Turning crushed drink cans into weighed aluminum ingots

Nov 01, 2026 Last reply: 4 days ago 4 Replies

Rate this thread:

bbarlow9 1 week ago

This project answers a question every caster has wondered about at some point: how much usable aluminum actually comes out of a big pile of crushed drink cans. The maker, Erik Espinoza, melts down 120 used cans in a small cylindrical propane-fired furnace and casts the result into ingots, weighing everything along the way to get a real yield number instead of a guess.

The setup is well thought out. A propane tank feeds the furnace through a connected hose, and a handheld yellow torch, tongs, and a slotted ladle handle the charging and skimming. Ingots go into molds, then get a water quench in a stainless steel bowl. A fire extinguisher sits nearby, and the maker wears a leather apron, gloves, and a mask. A Taylor TE10FT digital scale tracks every can individually, which is what makes the final numbers trustworthy.

The charge breaks down as 100 crushed 12 oz cans and 20 crushed 16 oz cans, weighing in at an estimated 1655 g of starting material. Individual cans came in at 15 to 16 g for the large ones and 12.5 to 13 g for the small ones. A 105 g aluminum starter ingot goes in first to create a molten pool, which shields the thin-walled cans from excessive oxidation as they are added to the running furnace. Dross and residual slag get skimmed off before pouring.

The result: six ingots totaling 963 g, or 2 lb 2.0 oz, a 58% capture rate. That leaves a fair bit on the table as dross. The maker plans to compare these results against an electric furnace with argon shielding, which should make for a interesting follow-up.

Project by Erik Espinoza. Watch it on their YouTube channel:

formatting link

Vote

jtate_1952 6 days ago

Nice yield documentation. One thing I am curious about: how did the maker get the melt started before any cans went in? Starting from a cold crucible with nothing but thin-walled cans seems tricky, so I would like to know what was in there first and how that first pool was established.

Vote

CedarHeron591 6 days ago

The starter ingot trick strikes me as the smartest part of the whole run. Pre-melting a 105 g chunk to form a molten pool gives the thin can material somewhere to drop into instead of sitting exposed to the flame, and the fact that it shields them from oxidation shows in that 58% figure. Without it, I suspect the capture rate would look a lot worse.

Vote

tundra_owl 6 days ago

One variation worth trying: compact the crushed cans much tighter before charging them. Loose, springy cans have a lot of surface area exposed to the furnace atmosphere, and that is exactly where oxidation losses come from. Densifying the charge into tight plugs could reduce the dross fraction and push the capture rate above 58% without changing any other part of the setup.

Vote

ZincLynx794 6 days ago

The maker answered that with a 105 g aluminum starter ingot. It went into the crucible first to create a molten pool, and once that pool was liquid, the crushed cans were inserted into the running furnace. The molten pool shielded the thin-walled cans from excessive oxidation as they melted in.

Vote