Introduction: Zero Cost Aluminum Furnace (No Propane, No Glue / Epoxy, No Welding, Individually Replaceable Parts)

Picture of Zero Cost Aluminum Furnace (No Propane, No Glue / Epoxy, No Welding, Individually Replaceable Parts)




As seen from the picture and title this instructable is about how to make a zero cost aluminum furnace from salvaged items from your local scrapyard or around the house.

Please be advised that though this aluminum furnace can truly be made at zero cost, for some steps you may want to substitute a tiny bit of money for less hassle and trouble, throughout this instructable i shall be pointing out these steps and how you could do so.

BTW this aluminum furnace build can seriously smelt down a lot of aluminum, but takes a while to heat up, so ensure that you have a large store of scrap aluminum before using, for maximum efficiency.


Step 1: Materials

Picture of Materials

If you're reading this step then i guess you're interested in building a Zero Cost  of your own, if so, go salvage / hunt for the following materials

1. Small stainless steel or steel pot, preferably without handles X1
(i found mine under the kitchen sink, by the looks of things it has been there a while)

2. Cement powder X min 1kg
(i got mine by sweeping up left over cement powder at a construction site after they had finished)

3. Milo tin / milk tin / circular biscuit tin , doesn't need to be empty X1
(i like drinking milo, so i already had one to begin with)

4. Very old but still working oven / toaster with heating elements that aren't in a glass cylinder X1
(i got mine when my mom had to get a new oven to prepare Christmas dinner)

5. Misc lengths of wire and both electrical and hardware connectors

6. Cardboard X1sheet

7. Small stainless steel or steel bowl X1
(kitchen stock)

8. Small ceramic plate or tile X1
(kitchen stock)

9. Scrap aluminum in the form of drink cans or otherwise
(beer cans ^_^)

Step 2: Furnace Casting + Coil Bending

Picture of Furnace Casting + Coil Bending

This step deals with Furnace Casting and Coil Bending of the aluminum furnace.

Furnace Casting
1. Take 1/4 of the cement powder you have (250g min) and mix it well with water
2. Pour it into the steel pot and let it set and harden into a level concrete surface
3. Make a cylinder that snugly fits around the milo tin using the sheet of cardboard and some tape
4. Center the cardboard cylinder in the pot, on top of the layer of hardened concrete, and place the milo tin in it, add some weights onto the milo tin just to stablise everything.
5. Take 3/4 of the cement powder you have (meaning the rest of it) and mix it well with water
6. Pour it into the steel pot around the cardboard cylinder and let it set and harden into a concrete ring
7. Remove the milo tin by sliding it out of the cardboard cylinder, then peel out the cardboard, if any cardboard residue remains, burn it off with a flame.

Coil Bending
1. Take apart the oven and remove the longest heating element available
2. Un-coil the heating element and proceed to re-coil it into concentric circles, as seen in the picture, at the bottom do a S-shape so as to give the bowl that is going to rest inside some support
3. Ensure that the individual coils don't touch each other or the pot for that matter, then place it into the concrete ring you've just made.

Step 3: Electrical Wiring

Picture of Electrical Wiring

This step deals with the electrical wiring of the aluminum furnace

Electrical Wiring
1. The heating element is no more than just a gigantic resistor, so all we have to do to is hook it up to mains voltage, and it will start to heat-up, eventually reaching aluminum melting temperature.
2. The problem lies with the nature of resistors, as they heat up their resistance decreases, thus allowing more current through them, which causes them to heat up further. This positive feedback loop can be devastating if left unregulated, as the metals in this furnace melt in the following order
aluminum scrap @ 661 °C
steel bowl and pot @ 1538 °C
heating element coil @ 2852 °C
after which the circuit would be broken thus breaking the feedback loop.
3. To solve this problem, i highly recommend having a thermocouple temperature meter, and if the temperature exceeds 800-1000 °C, cut the mains power until it cools down some or use a light dimmer circuit to regulate how much power gets into the heating element coil, thus slowing the heating process so much and so accurately to the point where heat production = heat desipitation, thus there would be no nett increase in temperature over any substantial length of time.

