This may be image-heavy or long. I’ve been busy with the thing, and it’s turned out/turning out nicely.

I haven’t been subjecting you all to every little step, and often when building something steps get dragged out across days or weeks. So it’s accurate to say that I’ve been working on this more or less constantly in the last month, but I haven’t actually done very much. Assembling something like this is, by design, a process that is a simple matter of adding pieces. Sometimes, adding a new piece takes just a minute. The problem is that sometimes those pieces take a few steps to fabricate, etc. Then I have to consider the design principles I don’t want to deviate far from: minimalism, maximum ease of assembly, utility, minimization of tooling or customization, etc. Up until this episode I have resisted the temptation to just grab my welder and start sticking things together. Instead of welds, I put a lot of thought (I don’t know if anyone can tell) into basically making the whole thing assemble with 4 wingnuts. That actually delayed things because sometimes I had to think a lot before taking the next step, and a big piece of that style of design is that you have to be extremely careful the whole time not to design yourself into a corner.

One part I did not undertake, myself, was the inner floor liner. See below:

The subfloor, if you will, is hard firebricks sitting between two more walls of hard firebricks. But flux and heat eat hard firebrick fairly quickly, so there is a bottom liner of kaowool felt, that holds in place an inconel tray. The kaowool felt is expected to keep things from rattling, but also to expand and contract with heat, instead of cracking the sidewalls. I don’t have any practical evidence, yet, but I think a lot of forges eventually fall apart because the materials don’t have anywhere to go, and just crumble. We’ll see.

Also, you’ll probably notice that things in the picture above aren’t very straight. The whole setup is designed to be aligned and leveled with the nuts at the bottom of the allthread, which serve to set the height and alignment for the whole thing. That also helps give it plenty of air-space underneath to keep it from setting fire to anything nearby.

The inconel floor was an interesting problem. I put a piece in a vise, heated it to dark cherry, and whacked it with a hammer. Normal steel would have bent a bit, but the inconel was springy. I took it up to orange heat and then it bent and stayed bent. But if I bent it myself, that long a piece, by heating and hammering, it was not going to be particularly straight. Also, because of how it’s positioned in the bottom, it serves to lock the sides in against the L-bracket bottom sides. I pondered making a handy sheet metal bender out of some square tube, but I frankly didn’t want to deal with the slight potential of something orange hot and springy bouncing around. So, I suppressed my ego and went in to TD Fabricators [td] in Clearfield, and asked the patient guy at the desk what he thought, telling him clearly that I was concerned that the inconel might be tough stuff. He said he was not very concerned because they have a genuine badass metal bender.

That is a 250-ton Accur press/brake and it’s rated to handle 9mm thick material for bends up to 3m. [In honor of the recent burst of US nationalism I am going to try to stop using freedom units of measurement]. Apparently the bender handled things just fine and very quickly because they didn’t charge me very much, at all. I have to say – I have been taking weird projects to TD for about 15 years, now, and they’re incredibly patient and have always done a great job. I usually come in with some kind of whack-ass sketch that’s not to scale and doesn’t have proper machinist measures on it, and wave my hands, and they make it. Since I’ve more or less kitted my operation out, I usually only darken their door when I need an oxygen tank or shielding gas or something else I probably should not have. At any rate, the tray fit machinist perfect and then I had to deal with some thoughts about interior heat management. The tray and kaowool felt insulation will take care of the bottom of the forge, but what about the top? My first thought was to just cover the top by putting insulation and firebrick on it, and calling it a day. But, that raises the problem that the L-brackets would transmit heat out of the interior and would also become a red glowing menace on the outside of the forge. That is a real concern. Perhaps you may remember my previous forge had an elaborate heat-shield/burner mount to keep any radiated heat from reaching the propane distribution hoses. I needed to solve the same problem.

That kaowool felt is amazing (more on that later) so I thought that I’d just cut two long strips and fold one end so it’s held in place by the L-bracket, and the other end so it’s held in place by the hard firebrick “ceiling.” In fact, I liked that idea so much that I decided I’d do likewise with the firebricks at the ends of the forge, so the heat blasting upward will have some insulation against it. Ta-da! Problem solved. In fact, that’s basically the forge. There’s just one more problem, and that’s the burners and burner mounts. The requirements on burners and burner mounts are not too particular and they’re going to be particular to the burners that are being used. For example, I could really easily rig this forge to run on one of nobox7’s rocket burners [nobox] which generate about 2200C. [I have my own version of a waste oil burner but have not published anything about it. Should I?] So I realized, looking at it, that my design with the brick sidings makes the whole thing extremely modular. I won’t bother including a picture, but what I did was I made a wooden replica of a firebrick, slightly thicker than an original, with a slightly cut-away lip at the bottom so it would fit, and made a silicone mold from it. Then, I positioned a small bottle of sand where I wanted the hole to be, and cast a firebrick out of mizzou. The reason I did the trick with the movable bottle is because 2 of my burners have one size nozzle and the other 2 have a different size – I was able to just put different objects in different positions to get the holes where I wanted them. I can also change the location of the burner ports just by swapping bricks around. I have no idea why I did not think of that, before.

