The short version: check the stamp, then a magnet, then density — those three are free and can tell you something is wrong. Only an XRF reading or a proper acid test can tell you a piece is right. No test you can do at home distinguishes 21K from 22K.
Almost everyone buying gold in Pakistan does one check: they look for the stamp. It is the right place to start and the wrong place to stop, because a stamp is punched by whoever made the piece. It records an assertion. Verification is a different activity.
Start with the stamp — and its limits
Genuine pieces usually carry a mark: 24K, 22K, 21K, 18K, or the equivalent numeric fineness — 916 for 22K, 875 for 21K, 750 for 18K. Absence of any mark is a reason for caution. Presence of one is not, by itself, evidence.
This is where Pakistan differs from India in a way worth understanding. India runs mandatory BIS hallmarking, where an independent assay office tests and marks the piece. Pakistan has no equivalent enforced across the retail trade, so the mark on a piece from an ordinary shop is generally the maker's own. Reputable makers stamp honestly, and in a bazaar built on multi-generational relationships there is a real cost to being caught. But "reputable" is a judgement about the seller, not a measurement of the metal.
The four checks, in order
Run them cheapest-first. Each step can eliminate a piece; only the last one can confirm it.
- Read the stamp. Ten seconds. No mark, or a mark that disagrees with what you were told, ends the conversation early.
- Magnet. A strong magnet, held close. Any pull means a ferrous metal is present. Gold is not magnetic.
- Density. Weigh the piece dry, then weigh it suspended in water. The difference gives volume, and mass divided by volume gives density. Gold is dense — roughly 19.3 g/cm³ pure, and still around 17 for 22K. Brass sits near 8.5.
- XRF or acid. Done by someone with the equipment. This is the step that produces an actual purity number.
The home tests, honestly assessed
Magnet — useful, narrow
Catches steel and iron cores, which is a common way to bulk out a fake. Misses brass, copper, lead and tungsten entirely. Tungsten is the awkward one: it is close to gold in density and non-magnetic, which is why serious buyers do not stop at density either.
Density — the best one you can do yourself
Genuinely informative and cheap. Its weakness is precision: you need an accurate scale, and any hollow section, solder join or set stone throws the result off. Use it to catch gross substitution, not to argue about karat.
Vinegar, bleach and "scratch" tests — skip them
These circulate widely and are close to useless. Plated pieces pass them, because you are testing the plating. Some of them will damage a genuine piece. There is no reason to pour household chemicals on something you just paid for.
THE PLATING PROBLEM
Every surface test — colour, acid on the outside, vinegar — is testing the outermost layer. A thick gold plating over brass passes all of them. This is precisely why density and XRF matter: one reads the whole object, and the other reads deeper than a plating.
The tests that give a real number
XRF fires X-rays at the piece and reads what comes back, returning a composition breakdown in seconds without damaging anything. It is the standard tool in the trade now. Ask where the machine is, watch the reading, and note that XRF reads the surface layer in depth — a competent operator will test more than one point on a piece.
Touchstone and acid is the older method: rub the piece on a stone, apply graded acids, and compare against known needles. In experienced hands it is accurate to a karat or so and costs almost nothing. It removes a tiny amount of metal.
Fire assay is the definitive answer and destroys the sample, so it belongs to refiners and bulk transactions, not to a necklace you want to keep.
What no test at home can do
It cannot separate 21K from 87.5% — those are the same thing — and it cannot separate 21K from 22K. That is a 4.1 percentage point difference in gold content, which is finer than kitchen-scale density work can resolve. If the distinction matters to your purchase, and it should because the two carry different prices, then the only honest answer is a machine reading. What follows from that difference is set out in 21K vs 22K gold.
The paperwork that protects you
Testing tells you about the metal today. The receipt is what protects you in five years, when you want to sell, exchange or insure the piece and nobody remembers the conversation. Insist that it records:
- Weight in grams, to two decimal places
- Karat, stated explicitly
- The rate applied, and the date
- Making charge, separately from metal
- Stone weight and price, if any, separately again
- The seller's buy-back or exchange terms
Keep it with the piece, not in a drawer of receipts. When you sell, the documented weight and purity make the conversation short — which is the whole point. More on that in buying gold jewellery in Peshawar.
Common questions
Is hallmarking mandatory in Pakistan?
Pakistan does not enforce hallmarking across the retail trade the way India does through BIS. Hallmarking schemes exist and some jewellers participate, but a stamp on a piece bought in an ordinary shop is generally the maker's own mark rather than an independently assayed certification. Treat it as a claim to be verified for anything expensive.
Does the magnet test prove gold is real?
No. It proves only that the piece is not made of, or cored with, a magnetic metal such as steel or iron. Brass, copper, lead and tungsten are all non-magnetic, and all have been used to fake gold. Passing the magnet test rules out one family of fakes and nothing else.
Can I test gold purity at home accurately?
You can test whether something is broadly gold-like using density, which is genuinely useful. You cannot get a reliable karat number at home. The gap between 21K and 22K is 4.1 percentage points, which is far finer than any kitchen method can resolve.
What is XRF testing and where do I get it?
X-ray fluorescence reads the metal composition of a piece without damaging it and returns a purity figure in seconds. Larger jewellers, assay offices and gold dealers commonly have a machine, and many will test a piece for a small fee or free for customers. For a large purchase it is the single most useful thing you can do.