Enter Inferno

Each spring, Baltimore Knife and Sword Co hosts their fabled Fire and Brimstone hammer-in. This past year, we made steel in just about every way I can think of (short of blister, shear, and brescian steel). Returning to me roots, albeit short ones, in steel making, the following details two hearth furnaces built over the weekend by two of the best authorities on the subject, Zeb Deming and Mark Green. 

Zeb's furnace was built out of firebrick on a steel tabletop, about as simple as I have ever seen and staggeringly effective. The tuyere he brought has a special fitting on the end that allows you to see through the pipe into the hearth, which is quite a sight (and difficult to capture the molten metal on camera). The air supply is ingeniously fed by a small vacuum and a dimmer switch, affording a good deal of precision. And there is a gate valve (just barely visible on the left side) to release excess flow.

Building the furnace Saturday morning with a bucket of cobb, wire, and brick as the day began to warm was the first part in an exciting weekend of melting and smelting. After leaving behind snow up north, Baltimore was a nice reprieve when the temperature rose above 50.

As with all furnaces, the entry angle and height of the tuyere will play an important part in the qualities of metal produced from the melt. Inside, the pipe enters about 1.5~2 inches from the top of the table, which has been insulated with a cake of ash (if I remember correctly).

Burning handfuls of charcoal throws so many sparks out of the hearth it is like a firework display. At night, the show is even better!

In Zeb's hearth a good amount of steel was made on Saturday. Above, he is running wrought iron from a wagon wheel through, carburizing it into steel. Inside a hearth furnace, there are three general zones, carburizing, decarburizing, and neutral. Lee Sauder, another expert in the field, offers a much better overview than I can, which can be found here- Aristotle's Steel.

Here is a view through the tuyere midway through the day's second melt. In this run, pure iron is being carburized into steel, originating as a 00Fe bar, meaning that the carbon content is <0.01% carbon.

The resulting puck came out solid and easy to work compared to conventional bloom, which is smelted from iron ore amongst other things. Above, Zeb is hammering the carburized pure iron, with the wagon wheel wrought on the right of the stone.

Also run through Zeb's furnace were a collection of forsaken blades of another smith who wanted to recycle them from their either broken or undesired form into hearth steel so he might then forge another knife from them.

Mark Green brought his Evenstad Furnace, which has a few notable differences than Zeb's. It is much shallower and wider, allowing for an easier examination of what is being melted. Control of where the metal enters the charcoal is a little more precise, and longer pieces do not need to be fed end up into the flames.

Into Mark's furnace went a collection of bloom steels from his other smelts, using the different zones to carburize the low carbon steels and decarburize the cast irons. The amount of control and predictability is astonishing.

I missed the first run Mark did on Friday afternoon, coming in at the very tail end. He ran the furnace using a double pair of bellows instead of an air pump, and was just as successful. For the rest of the weekend, however, electricity was used for sake of convenience.

Here, Mark is consolidating decarburized cast iron.

And finally, spark testing one of the day's many hearth melts.

In my continued exposure to the steel making process, and being able to talk with people far more knowledgeable in the subject than myself, I have been able to start experimenting in the processes on my own, and soon I will be using it to make blade steel of my own.

Greensand

There are various methods that can be used to cast metals, ranging from plaster investments with vacuum belljars to what is called 'greensand casting'. The latter can be made either with water, which as I understand it makes a true greensand, and oil. Water based greensand is relatively easy to produce, although requires some ingredients that may be difficult to find if you do not look in the right place.

For my greensand mixture, I used three things (in addition to water). First, and most obvious, is sand. I gathered mine from a beach a year or so back with some measure of foresight when I was in North Carolina. It is a relatively fine sand, the finer the better until you reach the consistency of flour (ask me how I know...). This can be found as playsand from most hardware stores, but needs to be sifted of all the coarse grains. Fine weave screens work well for this. Second is silica sand. This is not necessary if you cannot find it, which as I understand can be laborious and expensive, especially in small quantities. I believe you can find this at pottery suppliers or speciality art stores, but I got mine from a friend in the arts business. The silica sand I have is extremely fine (you don't want to breathe in the dust from this stuff). Finally, southern bentonite clay. This is what allows the sand to stay together and take a fine pattern from whatever you are casting. I ordered mine from a supplier online a while back, just enough to fill a flatrate shipping box. It was not too expensive if I remember correctly. Another place you can find it, so I have heard, is by grinding up certain types of cat litter.

