Buying guide
How to Tell If Gold Is Real: Stamp, Magnet, Weight and Lab Tests
Check the karat stamp first, then use a magnet and the piece's weight as quick screens. None of those is proof, because fakes can carry a fake stamp. The only certain answer is a jeweler's test: an XRF analyzer, which reads the metal without damage, or an acid test on a touchstone.
Solid 10K gold chains, weighed and stamped
See allThe order to check in
Work from free and quick to certain. Each step can rule a piece out; only the last one rules it in.
| Check | Cost | Can it prove fake | Can it prove real |
|---|---|---|---|
| 1. Read the stamp | Free | Sometimes (GP, GF, HGE marks) | No |
| 2. Magnet | Free | Yes, if it sticks strongly | No |
| 3. Color at wear points | Free | Often, for plated pieces | No |
| 4. Weight and density | Kitchen scale | Often | No |
| 5. Acid test | Jeweler, or a kit | Yes | Close, with limits |
| 6. XRF analysis | Jeweler or refiner | Yes | Yes, metal content in numbers |
Read the stamp: 10K, 14K, 417, 585, 750
Real gold jewelry sold in the United States is normally stamped with its karat. Look inside ring bands, on chain clasps and end tags, on earring posts and on the back of pendant bails.
| Stamp | Meaning |
|---|---|
| 10K, 10KT, 417 | Solid 10 karat, 41.7% gold |
| 14K, 14KT, 585 | Solid 14 karat, 58.3% gold |
| 18K, 18KT, 750 | Solid 18 karat, 75% gold |
| 24K, 999 | Pure gold |
| GP, GE, HGE, RGP | Gold plated or electroplated over another metal |
| GF, 1/20 14K GF | Gold filled: a thicker gold layer bonded to base metal |
| 925 with gold color | Sterling silver, usually plated (vermeil if the plating is heavy) |
A stamp tells you what the maker claims. It does not prove it, because a stamp is just a mark and counterfeiters use them. A missing stamp also does not prove a piece is fake; old and handmade pieces are sometimes unmarked. Our page on hallmarks and gold stamps lists the less common marks, and the gold karat guide explains the percentages.
Magnet test: what it proves and what it does not
Gold is not magnetic. Neither are silver, copper or zinc, the usual alloy metals. So if a strong magnet grabs the piece firmly, the metal under the surface is likely iron or steel, and the piece is not solid gold.
A magnet that does not stick proves very little. Brass, aluminum, many plated base metals and tungsten are not magnetic either, and they are exactly what fake gold is made of. Ignore the clasp spring when testing: small steel springs inside real gold clasps can pull slightly on a magnet.
Weight, density and color checks
Gold is very dense. Pure gold weighs about 19.3 grams per cubic centimeter, while brass, the usual base of fake jewelry, is around 8.5. Karat gold falls between the two but stays well above brass, so real gold feels heavy for its size. A large-looking chain that feels light is a warning, though hollow construction can also make real gold light.
For a loose, plain piece you can check density at home: weigh it, then weigh how much water it displaces, and divide. Pieces with stones, hollow links or springs throw this off, so treat the number as a hint.
Color tells you more at the edges than on the face. Look at clasp edges, the inside of ring bands and the ends of links, where the piece rubs. Plating wears through there first and shows silver, gray or a coppery color. Solid gold looks the same color all the way through.
Color differences between real karats
Real gold also varies in color, which confuses people. 10K is noticeably paler and slightly greenish next to 14K, and 18K looks deeper and warmer still. A pale chain is not necessarily fake; it may simply be 10K. Compare against a known piece of the same karat in daylight, not under store lights, before drawing any conclusion from color alone.
Plated and filled pieces
Plated and filled jewelry is real gold on the outside only, and the labels mean specific things under the FTC Jewelry Guides and trade standards:
| Term | What it is | Gold layer |
|---|---|---|
| Gold plated / gold electroplate | Thin gold layer applied by electricity over base metal | At least about 0.175 microns of fine gold to use the term electroplate |
| Heavy gold electroplate | A thicker electroplated layer | At least about 2.5 microns |
| Gold filled | Gold sheet bonded to a base metal core | Industry standard: at least 1/20 of the total weight in karat gold |
| Vermeil | Gold plating over sterling silver | Heavier plating over silver, not base metal |
| Solid gold | Karat gold all the way through (or hollow tubes of karat gold) | The whole piece |
None of these are illegal or bad on their own. They become a problem when they are sold as solid gold. Plated pieces also explain most green-skin complaints, covered in why jewelry turns skin green, and the full comparison is in plated vs filled vs solid gold.