An observation: build-blogging honestly about stuff you’ve made is kind of humbling. If you just start with the finished project and go, “here’s this great thing I built!” you are concealing all the headscratching and human mistakes that you made in the process. Unless you’re Jimmy Diresta, who does not appear to make mistakes.  But I guess that is my point: there are a lot of builders on the web who post these tightly-edited build videos that make it look like they just did it and everything went together perfectly. In the case of this forge, everything did more or less go together perfectly but that was because I put a lot of thought into it and was patient while my silicone cured and the mizzou bricks hardened (one week for each!)

The mizzou cast bricks are the ones that are graphite colored because there’s graphite in ’em. Now you may be thinking the arrangement is a bit funny, but I assure you there’s nothing humorous about it, at all. The first two burners, I want close to the mouth because I’m mostly going to be making knives and the occasional sword. So most of the time I’ll want 2 burners running and the third will be turned off. The dead space between Burner 2 and 3, I am going to block off with a brick wrapped with kaowool felt, so heat doesn’t shoot out the back of the forge, and the heat shooting out the front will make the area by the door nice and toasty.

Speaking of the door, here you can see the doorway with firebricks installed on “the porch” – that was made by just extending the L-bracket forward 25cm or so, and cutting back the top of the bracket so there’s about 5cm left to stiffen it. This means the porch will be very easy to refurbish, just put another brick on it. I may put some wires underneath in case I am worried about a brick breaking and falling toward my foot while it is hot. Normally I never get my feet that near the forge (I use long-handled tongs) Another thing you may notice on the door is an experimental mounting-rail for a forge door, to help it heat faster. That’s just more L-bracket, drilled and added to the stack on the allthread. The upward curve at the end of the L-bracket is to keep the door from rolling off that end, and I could have just put a bolt through it, but for fun’s sake I welded a piece on and ground it down a bit to make the shape. The door is still “in progress” because I need some heat-resistant wheels and a frame that will carry some more kaowool felt. I can make the frame by using smaller L-bracket and screwing it together but I’ll probably weld it because why be stubborn, there is a perfectly good arc welder that lives literally within reach of the forge.

The last problem involved the burner mounts. In my last forge the mounts were tacked onto a plate attached across the top and bottom L-brackets. But I realized that’s not very elegant, really. I could just attach something to the bottom L-bracket, which is what I decided to do. I took some mild steel strap, 3cm thick, and bent it into 3 big “L” shapes, then welded them to the bottom L-bracket under each porthole. If I decide to relocate a porthole I’ll have to grind the welds, but I could have used screws. At this point, I was like a guy with a welder in hand, just looking for stuff to stick together. That gave me an idea:

The overall assembly, with everything carefully stacked between the allthread, should be able to survive being picked up and inverted. That would make my welding a bit easier, too. So I thumbed the wingnuts tight, and flipped it over. It’s pleasantly light – about 11kg. My old one weighed 1/3 of that because it was foam brick, but it probably was too fragile to invert. I don’t expect I will ever need to travel with a forge but if I do I’d be pretty willing to trust it in the back of my truck. This view also gives a good aspect on the portholes. You can see two are the same size and the third is a bit smaller. The plan is to attach the brackets, flip it back, then use an angle grinder to cut and adjust the height of the brackets, and figure out a good way of attaching the burner.

I probably should mention, in case you think I am being particularly worried about the burners: they don’t output enough reaction mass that there is much chance of them firehosing around shooting burning propane death-beams. The danger is that the burner eventually falls out of the porthole – maybe you make some propane “pop” and the pressure blows the burner out – then the burner might cannibalize its own propane hose or another and that could get exciting. I will probably do one more episode about the forge including the propane feed system hows and whys but once the burners are in, it’s ready to use.

One problem I noticed with my forges is that the burner can’t project too far into the forge or the forge’s heat will consume the end of the burner. The answer to that, obviously, is to design a pressurized burner system with a mixing plenum that dumps propane/air straight out inconel tubes… But maybe when I’m older. Instead, I wrapped the end of the burner in kaowool felt that projects a bit past the burner, so that the burner is not fully inserted into the forge body. I adjusted all that (i.e.: fiddled with it) and drilled some holes in the corners of the strap, to feed a wire through to tie everything in place. That will not, can not, slip out and shoot around.

After that, I connected up the feed lines and did a basic test firing. You don’t let it run to full heat because you’re basically driving any excess moisture off of/out of things. Since youtub seems to want to make everything like a TikTok now, I can’t embed a link, which is probably good anyway. If you want to see three burners a’burning, here.

Part 1

Part 2