How to Tell If Silver Is Real: 7 Tests You Can Do at Home

In June 2023, a coin-grading company called NGC examined a 2023 American Silver Eagle that had a problem. It was too heavy. A genuine Eagle weighs 31.103 grams, and this one weighed 31.54. The counterfeiter had made a silver coin out of almost no silver, and then, in a burst of generosity, thrown in some extra.

That fake is the best short answer to how to tell if silver is real. No single home test proves a coin is genuine, but five cheap ones used together catch the common fakes: weigh it, measure it, then try a magnet, an ice cube and a glass of water. A fake can get one of those right. Getting all of them right costs a counterfeiter real silver, which defeats the purpose of counterfeiting silver.

Think of yourself as the bouncer at a bar on a Friday night. A fake ID can get the photo right, or the hologram, or the birthday, and plenty of them do. Getting all three right at once is rare. Your coin is about to be carded.

What are fake silver coins made of?

Mostly copper, with some zinc and nickel. When NGC ran its 2023 fake through an X-ray analyzer, it found 65% copper, 21% zinc, 13.5% nickel and less than half a percent silver. A fake 2000-P Eagle NGC examined three years earlier was 83% copper, 12% zinc, 3% nickel and 1% silver.

Copper, nickel and zinc together make an alloy called nickel silver, which contains no silver at all. It earned the name by looking like silver, and since the 1840s it has been the metal hiding under silver-plated cutlery. So a fake Silver Eagle and a hotel teaspoon are, give or take a design, the same object. Only one of them was sold to you as an investment.

Two facts about that alloy make the rest of this guide work. It is about 19% lighter than silver for its size, and it moves heat about thirteen times more slowly. Every good home test is built on one of those two gaps.

The fakes go where the money is. In a 2018 poll of American dealers by the Industry Council for Tangible Assets, 43.3% said customers had tried to sell them counterfeit Silver Eagles, the highest share for any bullion coin. More than 80% said they were seeing more fakes than five years earlier. Fake bars turned up too, led by copies of SilverTowne and Sunshine Mint bars. The more famous the coin, the more likely someone has copied it. Nobody forges a coin nobody wants.

Tests 1 and 2: How much should a silver coin weigh, and how big should it be?

Start here, because it needs nothing but a kitchen scale that reads to a hundredth of a gram and a pair of digital calipers. Weigh the coin. Measure its width and its thickness. Then compare all three numbers with the mint’s specification.

Sources: US Mint (Silver Eagle weight and width; modern quarter), Royal Canadian Mint (Maple Leaf weight and width), Numista (silver quarter) and the coins’ Wikipedia entries (thicknesses, Morgan dollar). Full links under Sources below.

The trick is to check weight and size together. A counterfeiter can hit the right weight or the right size, but not both, because a cheaper metal needs more space to weigh the same. A nickel-silver fake as heavy as a real Eagle is roughly a quarter bigger by volume. That is exactly how the heavy 2023 fake gave itself away: NGC called it “much too thick.” The 2000-P fake went the other direction. It came in light, at 30.6 grams, and narrow, at 39.9 millimeters.

The scale does other jobs too. A silver quarter and a modern quarter are the same size to the hundredth of a millimeter, but the silver one is 0.58 grams heavier. Sort a coffee can of quarters on a kitchen scale and you’ll find every silver one without reading a single date.

Two cautions. Old circulated coins lose a little metal to decades of pockets, so a worn 1942 quarter can read light without being fake. And NGC itself treated the 2023 fake’s 1.4% extra weight as a warning sign, not a verdict. A reading that is off means keep testing.

To skip the arithmetic, type your readings into Is My Coin Fake?, which checks them against the published specification for 20 silver and gold coins and tells you which measurement is off.

Into the weeds: bars are harder

Coins are made to one published size. Bars aren’t: a ten-ounce bar from one refiner can be a different shape from another’s, so there is no single number to check. Look up that maker’s own specification for that exact bar, and lean harder on the density test below, which doesn’t care about shape. Fake bars are also where a dense filler such as lead is most likely to hide, because a thick bar has more room inside for it.

Test 3: Does real silver stick to a magnet?

No. Silver isn’t noticeably magnetic. Hold a strong rare-earth magnet, the small silver-colored kind sold for hardware projects, against the coin. If the coin jumps to it, there is iron or steel inside, and you can stop testing.

There is a fun second half. Tilt a real silver bar or a large coin and lay the magnet on it. It won’t stick, but it won’t race off either. It slides down slowly, as if wading through syrup. Silver conducts electricity better than any other metal, so the moving magnet sets up small electric currents in it that push back. The same effect makes a magnet dropped down a silver pipe fall slowly.