All the recipes I have seen for greensand measure the quantities by weight, but since I did not have a scale, I did it by volume. It took some adjusting before getting right, so here is what I used-

5 cups fine sand
3/4 cups bentonite clay
1/2 cup fine silica sand
Water to taste

First I put the beach sand in my mixing tray. The larger the tray the easier this will be.

On top of that goes the silica sand (or bentonite clay, order does not matter)

And finally the bentonite clay. Now it is extremely important that you thoroughly mix the three things (or two if you do not have silica sand). As even as possible will yield the best result. The bentonite clay is what really absorbs the water, and if it is not mixed well there will be dry pockets and clumps. No good.

Next is the most important part to get right, adding the water. The fist time I tried this, it was a disaster. I added water from a measuring cup and it was near impossible to even out in the mixture. So this time, I used the spray function of an iron. A squirt bottle would be just as good (ideal) but I did not have one lying around and this was the best I could do. A fine, even misting to dampen the surface of the sand is all you need. GO SLOW! It is very easy to over saturate the mixture, and once that happens it is extremely difficult and tedious to fix. Once the surface of the sand is dampened, mix it around until the entire body of the sand is uniform again. It may be necessary to cut the mixture and fold it over itself.

Repeat the process of dampening and mixing until you can take the sand in your hand and clump it into a ball. The cluster should retain its shape while not leaving sand stuck all over your fingers (it is on mine because I did not dust them off after mixing). The cluster should also not crumble when you break it. As shown above, it should break cleanly and not disintegrate. You may need to dampen the sand again if you store it for a long period of time, but this mixture should keep for a very long time in a sealed bag or container.

That's it! Nothing fancy or special about it, and it can be made in about five minutes once you have your ratios of ingredients figured out for your own use and preference.

More later on how to use it.

Venturi Forge Build Part IV - Lining

Continued from Part I here

                from Part II here
         and from Part III here

After a little bit of a hiatus, I finally found the time to finish this project (although I suppose it has technically been done for a few weeks now). I will preface this with the following: I am showing only the simplified version of what I did with the lining, because I went about it in a very backwards, roundabout way. Had I a little more foresight, I would have done it as I am writing about it.

First, a warning. Kaowool is a ceramic fibre insulation that SHOULD NOT BE BREATHED. When cutting, it is critical that you wear a dust mask at the least.

That being said, the majority of the insulation in the lining comes from the kaowool. I used two layers of 1" blanket for a total of- difficult math here- 2".

Measuring the width and guessing the length (circumference, didn't have a flexible tapemeasure) worked just about perfectly. For the second layer, I left out enough room for the firebrick that I will use as the floor. This is for two reasons. First, it is more wear resistant, and second, flux tends to annihilate kaowool.

Notice that the forge housing is a little longer than the length of the two firebricks placed end to end. That would have been easily avoidable had I thought of that before cutting the housing. It still would have been fixable, but that would have meant cutting half that length off both sides, which with only a pair of tin snips, I was not about to do.

Next came the door. This is where the problems began. I should have, and I realized this far too late, cut the door like a trapezoid with the taper narrowing the hole the farther it went in. That would have prevented many of the sealing issues later on, but the best solution would to just not have a door at all. At least, not for a welding forge.

This is an in progress shot of the door to show how I attached the kaowool to the metal. There are wires pierced through the pieces of insulation, then fitted in the cracks between the door and the flange I riveted to it. Although not the best solution, it was all I could come up with after some deliberation. I also considered using furnace cement, but I do not think that would have worked well as the curing process would likely have lifted it apart (and it does not bond well to the raw kaowool).

