How to Spot Fake 916 Gold in Singapore
Somebody has given you a chain, or you bought one on a marketplace, or a piece has been sitting in the family box for thirty years and nobody is quite sure where it came from. It says 916 on the clasp. It looks right. And you have that low-level unease that you cannot resolve by staring at it harder.
So how do you actually check? Which of the tests people talk about online are worth doing, and which are theatre? And if you genuinely need a definite answer, who in Singapore can give you one?
Here is the honest version, including the parts that make jewellers uncomfortable.
Start here: a 916 stamp is not proof of anything
This is the first thing to understand, and most guides bury it.
A hallmark is a claim, not evidence. Stamping "916" into metal is the easiest part of making a convincing fake, and anyone counterfeiting gold will absolutely stamp it. A piece with a 916 mark has told you what the seller wants you to believe.
That does not mean the stamp is useless. A missing stamp on a piece sold as 916 is a genuine warning sign, because legitimate manufacturers mark their work. But present-and-correct proves nothing on its own, and treating a stamp as verification is the single most common mistake buyers make.
Which is why the rest of this page is about weight and physics rather than markings.
What fakes in Singapore actually are
Some reassurance, because the internet makes this scarier than it is.
Almost all fake gold jewellery is gold-plated brass. A base metal core, usually brass or occasionally steel, with a thin layer of real gold over the top. Plated pieces are cheap to produce and they look convincing until you weigh them.
You will read alarming things about tungsten fakes, because tungsten has a density of 19.25 g/cm³ against gold's 19.32, so it can pass a density check. That is a real problem, and it is almost entirely a bullion bar problem rather than a jewellery one. Tungsten is extremely hard, cannot be cast or soldered the way gold can, and nobody is producing convincing tungsten-cored rope chains. If you are checking a chain, a ring or a bangle, you are realistically looking for plated brass.
That is good news, because brass and gold are nothing alike by weight, and weight is something you can measure at your kitchen table.
The density test, which actually works
Gold is remarkably heavy. That is the property counterfeiters cannot fake with cheap metals, and it is the basis of the only reliable test a normal person can run at home.
Here is what different metals weigh per cubic centimetre:
| Metal | Density (g/cm³) |
|---|---|
| 999 gold (24K) | 19.32 |
| Tungsten | 19.25 |
| 916 gold (22K) | roughly 17.7 to 18.2 |
| 750 gold (18K) | roughly 15.6 |
| Silver | 10.49 |
| Brass | roughly 8.5 |
| Steel | roughly 7.9 |
Look at the gap between 916 gold and brass. Gold is more than twice as heavy for the same volume. That is not a subtle difference you might miss; it is enormous, and it is why this test catches plated brass every time.
To measure it you need a scale accurate to 0.01g, a cup of water, and something thin to suspend the piece with. Weigh the piece in air. Then weigh it fully submerged, hanging in the water without touching the sides or bottom. Then:
Density = weight in air ÷ (weight in air − weight in water)
Work an example. A 20g chain that is genuinely 916 gold will read about 18.87g submerged, because it displaces only about 1.13cm³ of water. The same chain in brass would read about 17.65g submerged, displacing 2.35cm³. Run the division and you get roughly 17.7 for the real one and roughly 8.5 for the fake.
The catch is the precision. On a 20g piece the difference between real and fake is about 1.2g of submerged weight, so a kitchen scale reading to whole grams will not do. A jeweller's scale reading to 0.01g will.
Every test, and what each one misses
No single test settles it. This is the honest ranking.
| Test | Reliability | What it misses |
|---|---|---|
| Fire assay | Highest. The reference standard | Destroys part of the piece. Lab only |
| XRF (X-ray fluorescence) | Very high, non-destructive | Reads the surface only, so thick plating can fool it |
| Density / specific gravity | High for jewellery | Fooled by tungsten. Needs a 0.01g scale |
| Acid test | Moderate | Marks the piece. Only tests the surface |
| Ceramic scratch | Low | Marks the piece. Easily misread |
| Magnet | Very low | Only catches iron or steel cores. Brass is not magnetic |
| Reading the hallmark | None on its own | Stamps are trivially faked |
Two rows deserve a second look.
