When creating a homemade refractory mix, the common suggestion is to add sawdust or foam beads to the clay and other materials. The result is a lighter, less dense refractory. As the refractory is heated up, the organic material burns away, leaving voids in the heat-resistant material. These voids transfer heat much less efficiently, increasing the insulative value of the refractory. The net result is that more of the heat stays in the crucible chamber and less leaks to the outside. For a foundry or forge, that’s a wonderful thing… we’re not buying propane to cause global warming!
The more heat that stays in the chamber, the faster the piece gets up to melting and forging temps which results in much more efficient operation. Theoretically, with perfect insulation a candle could melt iron (eventually), but there is no perfect insulator. As we add heat from the burner, heat is being lost through the shell as well as escaping out the exhaust, which is a shame, but a necessary evil unfortunately. Still, we want an insulator as good as we can get so that we can add heat significantly faster than it leaves, so that the temperature can keep increasing. As the temperature increases, so does the rate of heat loss, so ultimately good insulation is the key to high temperatures.
To this end, most hobbyists skip the refractory entirely and go strictly with a ceramic wool material that has VERY high insulation value and brings things up to heat quickly. The downside of this is mainly durability. Physically they can’t take much abuse, and welding fluxes and glass will destroy it quickly. That doesn’t make them bad, it just means that you need to be careful. You’d be well-advised to coat the material with Satanite and ITC-100 to give it some substance and resistance to flux. Putting a dense board on the bottom isn’t a bad idea either.
Why is the ceramic wool (kaowool, inswool, etc) so desirable? For the 8lb version at 2000 degrees F, it has a conductivity somewhere in the range of 1.8 Btu-in/hr-F-ft^2, while insulated casting 2600 is about 4.0 Btu-in/hr-F-ft^2, insulating castable 3000 is about 4.54 Btu-in/hr-F-ft^2, and NON-insulating castable has very poor insulative qualities at 7.4 Btu-in/hr-F-ft^2. Yikes!
That said, I want something that will last a lifetime and withstand the higher temperatures that I’m shooting for. Ceramic wool seems to top out at 2700 degrees F, while I’m looking for a little bit more. So, regardless of how great ceramic wool is I’m dealing with castable refractory as part of the solution. There are many brand names, but ultimately you’re looking at different temperature grades and insulating (eg. Kast-o-lite) vs non-insulating (eg. Mizzou) refractory. The insulating refractory is much less dense (90 lb/ft^3 vs 141 lb/ft^3) as you’d expect from our earlier discussion.
A common trick is to buy the non-insulating castable and add a healthy amount of the foam beads to it to turn Mizzou into an insulating castable refractory to make your money go farther. It also provides flexibility in that you can use the non-insulating castable in other ways in which the insulation would be a detriment, such as making your own crucibles. Some people just use furnace cement.
The web is strangely silent on whether it’s possible or desirable to add a smaller amount of insulation to already-insulated castable. I’ve decided to find out by mixing a small amount to be used as the blade-port plugs. If it turns out badly, then I’ll have not wasted a great amount of my refractory. If it turns out well, I’ll use the technique throughout the rest of my furnace to reduce weight and improve efficiency.
So first, we need the insulative burn-out material. Sawdust has been the traditional material in Firebrick. It’s harder to come by for some than others. I do a fair amount of woodworking, so not so hard. Still, that’s a LOT of sawdust. The other common suggestion are the foam beads used to put into bean-bags. The advantage is no prep.. just buy a bag and go!
I’m a bit leery of the size of the voids if they’re bead size though. I figured there had to be a better way, and think I found it.
The S-shaped expanded polystyrene packing “peanuts” are basically just waste. You get a ton of them shipped to you when buying things and ultimately they just go in a landfill someplace. Re-using them is not only cheap but environmentally friendly. But the S is definitely too big if the beads are too big, right? Those particular peanuts are inside a Food Processor. Add a little water and blend!
The end result is significantly finer (on average) than the foam beads, which should lend itself to a higher material strength. I added in my sawdust and I’ve got about a 1:2 ratio of sawdust:foam ready to go.
A quick note of warning! I also tried using the 8 peanuts in the food processor… that was a huge mistake to be avoided. Those particular peanuts don’t cut and crumble into smaller pieces, they merely decompress and too-often get between the blades and the walls of the food processor, leaving a plastic streak along the inside wall. Initially I thought I had permanently marred the processor, but the excess plastic was easily removed with a little elbow grease.
I mixed up the Kast-o-lite with the PeanutDust in about a 3:2 ratio and poured it into my molds. They’re curing now… getting more solid all the time. On the underside (ie. the outside) of the plugs I inserted a small hook in each one prior to pouring the refractory. This will give me something to grab on to when I need to remove the port for work.
Now we just have to wait for it to harden to figure out if the experiment is a success of not….
Update: The net result is a qualified success. The bricks hardened up nicely where there was sufficient binder. I forgot the problem that I’d noticed when casting my flare… the Kast-o-lite had separated in shipping. The finer binding particles (dissolve in water and act like a glue) have migrated primarily to the bottom of the bag, so my mixes from the top have had an aggregate/cement imbalance.
The result is that the bottom surface where the hooks are, is SOLID. The topmost 3/8″ was crumbly as it had nothing to hold it together. I’m remedying this the same way I did with the flare… a scoop from the center of the refractory and shaking it in a tray to separate the lighter big aggregate particles from the lighter ones (that include the binder.) Add some water and use it to dab on top to fill in the pores until it seems “wet.”
If I hadn’t had the same problem with a non-PeanutDust batch I’d say that the sawdust had caused it, but I know better. It looks like the particles have mixed all the way through. I can see bits of foam in the lowest levels, so it didn’t separate. Overall, a win. I think I’ll be dropping the ratio down to 2 parts Kast-o-lite to 1 part PeanutDust going forward though for the more structurally important parts of the casting, and adding the PeanutDust wet instead of dry so that the sawdust doesn’t absorb any of the liquids.