Step 4: Bowl + Lid + Replacing Parts

Picture of Bowl + Lid + Replacing Parts

Other misc stuffs

As with any furnace a crucible is needed, however since we are only melting aluminum, a steel bowl can be used instead, because steel has a much higher melting point than aluminum, it is also much easier to obtain as compared to ceramic crucibles

In the event that by some chance your heating element coil used in this furnace refuses to produce enough heat to melt even the smallest batch of aluminum, a ceramic plate or tile can be used as a lid and insulator to trap the heat generated in. This method can also be used to speed up the heating of large batches of aluminum or the initial start-up time required.

Replacing Parts
After prolonged use the constant heat could cause wear and tear in all parts of the furnace, by making parts that fit together instead of casting it as a whole, individual parts can be replaced without discarding the whole furnace. The most likely scenario for this would be the positive feedback loop problem described in "electrical wiring", gone wrong.


clothier_bruce (author)2017-07-08

Have you actually used this thing? Did it melt yoiur aluminium? Pics?

clothier_bruce (author)2017-07-06

Can you really bend a long, straight heating element like that without wrecking the inside? Sounds brilliant, but too good to be true.

rsteadman1 (author)2016-03-01

PS: Resistors have a POSITIVE temperature Coefficient; meaning that the hotter it gets, the more resistance it has. Resistors, and ALL conductors have this 1/1 Proportional temp/ohm ratio. Semiconductors, like Silicon, Germanium, (Used in transistors and Diodes,) Have a NEGATIVE temperature coefficeint; meaning that the hotter it gets, the lower the resistance. That's called a cascade, or Avalanche.

inanimatepotatoe (author)2016-02-11

What size pot did you use?

mike.valentine.718 (author)2015-04-08

Could wiring up a dimmer switch system help with controlling the temperature issues?

Dimmer switches don't work as well at higher wattage, they lower the current in the coil by increasing resistance in the circuit. This resistance creates a second point in the circuit where energy needs to be dissipated in the form of heat, where you would need a big heat sink. What you need is one of the thermostats from your stove. They work on a different principle.

Kinnishian (author)TylerM142015-09-23

This is not true. Dimmer switches are rheostats and they are triac devices that turn the AC voltage on after a certain portion of the 1/60th second cycle time of AC voltage. Then they shut off again when the voltage crosses 0.

Only very ancient dimmer switches would work by load-balancing the way you described. Maybe for old theater equipment and the like. I'm not totally familiar.

Kinnishian (author)Kinnishian2015-09-23

First rule of correcting should be to make sure you're correct. I am however wrong to call them "rheostats." Rheostats are usually used in the circuitry but the typical dimmer switches are actually called triac-dimmers, I believe.
So urhm take what I said with some measure of caution. Not a part of electronics I'm totally familiar. Also, now and days there are different dimmers besides triacs because triacs do not neccesarily work well with flourescents of LEDs.

How do the thermostats in your stove work?

piedade (author)lotzofgibberish.2015-07-07

bi-metalic strip that bends when heated and open the electric circuit they form ... after cooling they regain its original form and stablish the circuit again... on-off-on-off..............

but there are other methods!

Kinnishian (author)piedade2015-09-23

Hi piedade. Such metallic switches may have been used in ovens many years back, but they have tremendous hysterisis and no doubt they are not used anymore. I only see them for saftey failure switches. Maybe some thermostats for old rooms, as well, have them.

yes, you can use an electronic dimmer (those used to dim incadescent light bulbs). Just check the dimmer specifications to see if it can handle the amount of power needed for these oven heating elements.

