Supporting a big cylinder turns out to be pretty hard.
With the boiler shell basically ready to work on, it was time to finally get down to making some parts. The legs seemed like a good place to start, because they are pretty straightforward parts and make for a good way to ease into this hefty project. Little did I know the legs would be such an odyssey! Design iterations and failures abound in this seemingly simple job, so buckle up and witness me.
The purpose of the legs is twofold. First, the boiler needs to be off the ground so that the base plate doesn’t get burned (if wood) or rob us of precious heat (if metal). An air gap underneath it acts as an insulator. Second, and more importantly, there needs to be room for fittings on the bottom. Different boiler designs will involve different fittings, but it’s a safe bet there will always be at least a drain valve. Often there is also a blowdown valve (which may be the same as the drain valve) and in our case there’s also part of the sight-glass system for monitoring the water level. In a real stationary boiler, it would be typical for it to be sitting in a built-up cradle made of bricks or cast iron. We’re going make legs because it looks nice and is easy to do (for certain values of “easy”).
The design that D.E. Johnson came up with for these legs looks quite attractive and is straightforward to make, so I started out copying what he did verbatim. His design consists of four legs, each attached by a right-angle bracket that attaches with a ferrule through the side of the boiler shell.
I deviated a bit from that design straight away, because I didn’t care for the square piece that supports the legs. It felt awkward to me amongst all the round details, so I decided to make all these parts round.
Next, we need to make the bosses that attach through the shell to those ferrules, and support the legs. As I said earlier, I decided to make them round.
Next up are the legs themselves. These are made from hex stock, mainly for appearance. Hex stock is a lot of fun to work with, and we’ll be doing a lot more of it on this project. You can get very attractive and functional parts by turning down sections of the hex to varying degrees of roundness. You also get attractive effects with chamfering and filleting tools on hexagonal profiles. The one trick with hex stock is that it is difficult to get concentricity with it. If you use the three-jaw chuck, it’s easy to hold, but you’re at the mercy of how true the stock is. Since you have to preserve part of the original surface (otherwise why would you be using hex stock), you lose the ability to “cut below the run-out”. You can set up hex stock in a four-jaw chuck, but dialing it in is really tricky, and two of the jaws are holding on corners, which is awkward and not very rigid. The ideal chuck is an independent six-jaw, but those are hardly common in weekend-warrior machine shops.
The bottom line is, don’t use hex stock for something that needs good concentricity. This is rarely a problem in any case, because if you’re using it, you’re probably making bolts, plumbing fittings, or decorative elements. None of those need crazy high concentricity, typically.
Note that parting off hex stock can be slightly alarming. It’s a very interrupted cut, and parting blades do not appreciate being treated that way. Very low RPM and gentle feed are the rule until you get deep enough to have a continuous cut. It will sound like a little baby power hammer until then.
I was pretty pleased with myself at this point, but a dark cloud was coming. Over the course of many showers and commutes, I was thinking about this leg system and decided it had a critical flaw that I really didn’t like- it requires cutting gratuitous holes in the boiler shell. This is a pressure vessel, and every braze joint that we subject it to is a potential point of failure. I’m also not a particularly good brazer-person. So why would I ask for trouble by using a leg system that demands four extra pointless holes to be made?
I couldn’t bring myself to do it. Instead, I decided to head off the ranch and do something different. All the ideas I had revolved around some sort of clamping-band system that would wrap the boiler and hold it like you might support a pipe in a ceiling. My idea was to have a continuous band over the top, to which one pair of legs attach. The bottom clamp would be two pieces, attached to the legs, but with a gap at the bottom. This gap would be bridged by a bolt that can be tightened to get the clamping action. The bottom clamp would squeeze the boiler upwards into the top clamp, securing it firmly.
I decided to reuse the ferrules that are supposed to go through the boiler, and instead braze them to the bands that wrap around the boiler. This way, the bands would sit flush all the way around, which would be quite attractive, and the legs still screw in as before.
To make the bands themselves, I fished some 24ga brass sheet off the junk pile. Now, I knew I had a challenge ahead. Sheetmetal work is one of those things where, if you have the right tools, it’s easy and pleasant. If you don’t have the right tools, it’s hell and the result always looks kinda like garbage. I was in the latter camp, but determined to minimize the garbage part. I don’t have a press brake, a shear, seaming pliers, a nibbler, or any of the other things one should use for sheetmetal tasks. I wasn’t about to tool up a whole sheetmetal shop in my laundry room for this one little job though, so let’s see about making due.
I know from previous agony that drilling sheetmetal cleanly is a pain in the patouie. The drill deforms the piece, the piece grabs the bit and twists itself up into knots, cats and dogs live together- it’s terrible. Instead, I thought I’d try my hand at making a punch. I’ve never made one before, but the tolerances can be looked up and a punch is ultimately just a round thing with a slightly larger round thing for a die. Lathes are great at making round things, and I have a lathe. Onward!
Ideally, a punch like this would be made with tool steel, not mild steel like I’ve done here. However, this will only be used on brass, and only needs to survive for eight holes, so I’m not concerned. Okay, enough theorizing. It’s the moment of truth- does it work in actual brass?
The next challenge was getting nice one-inch-wide strips cut from my brass stock. Again, the correct tool for this (a shear) makes this a painless task (as long as you keep your fingers out of said shear) and delivers perfect results. I needed another way.
The rule of thumb on bandsaw blades is that three teeth should be inside the material during the cut. For 24ga material, that means you need a very fine-toothed blade. This idea was a no-go.
I also tried tin snips, out of desperation, even though I knew it would make a deformed, roughly-cut mess (because tin snips always do). I didn’t bother photographing that train wreck.
The one thing I found that worked okay was a jigsaw with a very fine blade. The material has to be well supported and clamped close to the cut on both sides to keep from vibrating and deforming, but the resulting cut is actually fairly decent.
I made four strips with the jigsaw method and set about making some holes.
With the clamps cut to length and punched, it was time to braze in the bosses. To the fire bricks!
This is the point where things started to go really wrong yet again.
To make matters worse, while fitting the bottom clamp, all my braze joints to the ferrules on the top clamp started cracking and failing. 24ga sheet just wasn’t enough cross-sectional area for a secure braze, and the joints gave up the ghost under very little provocation. Recall also that I wasn’t happy with the pieces made thus far- misaligned holes, saw cut marks, etc. Everything was just sub-par overall.
I had used up all my brass sheet stock in all these various failed attempts at cutting, punching, brazing, etc. Each step in the process failed multiple times, and that’s usually a sign that the whole enterprise is on the wrong strategy. It was time for a stop-and-think, as my mom used to say.
I decided to start over with a different approach. I would not rely on brazing- the ferrules would clamp through the bands with screws. That would mean the clamps would have a sort of oblate-sphereoid shape to them, but the system would be secure, reliable, and easy to make. Second, I gave up on cutting nice brass strips, and simply ordered 1″ wide material. It felt like cheating, but the result is much nicer. Finally, I gave up on my homemade punch, since getting repeatable well-aligned holes with it was proving impossible.
With all the band clamp pieces remade (dimensions adjusted for my new bolt-through ferrule strategy), it was time for a test fit. “This one time, at band clamp…”
The final step was to attach the legs to a brass plate. This plate serves to stiffen the whole leg system, and also protects the eventual wooden base under the boiler plant from the heat.
I could not be more pleased with how this leg system turned out. While it was a painful process, the end result was all the more satisfying for it.
Now we can get started making plumbing fittings and all the other bits and bobs that will make this empty shell into an actual boiler. Stay tuned!