Now the catch. Copper slows a magnet the same way, and nickel silver doesn’t stick to a magnet any more than silver does. Even an American five-cent piece, which is one-quarter nickel, isn’t visibly attracted to a magnet. NGC’s fake Eagles would have sailed through. The magnet is a bouncer who only checks whether you’re wearing a ski mask. Useful. Not enough.

Test 4: Does ice melt faster on real silver?

Yes, and it’s the most satisfying test on the list. Silver moves heat faster than any other metal, at 429 watts per meter-kelvin. Put an ice cube on a silver coin and another on a modern quarter at the same time. The cube on the silver starts melting at once, as if someone had set it on a warm stove. The cube on the quarter sits there like it has all afternoon.

Silver moves heat about 13 times faster than the metal most fakes are made of

Thermal conductivity, watts per meter-kelvin. Higher means an ice cube melts faster.

The chart shows the catch. Nickel silver, at 33, melts ice about thirteen times slower than silver, so the ice test catches the fakes NGC found. A coin with a solid copper core, plated in silver, would melt ice almost as fast as the real thing. The test also works best on a big piece of bullion. On a dime there just isn’t enough metal to tell much.

Test 5: What does real silver sound like?

Balance the coin on a fingertip and tap its edge with another coin. Real silver rings with a long, clear tone, like a small bell. Most base-metal fakes give a shorter, duller clink.

The ping is a comparison test, though, not a measurement. It works best side by side with a coin you already know is genuine, and it means little on a coin in a plastic capsule. Phone apps that listen to the ring exist. We like them as a hint and would never build one that says “genuine,” because a confident wrong answer would cost someone real money.

Test 6: How do you do a specific gravity test at home?

This is the test counterfeiters fear, because density is a property of the metal itself. You can’t plate it, paint it or engrave it. You need the same kitchen scale, a cup of water and a piece of thread.

1. Weigh the coin dry and write it down. Take it out of any capsule first.
2. Set a cup of water on the scale and press tare, so it reads zero.
3. Hang the coin from the thread so it is completely underwater but touches neither the bottom nor the sides. The scale now shows the weight of the water the coin pushed aside.
4. Divide the dry weight by that reading. The answer is the specific gravity.

Fine silver comes out at about 10.5. Silver that is 90% pure, like a Morgan dollar or a silver quarter, comes out at about 10.3. Copper reads about 9, and nickel silver about 8.5.

Try it on a real Silver Eagle. It weighs 31.10 grams dry and should push aside about 2.96 grams of water: 31.10 divided by 2.96 is 10.5. NGC’s heavy 2023 fake, at 31.54 grams, would have pushed aside about 3.71 grams, for a reading of 8.5. That is a fake ID with the wrong birthday.

Silver is denser than every common fake metal except lead

Density, grams per cubic centimeter (the same number as specific gravity).

Into the weeds: small coins, bubbles and lead

A one-ounce coin pushes aside about three grams of water, so a scale that reads to a hundredth of a gram gives a good answer. A silver dime pushes aside only about a quarter of a gram, and the last digit on the scale swings the result too much to trust. Test ten dimes together on the thread in a small bag instead. Tap the coin once underwater to knock off air bubbles, which make it read light. And because lead is denser than silver, a fake that mixes lead with copper could in principle land near 10.5. That is why density runs alongside the size check and the ice test, never alone: lead melts ice slowly too.

Test 7: What anti-counterfeiting features do real silver coins have?

Get a magnifying glass and look. The major mints now build in features that are expensive to copy, and the fakes keep getting them wrong.

• American Silver Eagle, 2021 and later: the edge is reeded with one reed missing, and the notch moves to a different spot each year. NGC’s 2023 fake put the notch in the right place, then gave the coin only 149 reeds, “dozens fewer than a genuine example would have.” It is hard to think of a more patient way to get caught.
• Canadian Silver Maple Leaf, 2014 and later: fine radial lines across the background, and a tiny laser-engraved maple leaf on the reverse with the year’s last two digits in the middle, visible under magnification. The Royal Canadian Mint also records each die’s security mark in a database it calls Bullion DNA.
• British Britannia, 2021 and later: tilt it, and the trident below Britannia’s gown turns into a padlock. The border also carries microscopic lettering.
• Anything stamped COPY: under the US Hobby Protection Act, a replica coin must be plainly and permanently marked “COPY” in capital letters at least 2 millimeters tall, on a face, never the edge. A coin that says COPY is a legal souvenir. A copy that leaves the word off and is sold as silver is a counterfeit.