MrMarkinator (author)2015-08-02

How would you set up the thermocouple?

fidgety2 (author)2013-11-27

What are you running your element from, household ect. I Am in the process of making this and currently I have my doubts as to the validity of this instructables alleged results

Flashflint (author)2009-12-28

Have  you actually used this?  It does not seem to me that the heating element would get hot enough or that that lid would hold up very well.  It is also dangerous to use concrete as it is prone to forming air bubbles and exploding.

quickshadow_2 (author)Flashflint2009-12-28

I have used this set-up to melt beer and soft drink cans before, and with satisfactory effect, however i must admit that i swapped out a new stainless steel bowl as the old one was badly stained and didn't photography well, the pot however remains badly oxidized. The heating element was taken from a 1000W oven thus has enough juice at least for aluminum especially with concrete and ceramic as the insulators, also the concrete i used doesn't seem to have any problems with the heat, as i have thoroughly bone dried it.

The exploding concrete you mentioned may have been caused by improper drying of the concrete, this residual moisture would form steam at high temperatures and provide the pressure for the explosion, this may also occur in plaster of paris.

NewB007 (author)quickshadow_22009-12-29

 If too much water was mixed with the concrete to start with, water droplets can remain inside, causing an explosion when the water forms steam.  I would consider a lid and safety glasses mandatory, though, just because of this risk.  It is not worth losing an eye just because you were counting on your mixture to be perfect.

Industrial furnaces of certain types use specialty concretes/mortars for their furnaces, so this isn't a bad idea in general.  Concrete depends on the chemical bonds formed with the water and mineral constituents for strength however, so I would be interested to know how the construction concrete is holding up after heavy or long-term use.  I would be surprised if it didn't slowly crumble as the water is driven off, but it was free, so perhaps that is acceptable wear-and-tear.

Overall, it was a clever use of materials and a simple design for this project.  Good job!

quickshadow_2 (author)NewB0072009-12-29

1st up, thanks for the compliment, and now that you mentioned it, the upper layers of the concrete seem to be ever so slightly flaking like a pastry then very slowly being burnt away by the heat over time, however the process is so pitifully slow that by my calculations, it should last the better part of a year at least, even with daily use.

sparhawk7 (author)quickshadow_22013-10-16

i would assume that the slow heating of the element, and the porous nature of concrete would allow concrete to be relatively safe. if you're really worried, a similarly cheap method could be a terracotta flower pot... don't pour water, or molten metal on the concrete/terracotta and you should be fine. good job.

laxap (author)2009-12-29
Interesting instructable.

  • Any safety measures? safety glasses, protective gloves, etc... May I suggest that you add a word about that.
  • Do you melt aluminium just for the fun, or is it part of another project?
quickshadow_2 (author)laxap2009-12-29

Ok..... this is kinda awkard, well, the truth be told i'm not exactly the a very safety oriented kind of guy, i just do what ever works, i improvise a lot, and that kinda almost always leads to some form of risk taking.

But then again not all of you are so poorly equipped as me, i only have hand tools to work with and no clamps or a dedicated work surface of any sort (i just use the floor), but if i had the dough i would say the min for any type of work DIY or otherwise would be leather gloves, hematically sealed googles, respirator mask and steel toed boots.

I'm melting down the aluminum i've gathered into ingots so that i have purified raw material for an up-coming instructable, but have you ever tried to CARVE STEEL MOLDS USING HAND TOOLS!!! IT TAKES FOREVER!!!

So pls vote for me,cos i hope to use the $100 voucher from shapeways to make some stainless steel moulds i could use in future instructables ^_^

sparhawk7 (author)quickshadow_22013-10-16

muffin tins. use muffin tins for ingots. cast iron works best, but just about any kind will work. DO NOT SPRAY THEM WITH PAM FIRST. that could be bad. ever had bacon grese spatter you? yeah... aluminum.

greenjedi (author)quickshadow_22011-01-27

When a fellow is Lukewarm about safety, His accident record is usually not so hot.

