Zinc Alloy Magnets from China: Cast vs Sintered, Heat Treatment and What Density Really Tells You

Two different products that look identical in a catalogue

A showroom in Yiwu will show you a "strong zinc mag" and a "heat-treated zinc mag" as if they're the same thing with a price difference. They're not the same thing, and the price difference is the only honest part of the comparison.

The fully dense version — you'll see it sold as die-cast zinc alloy, sometimes listed under a trade name like Zamak — is cast from a zinc alloy with a decent copper and aluminium content. It's heavy for its size, it doesn't soak up moisture, and when it fails it deforms. The other version is sintered from zinc powder: the powder is pressed into shape, and then either left as pressed or heat-treated to migrate the binder and seal the surface. Sintered zinc is lighter, cheaper, and if the process is imperfect it is spongy inside.

I've been buying rare-earth and ferrite magnets out of Zhejiang and Guangdong for about nine years now, mostly for door hardware, LED mounting and motor brackets, and the sintered-vs-cast question comes up on nearly every RFQ. Here's what actually decides it.

Why the density matters more than the grade

Zinc has roughly the same density whether cast or sintered in theory, but a sintered part is never fully dense. There's porosity between particles, and porosity is where problems start. A magnet that absorbs moisture through porosity will corrode from the inside. A magnet with a pin hole will crack when riveted. And a magnet that's 8% less dense weighs less, which surprises buyers who assume weight implies quality but at least gives you a quick field test.

A rule of thumb I've used for years: put an unknown zinc-housed magnet on a kitchen scale and compare it to the weight in the drawing. If the weight is more than about 8% under the calculated figure for full density, it's sintered, or it's got a big void in the middle. This isn't precise enough for a QC pass, but it catches the obvious bait-and-switch on the first box.

Cast zinc alloy's pull strength depends on how much of the part is actually magnet and how much is housing. A lot of "strong" magnets are 60% housing by volume once you cut one open, and the housing does nothing for holding force. Ask for a cross-section of a failed sample. Any factory that won't slice one open has something to hide.

The heat treatment question

Sintered zinc parts are often heat-treated — a low-temperature bake that migrates the organic binder to the surface and improves density and corrosion resistance. Done properly, it's a real improvement. Done cheaply, it's a five-minute pass through an oven that creates a surface that looks sealed but isn't, and the part rusts inside its packaging within six months.

What to ask for: the actual temperature and duration of the treatment, and a sectioned sample showing the surface layer. I've had good results asking the factory to include the treatment parameters in the production record, which at minimum tells you whether they know them or are making them up. The factories that can answer this in detail are usually the ones that actually do it in-house. The ones that scramble are outsourcing it and don't know what their supplier does.

Also worth knowing: heat-treated sintered parts have a slight dimensional change. If your application has a tight tolerance, ask what the post-treatment shrinkage is and get a first-article inspection from after the treatment, not before. I've seen a 0.15mm shift on a 12mm part cause an assembly line stoppage for a client making small door catches.

What the end use should drive

The decision is rarely about quality in the abstract. It's about what the part has to do.

For door catches, cabinet latches, and anything that gets hit or riveted a lot, I buy cast zinc. The deformation behaviour is predictable, you can drill it, tap it and rivet it without cracking, and the part survives thermal cycling around a steel frame. The extra cost is real — often 25–40% more per piece at 5000 pieces — but the return rate on the cheaper version more than eats that.

For LED mounting, decorative badges, and low-stress position-holding, sintered is often fine and saves meaningful money. The failure mode, when it happens, is a cracked corner or a slowly rusting edge, and neither is catastrophic for a decorative part.

For anything that will be exposed to salt air or condensation — outdoor lighting, marine hardware, anything near a coast — I want cast or I want sealed. Sintered zinc outdoors, even heat-treated, is a corrosion gamble I've lost twice and won't take a third time.

And one thing that trips up new buyers: the magnet inside the housing matters more than the housing. A cast zinc housing with a weak ferrite core gives you a beautiful part that holds nothing. Get the pull force tested on a finished assembly with the actual mating surface — not on a bare magnet, not in the factory's rig with a polished steel plate. Real surfaces are painted, machined or galvanised, and everyone loses 20–40% against the lab number.

How I audit the supplier beyond the sample

Samples prove nothing except that the factory can produce one good part when they know someone is looking. What matters is whether the 5000th piece matches.

• Ask for a weight certificate from the last production lot, not the sample lot. Zinc alloy density is stable enough that a weight chart is a useful untruth detector.
• Request five consecutive lot numbers with a retained sample from each. If they can't produce retained samples, the "quality system" in the brochure is fiction.
• Test pull force on 20 pieces from three different cartons, cold. Cold is where cheap sintered parts fail, and the difference shows up around −5°C to −10°C rather than at room temperature.
• Ask what happens to a drop test failure — does the part go back into the bin, get reworked, or get scrapped? The rework answer is where most of your field failures are born.

None of this is unusual for experienced buyers. The problem is that most buyers never get past the sample and the brochure, and the difference between a $0.42 part and a $0.61 part is usually exactly what the audit finds.

If you're comparing quotes from two suppliers and the difference is 30%, the cheap one isn't being clever about overhead. They're using a different process or a thinner coating or a recycled magnet core. Worth finding out which.