On older coins with no security features, compare the details against a photo of a certified example. NGC found cartoonish berries on the 2023 fake. Real engravers don’t draw cartoon berries.

Which home test catches which fake?

This is the bouncer’s whole checklist on one card. The rows are the kinds of fake the metal facts above make possible; a “yes” means that test can physically detect that fake.

Based on the densities and heat figures in the two charts above. The ring test isn’t in the table because it is a comparison, not a measurement.

Every row has at least two yeses. That’s the point of using several tests: each one has a blind spot, and no blind spot is shared by all of them.

Which silver tests should you skip?

• Acid test kits work, but they work by scratching a mark into the metal. That is fine for a lump of scrap and a bad idea for a coin you may want to sell one day.
• Bleach turns silver dark quickly, which tells you very little and leaves you with a coin that looks like it spent a decade in a basement. A different basement.
• Biting is for gold, and mostly for old movies.

When should you have silver tested by a professional?

When the coin is worth much more than its metal, when it’s a large bar, or when your home tests disagree. Grading companies use X-ray fluorescence, the analyzer NGC ran on its fakes, which reads the makeup of the surface. In the 2018 dealer poll, 60.6% said they used counterfeit-detection equipment, such as a handheld verifier that reads the metal’s electrical properties underneath any plating, and even through a plastic capsule. One popular model, the Sigma Metalytics verifier, was listed by one dealer in October 2026 at about $1,300, which is why most stackers borrow the shop’s rather than buy one.

Isn’t it easier to just buy from a reputable dealer?

Yes, and that is the honest best defense. The industry’s own anti-counterfeiting director advised buying from knowledgeable dealers or directly from the mints, and dealers already run these checks every day. If every ounce you own came straight from a mint, you need this page less than most people.

But a lot of silver never passes through a dealer. It comes from a grandparent’s dresser, a coin show table, an online listing or a friend selling a tube. Learning the tests takes an afternoon and a kitchen scale, and the same checks are the first thing a dealer does when you sell. It’s worth knowing what they’ll find.

Why can you test silver at home at all?

Because silver carries its own proof. Its weight for its size, the speed it drinks the cold out of an ice cube, the ring when you tap it: none of those is a promise somebody made. They are properties of the element, and they were the same in a Roman mint as they are on your kitchen counter. No board meeting can revise them, and no counterfeiter can copy them without using actual silver.

Most of what people call money asks you to trust a statement. A coin lets you check. Silver is the only customer in the line who is happy to show ID.

Sources

• The heavy 2023 fake (31.54 grams, 65% copper, 21% zinc, 13.5% nickel, under 0.5% silver, 149 reeds, berries): NGC, “Counterfeit Detection: 2023 Silver Eagle,” June 13, 2023.
• The light 2000-P fake (30.6 grams, 39.9 mm, 83% copper, 1% silver): NGC, “Counterfeit Detection: 2000-P American Silver Eagle,” June 28, 2020, from The Numismatist, March 2020.
• Dealer poll (43.3% saw fake Silver Eagles; more than 80% saw more fakes; 60.6% use detection equipment; fake bars): CoinWeek, June 28, 2018, and the Anti-Counterfeiting Educational Foundation. Industry Council for Tangible Assets online poll, June 7–14, 2018.
• Silver Eagle weight and width: US Mint. Thickness and the 2021 edge notch: Wikipedia, American Silver Eagle.
• Maple Leaf weight, width and security features: Royal Canadian Mint. Thickness and the 2014 start of the radial lines: Wikipedia, Canadian Silver Maple Leaf.
• Britannia security features: The Royal Mint, Britannia 2021 1 oz silver coin.
• Morgan dollar: Wikipedia, Morgan dollar. Silver quarter, 1932–1964: Numista. Modern quarter (5.67 grams) and the five-cent coin (25% nickel): US Mint, coin specifications.
• Silver’s density (10.503 grams per cubic centimeter at 20°C), heat conductivity (429) and electrical conductivity: Wikipedia, Silver. Copper, iron, zinc and lead: their Wikipedia entries. Nickel silver (8.5 and 33): makeitfrom.com, alloy C75200; its use under plated cutlery: Wikipedia, Nickel silver.
• Magnets and silver, and the five-cent coin: University of Maryland, “Is silver magnetic?” and “Is nickel magnetic?”
• The COPY rule: Federal Trade Commission, rules under the Hobby Protection Act (16 CFR Part 304).
• Verifier listing: SD Bullion, $1,299.99, read October 17, 2026. No affiliation.

