ring that lasts a second or two; a base-metal fake tends to sound dull and short.
You don’t have to trust your ear alone. Free phone apps record the tap and analyze its frequency against known-good coins, and — usefully — a tungsten-cored fake rings very differently from real gold even though it weighs almost the same. Do it in a quiet room. The honest limits: it works best on loose coins (bars and slabbed, encapsulated coins don’t ring cleanly), and a counterfeit struck from the correct alloy will ring true, so the ping test checks the metal, not whether the coin’s design is genuine.
Test 7 — The density (water) test
This is the most underrated at-home test: non-destructive, needs no special kit, and hard to fool. It works because gold is remarkably dense — about 19.3 grams per cubic centimeter, far denser than the cheap metals used to fake it.
You need a scale that reads to 0.01 g and a way to weigh the piece while it’s submerged. The method:
- Weigh the piece dry, in air. Call that W(air).
- Weigh it again fully submerged in water (suspended by a thin thread so it doesn’t touch the container). Call that W(water).
- Divide: density = W(air) ÷ (W(air) − W(water)).
A result near 19.3 g/cm³ points to high-purity gold; lower karats land lower (18K sits around 15.5, 14K around 13). If your number comes out near brass or silver territory, you’ve caught a fake. The one thing density can’t catch is the rare tungsten-core fake — tungsten’s density is nearly identical to gold — which is exactly why coin buyers reach for the ping test too. (Any 0.01 g jewelry scale works; if you read how much your gold is worth, you already know why an accurate scale is worth owning.)
The seven tests at a glance
No single test is the whole answer — the smart move is to stack a few. Here’s how they compare, from free to paid:
| Test | Cost | Damages the piece? | What it actually proves |
|---|---|---|---|
| Hallmark stamp | Free | No | What the piece claims to be |
| Magnet | Free | No | Rules out magnetic fakes only |
| Skin / wear discoloration | Free | No | A hint of plating, nothing firmer |
| Acid (scratch) | Low | Yes, mildly | Real gold vs. fake, and the karat |
| Electronic tester | Medium | No | Surface karat on solid, unplated pieces |
| Ping (ring) test | Free | No | Coins/bullion: catches base-metal & tungsten cores |
| Density (water) | Free–low | No | Solid gold vs. most fakes (not tungsten) |
Read that table as a recipe, not a menu: the stamp and magnet together take seconds and catch the obvious fakes; a density test or an acid test settles most of what’s left; and anything still doubtful goes to a professional.
When to stop testing and pay a professional
There’s a point where guessing stops being smart. If a piece is heavy, valuable, unmarked, or you simply can’t get consistent answers, take it to a jeweler or gold buyer with an XRF analyzer — an X-ray fluorescence scanner that reads the metal’s exact composition in seconds without a scratch. Most gold buyers have one, and many will scan a piece for free while quoting you.
The math is simple: if a wrong guess could cost you real money, a professional test is cheap insurance. Pushing a $20 acid kit is fine; pretending a magnet settles the value of an inherited coin collection is not.
You’ve confirmed it’s real — now what’s it worth?
Knowing gold is genuine is only half the question. The other half is what it’s actually worth, and that’s just arithmetic: the weight in grams, times the purity (0.585 for 14K, 0.75 for 18K), times the live gold price per gram. That’s the melt value — the ceiling — and a buyer pays a percentage of it.
Don’t do the multiplication in your head under pressure at a counter. Our gold melt-value calculator pulls the live price and does it for you, and the gold price page shows where the market sits today. If you’d like the full walkthrough — reading the stamp, weighing correctly, and what different buyers actually pay — the companion guide on how much your gold jewelry is worth covers it end to end, and our step-by-step selling guide shows how to compare offers without getting talked down.
So — how do you tell if gold is real? Read the stamp, test with a magnet, and if you still aren’t sure, reach for an acid kit or a density test before you trust anyone else’s word. Rule out the fakes first; then let the numbers tell you what the real thing is worth.
Common questions
How can I tell if gold is real at home?
Start with the free checks: read the hallmark stamp (585, 750, 14K, 18K mean solid gold; GF, GP, HGE, RGP mean plated or filled), then hold a strong magnet to the piece — solid gold is not magnetic, so a firm pull means there is base metal inside. Neither is conclusive on its own. To go further, an acid test kit or an electronic tester narrows it down, and a density (water) test is a good non-destructive check. Nothing at home is as certain as a jeweler's XRF scan.
Is gold magnetic?
No. Solid gold is not attracted to a magnet. If a piece pulls noticeably toward a strong rare-earth magnet, it contains magnetic base metal and is not solid gold. But the reverse is not proof: many fakes — brass, copper, and tungsten among them — are also non-magnetic, so passing the magnet test only rules out one kind of fake.
Does the acid test damage the jewelry?
Mildly, yes. The standard method rubs the piece against a testing stone to leave a streak, and applying acid can etch or discolor the spot if it touches the item directly. On an inconspicuous area the mark is tiny, but it is not fully reversible. For a signed, antique, or otherwise valuable piece, use a non-destructive test — or a jeweler — instead.
Are electronic gold testers accurate?
They are useful but limited. Electronic testers read the surface conductivity of the metal, so a base-metal piece with a thick gold plating — or a tungsten core — can read as solid gold. They are reasonably reliable on solid, unplated items and don't damage the piece, but they cannot tell you what the metal underneath actually is.
What is the most reliable way to confirm gold is real?
A professional XRF (X-ray fluorescence) analyzer, which reads the metal's exact composition in seconds without damaging it. Most gold buyers and many jewelers have one. For high-value bars, pros pair XRF with a density check, because XRF reads only near the surface. If real money is on the line, pay for a professional test before you sell.
Does real gold change color or leave a mark on your skin?
Usually not, but this is a weak signal, not a test. Cheap alloys and plated pieces can react with sweat and leave a green or black mark, while solid higher-karat gold generally doesn't. Some people's skin chemistry reacts even with real gold, though, so treat discoloration as a hint to test properly — never as proof either way.