The first covering over the kaowool is satanite. The purpose of this is to seal the kaowool, better the insulation, and serve as a base for the subsequent treatments. Below are the directions from the distributor-

Mix the Satanite to a thick paste...just keep adding water slowly until you get a pasty consistency that you can paint on with a paintbrush....roughly the consistency of sour cream. Spray the ceramic fibre insulation down using water with a hand sprayer to wet it lightly. Next, apply the Satanite to the wool using a paintbrush, covering all exposed wool surfaces. To cure it, you want to dry it slowly. First, let the forge sit for a few hours minimum to air dry a little, then fire up the forge just briefly and shut it down. Do this several times, allowing it to cool down in between and increasing the on-time with each subsequent cycle. You'll see water vapour evaporating the first few times you do this. Finally, fire it up and bring it up to full temp to fully cure it. You will probably want to apply at least two coats of Satanite in this manner...it's a little time consuming (do it over a couple of day period) but makes for a more robust coating. a 1/4" layer is a good thickness to shoot for. If you are going to apply ITC-100 over top of the Satanite, be sure to fully cure the Satanite first.

Above is only the first coat of several.

Over the satanite I used ITC-100. This is an infared refractory which supposedly increases the thermal efficiency of the forge by a significant amount (upwards of 30%). Again, below is the recommended application-

How do I mix and apply ITC-100?
For ITC-100, the manufacturer recommends to mix it 2:1, so if you have a pint, mix it with a half pint of water. My experience, indicates that mixing it a little thinner is just as good if you are using Satanite as a basecoat first. Since you're using the Satanite as a protective coating, the ITC-100 doesn't need to serve this function. Mix it thin, and apply the coats evenly. Applying several thin coats is better than applying a single thick coat. You'll likely have some left over for future patching. Apply the ITC-100 over the Satanite only after the Satanite is fully cured. You can use ITC-100 alone without first applying Satanite, you will just need to use more of this material.
How much area will ITC-100 cover in a forge?
ITC-100 will cover 6 to 12 square feet per pint, or 3 to 6 square feet per half pint. If you apply a basecoat of Satanite to your forge first, you can get by with the larger number for square feet coverage. An additional benefit to doing this with Satanite first, is that Satanite is cheap and by building up a 1/4" layer of Satanite over you Inswool liner before applying the ITC-100 your forge will be more robust.

All the steps in this process require the coatings to be cured, by gently heating the forge over several hours (which quickly turns into days). This is after the ITC-100 has been fully cured.

Finally I mixed some bubbble alumina to serve as a flux resistor on the edges where lining meets firebrick and around the door opening. I did not think the stuff would actually have bubbles in it, but it does. They are the rigid sort of bubble almost like an empty eggshell.

Mix the Bubble alumina to a troweling consistency and trowel it in place onto dampened Inswool ceramic fibre blanket Let it dry for a few hours and then slowly fire up your forge to fully cure it. This is one of the most flux resistant coatings we have found.

I originally intended to use this on the firebrick too, but later decided against it due to the difficulty in making the surface flat, which is important (to me) while heat treating thin blades.

To keep the alumina from seeping over the edges of the firebrick, I folded some paper and made a barrier to keep it in place while it set.

And what it looked like after. As I said, it was very difficult to keep neat.

Finally, it was time to seal off the ends, the back one here. While planning the design, I intended to forge a rim around the edges of the doors, but abandoned that after cutting the sheets with the tin snips again. To keep the kaowool in place, I bent the material from the opening upward. That flange will slide inside a layer of the wool.

In the front (and back, later), I used the 1/8"x2" rectangular bar I intended to reinforce the burners, but did not need, to guide the flame out of the forge body so it would not creep up the flat surface. I bound it with rutland's furnace cement, but some sort of mechanical reinforcement is recommended (rivets, bolts, etc.)

After the two ends were screwed in place, which was a trial not having a drill to use, I filled finished the last of the insulation. With the scraps from the initial lining, I cut strips and fit them in place, finally sealing it using the same process as the main body. (This one is an in progress picture and is not actually that sloppy).

That's it! After doing a stress test, it easily reached 2200 degrees (F) at 10psi. I had to take out the thermocouple before melting it, but since I have done a little welding and it runs great. I did make a few buffers to reduce the chamber volume, but no pictures of those at the moment. For use with the door, however, they need to be removed.

Feel free to ask any questions in the comments.

Cheers!