This is an old revision of the document!
“If you're gonna' play with fire, you're gonna' get burned.”
Easily one of our most fun, most dangerous, and least productive projects ever was casting. During the several weeks we attempted this, we learned a lot about what to do and especially what not to do. No joke though - molten lead and aluminium is incredibly deadly. For one thing, there is an absurd amount of energy pent up in liquid metal - enough to cause the water in solid concrete to boil out and explode. Likewise, molten metals let off all sorts of vapors - the kinds that give people cancer and gross smelling clothing. Seriously, not once did we get compliments on how great we smelled when we were casting aluminium.
If you pursue any of what we did, spend the money on good equipment. If we had the money we would have done a much better job.
The most important part of any casting process is the furnace. This may not be objectively true, but it's what we found to be the most important. The furnace turns cold solid metal into incredibly hot liquid metal. The furnace we designed was incredibly simple, insanely cheap, unstable, and a whole slew of other negative adjectives. In short, it worked. But… when it didn't work, it really didn't work.
As you can see, it's sort of grungy. In case the setup is not clear, the system is made up of two paint cans stacked on top of a coffee can held up by two cinder blocks. We could say these cinder blocks were there for their insulatory properties, but we'd lying. In the early days those cinder blocks were bricks. My, how times change. In this particular image, the tray of brown dirt on top is supposed to be “greensand.” Greensand is used in casting to form a negative into which metal is cast. Our “greensand” never worked properly, and only ever smelled horrible. That's why I put it in quotes.
If you look closely at the image to the left, you can see that the crucible being used is actually a steel soup can. These actually worked pretty well until the furnace got “too hot” and the steel cans melted then disintegrated. The melting point of steel is about twice that of aluminium, so while steel crucibles worked they were incredibly weak when actually used. These had a success rate of less than 50%, and were eventually replaced with a proper ceramic crucible. Another interesting thing to note about this photograph is the fuel being used. This is standard Kingsford brand charcoal. It worked very well in our furnace because the briquettes were consistent in their shape and held up nicely when being burned. These two things were great because they meant the crucible was very firmly held in place and received a lot of contact with enough heat to be effective. Unfortunately, though, due to the nature of this type of charcoal a lot of ash was generated - even when burned at such high temperatures. This ash makes the work area quiet messy, and can eventually clog up the fan.
This is, I believe, the most complex object we ever cast. When you look at something like this you may be surprised to learn this is more or less the same technique used to cast Ferrari engine blocks1).
- A small(ish?) coffee can with a large hole/opening in the side so that a small hairdryer can be positioned to blow air in through the side. Aluminium tape was used to seal space around the dryer, permitting as much positive pressure as possible. We initially had the hair dryer standing up vertically under the entire rig, but very quickly realized this is a bad idea; when our earliest (steel can) crucibles failed, the molten metal had a tendency to drip straight down and, as you might imagine, flow into the hair dryer - and trust us, molten metal can make very short work of a $7 hairdryer. Luckily we only had one hair dryer ruined this way and quickly restructured the rig so it wouldn't happen again.2)
- A normal steel paint can with a hole cut in the bottom, made to mate with the coffee can. This acted as an adapter and an ash trap.
- Another paint can, this one with many holes punched in the bottom. We used a large nail and making holes about 1/8“ in diameter. We probably punched around 20 of these in various radial patterns. The holes needed to be big enough to allow for ash to regularly sift down, but small enough that no pieces of still-burning charcoal would make their way down. Too few holes would not allow enough air to get through at regular intervals, while two many would cause the bottom to lose its structural integrity - and remember, it needed to be able to hold the weight of the charcoal, the crucible, and all the metal being melted - all while being heated to extreme temperatures.
This top can, the furnace itself, did not last very long at all. In fact, we had to replace them after what was likely less than two hours of full-heat operation. The cans we got new from Lowes (at about $5) started out very shiny and pretty. The inside was lined with some sort of plastic. After a few minutes of the fan blowing and the coals coming up to temperature, the can lost all of its previous luster and was left a dull dark brown. The heat also accelerated oxidation, so when it was humid or raining (almost always) the walls of the can would rust like you wouldn't believe, and enormous holes would form. These holes caused three things to happen: air pressure and heat would leak out, coals would drop out, and the overall structural integrity of the can would drop.. thus making it that much more dangerous to operate.
The three cans stood around two feet tall, and were supported on two sides with cinderblocks, as mentioned above. Starting the furnace was very simple, though it took a long time to get it hot enough be an effective furnace. First a bed of charcoal (sometimes hardwood, sometimes briquettes) and crumpled up newspaper was laid across the bottom in two or three layers. We lit the newspaper with wooden matches, then turned on the hairdryer. This created a positive pressure chamber in the middle paint can, and air was forced upwards into the furnace itself. For all intents and purposes, it was a charcoal grill with a plain-air injector. After a few minutes the newspaper would be nothing but ashes (ashes that were then blown everywhere) and the coals would start to burn on their own.