Hollow and heavy are not the same as fake and real
Weight checks have one big trap: hollow gold. Many real gold chains, hoops and bangles are made from hollow tubes to give a big look with less metal. They are real karat gold and carry real stamps, but they feel light for their size. A light piece is a reason to look closer, not proof of a fake.
The opposite trap exists too. A heavy piece can still be fake if the core is a dense base metal. Weight is a screen, never a verdict. Our Hollow Cuban Chain 6mm is real 10K gold at 16.55 g, named hollow in the title, which is how a seller should describe it.
Acid test: use it carefully
The traditional jeweler's test rubs the piece on a black touchstone to leave a streak, then applies acids made for each karat. If the streak survives a 14K acid but dissolves under 18K, the metal is about 14K.
- It removes a little metal. The streak is gold taken off the piece, so it is done in a hidden spot.
- It rounds to the nearest acid. It tells you "about 14K", not an exact number.
- Thick plating can fool it. If only the surface is tested, a heavy plating can read as gold. Jewelers file a small notch first to test below the surface.
- The acids are dangerous. Nitric acid burns skin and damages stones and finishes. Home kits exist, but this is a test best left to someone who does it daily.
XRF testing at a jeweler
X-ray fluorescence (XRF) is the modern answer. The analyzer points X-rays at the metal and reads the energy each element gives back, then reports the composition in percentages: gold, silver, copper, zinc, nickel and so on. It takes about a minute and leaves no mark, which is why refiners, pawn dealers and many jewelers use it.
XRF reads the surface layers, so very heavy plating can still mislead a basic reading; good operators test several spots, including a worn edge or clasp, and many machines flag plating. Even so, an XRF number from a professional is the closest thing to proof without melting the piece.
Home tests that do not work
The internet is full of gold tests. Most are unreliable, and some damage the piece:
| "Test" | Claim | Reality |
|---|---|---|
| Bite test | Real gold dents | Only near-pure gold is soft enough; 10K and 14K do not dent, and you can damage teeth or the piece |
| Vinegar soak | Fake gold changes color | Proves little; plated pieces can survive a short soak |
| Ceramic scratch | Real gold leaves a gold streak | Scratches your piece, and plated items can streak gold too |
| Flame test | Real gold does not darken | Can damage stones, solder joints and finishes |
| Floating test | Gold sinks | Every metal used in fake jewelry sinks too |
Stick to the stamp, the magnet, weight and color as screens, and let a professional test with XRF for the answer.
Getting gold tested in Miami
Ask any established jeweler, refiner or gold buyer whether they test with XRF; it is the method to prefer for a piece you care about because it does no damage. Bring the receipt if you have one, and ask for the reading in writing.
If you want to compare with known solid gold, our pieces are listed with metal and karat, and many list gold weight in grams. Examples: the Franco Chain 3mm is 10K and 24.26 g at $3,425, and the Classic Ball Earrings Large 7mm are 14K at $315. You can see the chains in person at the showroom, 7900 NW 27th Ave, Unit 506, Miami. Chains get their own step-by-step in how to tell if a Cuban link is real; diamonds are covered in how to spot fake diamonds, and our solid chains are in gold Cuban link chains.
What fake gold is usually made of
- Brass: copper and zinc, yellow and cheap. The most common base under gold plating.
- Stainless steel with gold-colored coating: durable and often marketed as "won't tarnish". Weakly magnetic in some grades.
- Sterling silver with gold plating: real silver, stamped 925, sold as vermeil when the plating is heavy.
- Tungsten-core fakes: rare in jewelry but used in fake bars because tungsten is nearly as dense as gold. Only XRF or deeper testing catches it.
Buying gold safely in the first place
The easiest fake to catch is the one you never buy. Buy from sellers who state metal and karat on the receipt, who list gold weight in grams, and who name hollow or plated construction plainly. Be wary of prices far below the metal value: if a 14K chain weighing 30 grams is offered for less than the gold in it is worth, something is wrong. Check the day's gold price, multiply by the karat share and the weight, and compare.
Questions people ask
Can fake gold have a 14K stamp?
Yes. A stamp is only a mark, and counterfeit pieces often carry 14K or 585 stamps. Use the stamp as a starting point and confirm with an XRF or acid test.
Is real gold magnetic?
No. Gold and its usual alloy metals are not magnetic, so a piece that a strong magnet grabs is not solid gold. A non-magnetic piece can still be fake, because brass and many plated metals are not magnetic either.
How do pawn shops test gold?
Most use an acid test on a touchstone, an electronic tester, or an XRF analyzer. XRF is the most precise and leaves no mark; acid testing removes a small streak of metal.
Sources
See it on you, at Suite 506.
Every piece has its price online. Buy it here, or come try it on in Miami.
















