Furnace (part III)

(Lost foam casting)

This project has been divided into sections. Navigate through them using the links at the bottom of the page.


Alas, upon the first firing of the furnace, disaster struck! My propensity and unswayed determination to use soup cans as crucibles finally fought back. Not realising that the crucible had developed a leak during the melt, I kept adding more and more aluminium thinking.... "wow this crucible can certainly take a lot of metal...".

A low yield from the subsequent pour hinted that something was drastically wrong. Next came the stench of burning PVC. Luckily, I had employed a segment of foam between the PVC and blower, which acted as a 'fuse' and let the aluminium spill before entering the blower.

Mr. PVC says "I give up, you win..."

The foam block used for the first casting by the new furnace with sprue attached

After repairing the blower pipe, the furnace was fired up for the second time. I had no ingot moulds at the time, so decided to do a quick lash up 'lost foam' casting.

Lost foam casting involves a piece of foam shaped to the desired metal shape with a sprue attached. The sprue acts as a gateway for the metal pour. The piece of foam is then stuck in fine grain sand with only the sprue visible.

When the molten aluminium is poured onto the sprue head, the foam quickly gives way and the liquid immediately follows before any sand can fill the gap. Once the mould is full (and all the foam is gone), the metal solidifies in the shape of the foam, supported in place by the sand.

The sprue is then cut off the casting after it has cooled.

The foam piece held in place by sand (albeit not fine-grain sand)

The first casting - not great, but not too bad

Here is the result of the first attempt at lost foam casting. I had tried to engrave "Penguin's Lab" into the foam, but this did not turn out very well.

The cavities in the casting are due to bits of slag which happened to find their way into the pouring process. I later concluded that the aluminium was being poured too cold, and that the furnace did not meet the temperatures required to fully melt the aluminium.

The fault was identified at the bottom of the furnace. Where the air distributor contacted the crucible, the floor of the crucible was being cooled by the incoming air. The arrangement was such that the two would be touching, as the soup cans were too high to allow for any coals inbetween. This was easily fixed by cutting down the crucible height.

Another improvement was made through the introduction of a new blower - this time in the form of a hairdryer. I removed the heating coil in the dryer, and rewired it to run on 12VDC - the specs of the motor. This sucker pushes through so much more air than the last blower.

The new improved blower system, complete with 240VAC - 12VDC transformer

A few kilos worth of aluminium scrap! Beautiful.

While the improvements were being made, I also abused some old circuit boards I found to scavenge some very nice aluminium boards, heatsinks, RF shields and chassi.

As most of these pieces are too large to fit in the crucible, I usually dump them in the furnace after a casting session, then perform a crude hammer, fatigue, and rip strategy when the aluminium is close to its melting temperature. At this temperature, aluminium is very brittle and is easily smashed into smaller pieces.

You can easily tell when the aluminium is getting soft by hitting it with a hammer. If it still 'rings' like usual, then its still cold. If it sounds 'dull' and rather unresponsive, then you know its nice and hot.

Here is the new furnace arrangement being fired up, and it sure is firey - the crucible is glowing bright orange and flames are being sent whipping up through the vent hole.

The loaded crucible glowing orange-hot

Another piece of foam ready for casting

Here is a new foam mould ready to be packed in sand and cast upon! The new furnace arrangement is a LOT more efficient than previously, and takes roughly 15 minutes to completely transform room temperature aluminium into its liquid form!

I have also found that there is considerably less slag output and no aluminium sticks to the bottom of the crucible after pouring.

Here is a crucible load of aluminium being poured into the foam mould pictured above. When pouring lost foam castings, it is important to keep pouring when the foam bursts into flames, as this is when the liquid must keep up with the rapidly dissappearing foam before the sand moves at all.

This shot is a still capture from a video, since I didn't want to waste time trying to get the self timer to go off at the right moment. However, I might have to do this in the future, as the resolution from these videos is really quite baaaddddd....

Molten aluminium being poured into a foam mould using the lost foam casting process

Some castings produced from the furnace

Finally, we have here two more "acceptable" castings. The right most one is the result of the pour pictured above. I tried engraving my surname "Peng" into the foam at the last minute, but unfortunately I did not pour the aluminium fast enough for this mould, as a large chunk of it was consumed by sand. The sprue hasn't yet been removed, and the casting is upside down.

The left most casting is quite a nice one. Although I had no idea what I was making (only that it was something vaguely geometrical), the aluminium filled the mould perfectly. The sprue has been cut off this one. Unfortunately I haven't found a use for any of these castings yet. Maybe I should cast something useful in the future...

Next, I try my hand out at greensand casting too- another casting method which involves a sand and clay mixture which is shaped into the desired mould. The main advantage of this method is that you can easily produce multiple castings, whereas the lost foam process sacrifices a load of sweat, tears (if badly performed), and of course foam everytime a casting is performed.

More random lost foam castings



Next - Furnace (part IV)

Previous - Furnace (part II)


Back to Penguin's Lab


© Penguin's Lab 2012