Once we were sure the coals were lit, we nestled the crucible into the coals, trying to get it as centered as possible. One of us would hold the crucible upright with a pair of pliers or tongs while the other would evenly stuff in fuel around it, so that when released the crucible would stay perfectly upright and not tip one way or another, even when loaded with half a kilo of metal. This was one of the trickier procedures for two reasons. Not only was the furnace consuming fuel at an incredible rate, causing the older pieces of charcoal to crumble and fall, but due to the air being injected in from the below meant there was a deadly column of hot air and flames blasting out from the top of the furnace, so our hands were likely to be burned. Gloves were helpful.
At this point all we had to do was load the crucible and wait for the metal to melt. A few times we used lead, but we grew to prefer aluminum. We had wanted to use lead all the time, because of the lower melting point, but it was hard to find lead and we had an essentially bottomless stockpile of aluminum - so aluminum we used.
I'm going to take a moment to quickly describe our environmental conditions, because I think they're very important for understanding this endeavor as we experienced it. We did all of this in the Winter of 2010. Now, a winter in eastern North Carolina isn't much, I understand that, but still, it's something we weren't fully prepared to deal with when it comes to casting aluminium out of a homemade furnace. For starters, the air temp (I'm going to be very generous and say it was consistently in the 40s, sometimes dipping as low as the mid 30s) kept us from doing quality work, because our legs and arms and hands were cold - then we would dead with these bursts of astoundingly hot furnace which was generally uncomfortable. I don't think the furnace had much trouble because of the air temperature, but my oh my, is it ever something we complained about.
Of course, had we done any of this mess in the summer (like we did with forging) we would have been absolutely miserable. We ran the furnace outside of Lea's garage and when it was burning full tilt there was probably a 10ft radius around the thing where you could be significantly warmer. It was also raining constantly. Snow would have been fine, but it didn't. It rained. This kept the concrete nice and damp, so should we spill any liquid metal, we would inevitably be greeted with an explosion of metal, steam, and concrete. Kept us on our toes, though.
Melting 400g of aluminium to completion typically took 20-30 minutes in our furnace. Which was, as stated before, insanely inefficient. I would suggest no more than 30% of the energy put into that thing ever made it to the metal, most of it probably went straight up or out, through the thin steel walls. That said, it did melt the aluminium eventually, and in those 30 minutes of waiting we embraced our freedom, mostly lingering two or three minutes before loading the furnace with more fuel. In those two or three minutes we debated about whether or not to put a lid onto the furnace, how to make the walls more efficient, whether or not the rods of still-solid aluminium sticking out of the crucible were radiating off extra heat that would otherwise be used to melt said aluminium, and how to make molds - and of course, what to make molds of.
In short, it was a very lively time.
I think it's worth noting that we did actually have plans for either a ceramic or concrete furnace that would have incredibly good insulation. Fueled by propane, too. So powerful, so efficient… Why, exactly, we never made such a furnace I don't know.
Money. Money was always the problem. Always has been.
Once we had the molten metal, we had to do something with it. We pursued two types of casting for a few weeks and had semi-decent results at best.3)
One of the hardest parts of handling molten metal is that it's really, really, hot. Like, indescribably hot. We had a proper handle for the crucible we inevitably bought, but because of the way the crucible had to be lowered nestled into the coals it was impossible to use it properly. What we usually did was use a pair of channel lock pliers to raise the crucible, by the lip, about 10” out of the furnace. This was an astoundingly challenging process simply because of the sheer weight of what you were lifting, and the fact that it was literally hot enough to melt solid metal. The other guy would hold the proper handle thing a short distance away, and as soon as the crucible was clear the coals he had to swoop in and grab it from the pliers, which would inevitably be slipping at this point. Then, the guy with the crucible-holder and crucible would have to get to the mold as quickly as possible and dump the metal in there before it cooled too much and hardened.
This was one of the first casting techniques we attempted, and it mostly ended in abject failure. We first attempted to make our own greensand, mixing ground cat litter (for the clay) and sand. We then built a medium (1' x 1' x 1') wooden box to hold the sand. This was such a failure, the making of all of this won't even be documented here, it's not worth it.
We also bought greensand, which was of much higher quality. This also required a box to hold the sand mold, however it was much smaller, around 1' x 6' x 3'. It was with this that we almost succeeded in casting our aluminum cannon, except we overlooked the fact that aluminum could make the cope (top part of the mold) float. So with our cannon almost cast, seeing the aluminum fill up the mold, suddenly the mold lifts and aluminum pours out the sides. Very discouraging, especially as that was probably our 7th attempt with the (bought) greensand mold method, and each attempt took around 1hr, and much charcoal. This was our surrender point.
Plaster casting worked fairly well, though it was very dangerous. When cured plaster comes in contact with molten metal, the heat of the metal causes the water that was otherwise sealed in the plaster to vaporize and try to escape under high pressure. This happened so quickly and with so much violence that the aluminium often was spewed out of the cast's overflow ports in a manner not unlike that of a violently erupting volcano. Needless to say, it was pretty dangerous.
Despite this, we were able to successfully cast a few things. The precision of the casts was diminished because of the plaster walls being so greatly pitted, but it worked for the most part. Supposedly adding talc to plaster is a good way to strengthen it and displace what space would have been occupied by water, but we were never able to find a good source of talc so we never tested this.