The magnet test is nearly worthless, and it is the one most often repeated online. Gold is not magnetic, true. Neither is brass, which is what your fake is most likely made of. A piece passing the magnet test has told you almost nothing.
XRF reads the surface, which surprises people because XRF is what the professionals use. Fire an X-ray beam at a gold-plated brass chain and it will report gold, because gold is what the beam is hitting. This is precisely why density and XRF are a strong pair: plating fools XRF and gets caught by density, while tungsten does the reverse.
What a proper Singapore hallmark looks like
Here is the specific thing to look for, and almost nobody knows it.
The Singapore Assay Office, set up in 1979 and part of Singapore Test Services, is the only recognised independent body in Singapore that tests and certifies precious metal articles. When SAO certifies a piece, it applies a set of three marks: the jeweller's own mark, the SAO mark, which is a lion's head, and a standard mark showing the purity.
So a fully SAO-hallmarked piece carries a small lion. If you have a piece with one, that is a meaningfully stronger signal than a bare "916" punch, because it means an independent laboratory tested the metal and found it at or above 91.6% gold.
Plenty of perfectly genuine gold sold in Singapore does not carry the SAO lion, so its absence is not evidence of a fake. Its presence is good evidence of the opposite.
Where to get a definite answer
If you need certainty rather than probability, stop testing at home and get it assayed.
The Singapore Assay Office tests using XRF and fire assay and will tell you what the metal actually is. That is the end of the argument, and it is worth the trip for a piece of real value or unclear provenance.
A jeweller with an XRF machine can also give you a fast, non-destructive reading, which is enough for most situations as long as you remember the surface-reading limitation above.
The reframe: the best test is not a test
Here is the thing worth taking away, and it is more useful than any of the above.
Every test on this page is something you do after you have already handed over money. The real protection happens before that, and it is not a test at all.
Buy from a shop that will buy the piece back. A jeweller who guarantees buyback at the day's rate on pieces they sold has put their own money behind the purity. If the metal were not what they said, they would be the one out of pocket, every single time, forever. No home test carries that kind of guarantee, because no test is backed by anyone's balance sheet.
Then add the two habits that make you almost impossible to defraud. Get the weight and purity written on the receipt, because a written claim is enforceable in a way a verbal one is not. And keep the receipt, because it is the document that makes any later valuation, insurance claim or dispute straightforward.
Almost every fake-gold story starts the same way: a cash purchase, no receipt, no named shop, a price that seemed too good. The physics is interesting, but the paperwork is what actually protects you.
If you think you were sold fake gold
Get it assayed first. Have the metal tested before you accuse anyone. You want a result, not a suspicion.
Go back to the seller with the receipt. Legitimate businesses make mistakes and legitimate businesses fix them.
Escalate to CASE if the seller will not engage. The Consumers Association of Singapore handles disputes over misrepresented goods.
Report suspected fraud to the police. Deliberately selling base metal as 916 gold is not a customer service issue.
At JJ Gold, every piece is real 916 or 999 gold. Every piece is weighed in front of you, hallmarked, and put on a receipt with the weight and purity on it, and we buy back what we sell at the day's rate. That last part is the only guarantee in this whole article that costs us money, which is exactly why it is worth more than the rest.
Run the density test if you have a piece you are unsure about. It is genuinely good, and on plated brass it is decisive. But the reason to buy somewhere with a receipt, a hallmark and a buyback commitment is that you never have to run it at all.
Bring anything you are unsure about to City Plaza #01-47 (810 Geylang Road, Singapore 409286) or Lucky Plaza #02-47 (304 Orchard Road, Singapore 238863) and we will weigh it and tell you what we think, whether or not you bought it from us. Call 8950 4377, or browse the collection at jjgoldjewellery.com. If you want the wider picture on how gold is priced and what to ask before you buy, start with our full buying guide.
Weigh it. Get it in writing. Buy where they will take it back.