_Scratch_ (author)2010-01-03

nice instructable, have you ever heard of a induction furnace? cuz that may heat up faster and get hotter. but really nice instructable =P

quickshadow_2 (author)_Scratch_2010-01-04

An induction furnace only works with ferrite metals (AKA metals that are magnetic) thus it won't work on aluminum, nice try though

sparhawk7 (author)quickshadow_22013-10-16

when you said FERROUS, you probably meant conductive. which is like... all metals. it is true however, that ferrous metals melt faster in induction coils. that's why graphite crucibles are used in induction furnaces to melt glass. because graphite conducts. so it heats up. thanks for the instructable though, it's super simple and easy to follow.

holzmanj9 (author)quickshadow_22011-04-17

Induction furnaces melt aluminum and gold quite nicely. Have you ever used one? My engineering school used an induction furnace for all the aluminum casting we did.

msw100 (author)quickshadow_22010-09-04

An induction furnace will melt any type of metal not just ferrite, where did you get that from?

trustr (author)quickshadow_22010-02-24

but the induction could heat up the steel bowl ---> the aluminum melts

xenor (author)quickshadow_22010-01-18

Actually, induction furnaces are used for all sorts of metals, including non-magnetic ones. A lot of commercial foundries use induction in preference to the old methods as it's a lot cleaner and can be very accurately controlled.

timothy demaegdt (author)2013-08-11

can' you measure what amps the resistor needs to melt the aluminium and then just buy an circuitbreaker that is slightly higher rated (for example when 5A is used to melt aluminium just put a 6A circuitbreaker in) so when the element gets to hot it shuts of and you can only restart when it is cooled a bit ( i didn't do any specific research tho this is just a random thought)

fr!th (author)2011-06-05

Cool concept, and I like your recycling of the heat element. Seems like a solid idea.

But I had to (create an account and) comment - this is outrageously dangerous! It looks from the image like you are just using crocodile connectors to attach *live mains* to your heating element.

At the very least, can I suggest you put a warning as to the highly dangerous combination of exposed high voltage connections along with a steel outer vessel and molten aluminium. Its a small misstep away from a Darwin award.

I would suggest the following: solder the mains connections and insulate with shrinkwrap. Or use `proper' terminal blocks (there might be one in your salvaged stove).

alternatively, can you bury the connections inside the concrete?

Addtionally, I would suggest that the exterior of the vessel (the SS pot) be connected to the earth connection of the power connector, and the lid if possible. This way if anything bad happens, the RCD will trip instantly - you do have Earth Leakage breakers right?

I don't think this would add much cost/time, but it would make the object significantly safer...

pbates123 (author)fr!th2012-02-05

I know this is a rather late reply - but I just couldn't leave this alone. I hope you know that concrete harbors water which makes concrete conductive.

In fact according to the UBC (Universal Building Code) used in the US it is legal and safe to use what is called a UFER ground. Essentially the ground of a building is connected to rebar that is buried in the concrete foundation.

Never touch a live wire while standing on concrete!!

Never bury unprotected or unapproved electrical connectors or wires in concrete. Darwin has claimed many lives of those who have fallen victim particularly in third world countries without building codes...

You raised some valid points though - Kudo's to you!

Laminarin (author)2011-11-11

Most substances increase their resistance with temperature. Notable exceptions are carbon, germanium, and silicon. A quick search for what oven elements are leads me to believe that they are made of a tungsten or nickel-chromium core surrounded by (electrically) non-conducting heat transfer materials. I'm betting the resistance actually increases with temp.
That being said, it's probably safest/most efficient to stay at as low a temperature as will work for your project. Do you have a way of measuring amps on this setup so one might make a PWM regulation system for regulating the power output?

martins999 (author)2011-05-14

you say that "The problem lies with the nature of resistors, as they heat up their resistance decreases, thus allowing more current through them, which causes them to heat up further" but actually resistance is (almost) proportional to temperature, so if temperature increased so does current. It's a negative feedback, not a positive feedback. Current in a resistor is higher when the heating element is cold and decreases until working temperature, when generated heat (power) equals dissipated heat. "electrical resistivity of metals increases with temperature" (

pbates123 (author)martins9992011-06-05

Actually Martins999 the temperature coefficient of a resistor depends on the material it is made of. As an example aluminum, known as a conductor does have resistance which becomes less or decreases as temperature increases. Resistors can have either a positive or negative temperature coefficient. The wiki most likely refers to carbon resistors.

