Melting Metal With A DIY Foundry Furnace

Foundry Furnace

If you want to do casting at home, you’ll need a way to melt metal. [Jake]’s DIY foundry furnace gets hot enough to melt aluminium, and is built out a mix of scrap parts.

The chamber of the furnace is built out of a water heater tank which has been lined with a special cement that refracts heat. The furnace is heated by a Babington burner. This type of burner works by atomizing the fuel and injecting it into the furnace. They are good for burning waste oil to achieve high heats.

A scrap Volkswagen oil pump and a cordless drill are used to feed oil into the burner. Once it’s fired up, the furnace takes about 10 minutes to melt the 11 pounds of metal that it can hold. [Jake] melted about 40 pounds of aluminium alloy from scrap alloy wheels in 2 hours, which should be more than enough for a home casting project.

After the break, check out the overview of the device and a demo of melting aluminium.

33 thoughts on “Melting Metal With A DIY Foundry Furnace

  1. Use of a dictionary would aid in Eric’s writing. refractory material doesn’t refract heat. it is any material that has the ability to retain its physical shape and chemical identity when subjected to high temperatures.

  2. “…special cement that refracts heat”

    Refractory cement is a cement that is refractory to heat (i.e. resistant to, or retains strength in the face of). Refractory cement does NOT change the angle of incident heat beams according to Snell’s law.

    Disclaimer- physics Nazi only slightly better than grammar Nazi.

    1. To expand on rdasx a bit – A #80 drill bit is 13.5 thousanths, so that’s a pretty tiny hole, but not out of what is doable with a pin vise and tiny bit. A good dremel drill press could probably do it powered, too.

      1. I have heard reports of success drilling by hand, but it’s chancy. A drill press is recommended. I used a CNC mill at the highest spindle speed and a really low feed-rate.

      2. Key is to to get the drill cutting perpendicular to the surface. If attempts are made to drill at an angle the bit will tend to skate off the surface, continually bending the bit while it’s rotating and that soon snaps the bit.That’s one reason why low feed rates really help at the start – it gives the bit more time to “mill away” any misalignment, and start to form the dimple to stop the bit wandering.

        Have wondered if spark erosion might be a easier way to make these small holes.

  3. MUCH easier to get an old oil furnace, which has 100 psi oil pump, blower, igniter already built. Furnace folks are always pulling them out and throwing away. I have only to call and ask if one is sitting there, and I’ve never been asked for any payment.

    1. I’ve got one of those and will probably transfer the ignitor over. The beauty of the Babbington is that the chunks of batter, french fries, and turkey skin in the waste oil can not clog the nozzle. There are special waste oil nozzles of oil burners, but they are made for motor oil and still need filtering.

  4. I recently have been experimenting with different heat sources for such furnaces, and have found that nothing — not even waste oil — can beat the heat capacity per dollar of Pennsylvania Anthracite coal. That is, if you can get any. Forced air is a must, but after that, it can burn hotter than oxyacetylene! $5/50lbs or better by the ton.

      1. The biggest advantages of coke are involved when you’re trying to use the coke to smelt iron ore, as it’s strongly reducing (and makes huge amounts of CO)…other than that, it’s cheaper per BTU, but not necessarily more dense than anthracite, which is approaching graphite (which is part of what it makes darn hard to light)…

  5. You have to be careful when playing with molten aluminum. When you’re skimming your aluminum you just picked up your “skimming device” off the ground where there is snow all around. Aluminum will explode when mixed with moisture, and also you weren’t wearing a face shield while doing it. You have to watch out when pouring your ingots with moisture in the molds. Both your molds and your skimmer should be heated up before tapping. Rule of thumb, you shouldn’t have any snow/ice/water around.

    Source: Run a smelter

    1. As another who has run a smelter(my uncle runs an aluminum smelter for fun and profit at his junk yard) there are other things to beware of as well.

      1. Make sure that’s aluminum you are putting in there. We once got an old transmission from a motorcycle that was MAGNESIUM. Very, very, very bad day. ended up with massisve overhaul to the melted remains of our smelter.

      2. Watch out for die cast aluminum, which can contain high amounts of zinc. Zinc fumes can KILL you. Watch for yellowish fumes.

      3. As above, Make sure there is no water in your equipment…we have had aluminum explosions…got lucky and no burns, but definitely brown trousers!!!

      Our burner is based on one my grandfather built based on his experience in iron smelting when he lived in Rockford, Illinois. it is a simple design that is a tube within a tube within a tube. the oil is injected into the outside tube, flows down to the end and returns up the center tube. There are simple drilled ports for the vapor to escape at the end where our forced air(old vacuum cleaners!!) enters..I’ll make a drawing and post it when I get a chance. We run on diesel and used oil. This burner has run for nearly 15 years with minor maintenance. I’m not sure how hot it can go, but we have melted down forged steel pistons…ended up making a whole batch of 45% steel/55% aluminum ingots.


        link to the (very crude) drawing of our burner.

        Startup sequence is
        1 preheat burner. we use an acetylene torch.
        2. once vaporization is underway( thick white smoke that easily ignites) begin adding forced air. add slowly to full force.
        3 adjust fuel flow for wanted heat/optimal burn

        shutdown is simply removing fuel and allowing burner to cool via forced air.

        no fuel source is required other than your liquid of choice once full burn has started.

  6. I can’t believe nobody else noticed the fact that there’s a small gasoline lantern heating the 40# propane tank. Moonshiner techniques at their best, and a great way to blow yourself up. Personally I prefer the heated water bath technique. I guess tank rupture is probably not the primary safety concern when operating something capable of making molten metal.

      1. There’s still a problem with the oil in that it could heats up to the oil’s flash point. At bare minimum I’d want a thermometer on that oil tank, and if I was operating it I would be far happier with some some kind of thermostat of temperature limiter on there.
        Water bath just limit the temperature to 100 deg C. Old deep fat fryer = couple of kW oil heating element +thermostatic control up to aprox 200 deg C.

        stove might potentially a problem. I have one of those coleman stoves myself(got sick of gas camping stoves when the fittings kept changing every few years) and they operate on same principle as tilley lamps(to pressurise the fuel you pump air in and that pushes the fuel out) The fuel regulating valves are simple flow restriction valves so if the tank pressure rises more fuel comes out. Now they could enter thermal runaway if loads of hot oil fell down on top, and began heating the air inside the tank.

  7. Well done Jake.
    as described on it’s page the “brute burner” ain’t a babington but a drip feed. Babington’s use the pump for fuel transfer rather than pressuring the oil(compressed gas is what does the atomising in a babington).

    Maybe consider adding a worm drive to the shaft of the oil pump, and powering from a motor. If your oil feed rate is low(and reduction is high enough), and high pressure isn’t required then it doesn’t take a large motor to power the pump. Indeed I’ve pumped sufficient oil for a calculated 24kW babington (if combustion was complete) using 12V automotive washer gear pumps with PWM for speed control. One thing to watch – washer pumps don’t like getting hot and they seize after a few minutes of pumping hot oil. Think that must be due to a mechanical issue in the pump perhaps parts expanding-They were as good as new after cooling down.

    Talking of babington’s they don’t need a ball
    Last year I knocked up a prototype in half an hour out of some 8mm stainless tube and some old tin cans. It wasn’t that successful due to a fuel leak, though air inlet holes, and the washer pump cutting out, but the video of it should demonstrate the principles.of how twin tangential atomizing jets can add swirl stabilisation helping to keep the burner lit, and increase the intensity of combustion.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.