Figures read October 17, 2026. Worked out by us from the sources above: nickel silver is about 19% less dense than silver and conducts heat about 13 times more slowly; a nickel-silver coin at full weight takes up about 24% more room; a silver quarter is 0.58 grams heavier than a modern one; 90% coin silver has a specific gravity of about 10.3; and the water readings of 2.96 and 3.71 grams in the worked example.

Questions people actually ask

How can you tell if silver is real at home?

Use several cheap tests together. Weigh the coin on a scale that reads to 0.01 gram and measure its width and thickness, then compare all three with the mint's published specification. Then check that it does not stick to a strong magnet, that it melts an ice cube quickly, and that its specific gravity in water is about 10.5 for fine silver or 10.3 for 90% coin silver. One test can be fooled; a fake rarely passes all of them.

Does real silver stick to a magnet?

No. Silver is not noticeably magnetic, so a silver coin should not jump to even a strong rare-earth magnet. If it does, there is iron or steel inside. A magnet will slide slowly down a tilted silver bar because silver conducts electricity so well, but copper does the same, and most fake silver coins are made of copper, nickel and zinc, which do not stick either. The magnet test catches some fakes, not all.

Does the ice test work on silver?

Yes, especially on bullion. Silver conducts heat faster than any other metal, 429 watts per meter-kelvin, so an ice cube placed on a silver coin or bar starts melting almost at once. Nickel silver, the copper-nickel-zinc alloy found in many fakes, conducts heat about 13 times more slowly. The weakness is copper, which is nearly as fast as silver, so a silver-plated copper fake can pass.

How much should a 1 oz silver coin weigh?

A one-ounce coin should weigh one troy ounce of silver plus any alloy. An American Silver Eagle weighs 31.103 grams and is 40.6 millimeters wide, according to the US Mint. A Canadian Silver Maple Leaf weighs 31.11 grams and is 38 millimeters wide, according to the Royal Canadian Mint. Check the weight and the size together, because a fake can match one but rarely both.

Can a fake silver coin weigh the same as a real one?

Yes. A counterfeiter can make a fake of exactly the right weight, and NGC examined a fake 2023 Silver Eagle that actually weighed more than a real one, 31.54 grams against 31.103. But the cheaper metals used in fakes are less dense than silver, so a fake at the right weight has to be bigger. NGC found that coin much too thick for its weight.

What are fake silver coins made of?

Usually a mix of copper, zinc and nickel. NGC's analysis of a fake 2023 Silver Eagle found 65% copper, 21% zinc, 13.5% nickel and less than half a percent silver. A fake 2000-P Eagle was 83% copper, 12% zinc, 3% nickel and 1% silver. That mix is close to nickel silver, an alloy long used under silver-plated cutlery, which contains no silver at all.

How do you do a specific gravity test on a silver coin?

Weigh the coin dry. Put a cup of water on the scale and press tare. Hang the coin from a thread so it is fully underwater without touching the cup, and read the scale. Divide the dry weight by that reading. Fine silver gives about 10.5, 90% coin silver about 10.3, copper about 9 and nickel silver about 8.5. For small coins such as dimes, test ten at once for a reliable reading.

What does real silver sound like when you tap it?

Balanced on a fingertip and tapped with another coin, real silver gives a long, clear ring, like a small bell, while most base-metal fakes give a shorter, duller sound. It is a comparison test rather than a measurement, so it works best next to a coin you know is genuine, and it does not work on a coin in a plastic capsule.

What does COPY stamped on a coin mean?

It means the coin is a legal replica. Under the US Hobby Protection Act, an imitation coin must be plainly and permanently marked COPY in capital letters at least 2 millimeters tall, on the face of the coin rather than the edge. A COPY coin is a souvenir, not silver. A replica sold without the mark and passed off as real is a counterfeit.

Which silver coins are counterfeited the most?

The most popular ones. In a 2018 poll of US dealers by the Industry Council for Tangible Assets, 43.3% said customers had tried to sell them fake American Silver Eagles, the highest share of any bullion coin. Among silver bars, fakes of SilverTowne and Sunshine Mint bars were reported most often. Counterfeiters copy what people already want to buy.

How can you tell if a silver bar is real?

Bars are harder than coins because each refiner makes its own sizes, so compare the bar with that maker's published weight and dimensions. Then lean on the tests that do not care about shape: a magnet, an ice cube and the specific gravity test in water, which should read about 10.5 for fine silver. Large or doubtful bars are worth having checked by a coin shop with a verifier.