martins999 (author)pbates1232011-06-05

Yes, there are materials that show a negative thermal coefficient (NTCs are used in electronics) but this is clearly not this case

autoflow (author)2009-12-28

 Can you go into a little more detail on the wiring of these extra parts that stop it from melting itself. I like the dimmer part to get it balanced but im not sure how to achieve this.

quickshadow_2 (author)autoflow2009-12-28

Ok, it is pretty simple, just go out an buy a light dimmer made for indecent light bulbs, it should look like a box with a knob on it, then wire it in series with the heating element, the light dimmer should come with instructions on how to do this (with lights that is, it's the same thing, just copy). To adjust the amount of power you want to feed into the heating element, just turn the knob, but a word of caution, DO NOT USE WITH ANY FORM OF COMPACT OR NORMAL FLUORESCENT LAMPS, as it will cause them to flicker and burn out, in due course possibly releasing the poisonous mercury contained within them, they have special dimmers for fluorescent lamps, please do take note.

thebriguy (author)quickshadow_22011-05-04

and special CFL Bulbs. Standard CFL's cannot be dimmed. The dimmable CFL's are only okay at best compared to incandescents (they do not dim as low and are very touchy, and I swear they burn out pretty fast.) I have not come accross any CFL that lasts as long as the packaging says it will. I get a 2-3 years at most and as little as 6 months. This is on bulbs rated at 7 to 10 years(!)
Sorry I got off track...

autoflow (author)quickshadow_22009-12-29

 do you use a 110 or 220 heating element. Also do you prep your cans at all. (crush or shred them) before you melt them? Ive found it real hard to melt a can in an empty container.

quickshadow_2 (author)autoflow2009-12-29

heating elements are rated in Watts not Volts, as P=V*I and V=R*I, thus if you have more voltage then there would have to be more current for the same amount of resistance (R= the heating element), in so generating more heat or power, in short all heating elements will work with all voltages, just that the heat generated by each will differ

Also, i find it easier to melt cans down if i shred them up 1st, guess i forgot to add that point down.

temper (author)2010-01-17

I had to vote you low because your title is a total misnomer. Virtually nothing in this instructable is likely to be lying around, and in fact the parts tend to be quite expensive.

That's a pity really because otherwise it is quite a good instructible.

tulekah (author)temper2010-06-23

drive down alleys and dumpster dive. i've make a shop heater out of the elements from trash electric stove, and also the rheostat from the same stove. "trash is a failure of imagination" p.s. put a "wanted" out on FREECYCLE!!!

thebriguy (author)tulekah2011-05-04

Agreed - there's materials EVERYWHERE for FREE. Just post a "wanted" on Craig's List and people are happy for you to take away what they consider "crap" - but what are excellent parts materials!

MillenniumMan (author)temper2010-03-05

I have my own hauling business, and run into these parts frequently. Not only could I build this for nothing, I could build several of these for nothing.

Right now as a matter of fact, I have a broken electrical oven, a 20qt stainless steel stewing pot and a sack of quickcrete sitting in the back of my truck from a cleanout I just did. A little work and those parts could make a large aluminium foundry. And since I run into much aluminium in my line of work, maybe I'll give this a go.

orksecurity (author)2011-01-30

Would have been nice to have a few "prover" photos of the widget in use.

Kevanf1 (author)2011-01-15

Use fire clay instead of concrete. It won't blister and there is no danger of it exploding. Though, concrete will not explode as long as it is not 'green'. However, having said that, it can actually take concrete 10 yrs or more to actually properly dry and set. I know, I used to manufacture concrete products for a living.

Bloody good 'ible' anyway I may give this a try.

tulekah (author)2010-06-23

could you bury the coil in sand in a big ceramic pot and make your crucible compartment with a thin layer of fire clay or some such?

About This Instructable




Add instructable to: