When You Feel Bad About Having A Better Idea
Just a few weeks ago I did a blog post about how incredibly pleased I was with my new internal combustion engine model. And it is a really neat model! Nothing wrong with it!
Except that about a week after I started selling it, I had a better idea.
For months before finishing the first model, I had been trying to figure out how to make an engine model in which the pistons, connecting rods, and valves all lie in a single plane with each other, so they can all be seen operating at the same time, from a single viewpoint. The fundamental problem is that the crankshaft must necessarily be running parallel to that plane, and therefore the connecting rods (which link the crankshaft to the piston) must swing up and down, outside that plane, moving perpendicular to its surface. Which is difficult if you want the model to be flat.
I was mentally stuck in a rut with a model of connecting rods working like you see here, with a ball-and-socket joint, all in one plane. This is how all my steam engine models work, and I just couldn’t come up with any other style of linkage that seemed like it would work.
So I designed a whole engine around the idea of flat pistons and connecting rods, stacked up in parallel along a crankshaft running perpendicular to those piston/rod planes.
To be clear, I spent a long time with this mental block. Bringing the design from concept to final retail package and listing on my website took several months on-and-off. The whole time I was thinking there MUST be a better way, and the whole time I was not coming up with anything.
Why do I want everything flat and visible from a single viewpoint? Because my goal in designing these models is to make the physical equivalent of a schematic diagram that explain the essence of a mechanical device with a minimum of distracting details. Timing is always an important part of the operation of an engine, whether it’s steam, internal combustion, or electrical. And to appreciate the role of timing and synchronization between moving parts in an engine, you have to be able to see those parts moving in relation to each other.
An example of a problem with this engine is that it’s very hard to see and appreciate the relationship between the intake and exhaust valves, because they are on opposite sides of the engine. Yes, that’s the way it is in many real engines, for good and practical reasons (like that there is more room to the side of each piston than in the space between pistons). But for the purpose of understanding their motion, it’s not ideal.
Then it hit me: a trivially simple way for the connecting rod to meet the piston at a joint that turns a 90 degree angle, with the connecting rod moving perpendicular to the plane of the piston. At first I thought there’s no way it could possibly operate smoothly, but a simple experiment proved me wrong on that: it’s every bit as smooth as my earlier ball and socket joints, despite having a square shaft turning in a round hole. Acrylic is really very slippery stuff, not quite like teflon, but slick enough that loose joints move with very little friction.
Using this simple idea I was able to adapt the design of my first engine to create a new all-in-the-same-plane 3-cylinder engine. Instead of two overhead cams operating the valves, it has a single camshaft with six cams operating an alternating set of three exhaust (red) and three intake (blue) valves, all in one plane.
Just like in real engines, this design leaves less room for large valves, and makes it nearly impossible to have four valves per cylinder (which increases efficiency and performance). It’s really quite crowded along the top there!
But just look at it! You can actually see all three cylinders and all six valves doing their thing in perfect synchronization, all from a single viewpoint. Especially photographed the way it is here, it achieves exactly what I had hoped to: you see almost exclusively active, moving parts, with just enough context to relate them to each other and to the overall device they are part of.
Concentrating on the camshaft, you can see immediately that the exhaust and intake valves for a given cylinder are rotated 90 degrees from each other on the shaft, with the exhaust cam leading the intake cam. That’s because the intake stroke immediately follows the exhaust stroke, and each stroke represents the piston moving from top to bottom, or bottom to top, which requires a 180 degree rotation of the crankshaft. Because there’s a 2-1 gear reduction from the crankshaft to the camshaft, 180 degrees of crank equals 90 degrees of cam movement. (In a real engine the cams may not be exactly 90 degrees apart, because the dynamic behavior of gas flowing in and out of a cylinder is never that simple.)
I intentionally arranged the firing order and placement of the valves so that all six valves open in order from front to back (left to right in this view). There’s 90 degrees from exhaust to intake, and then there’s a further 30 degrees from the intake of one cylinder to the exhaust of the next cylinder in line. That’s because there is 1/3 of a rotation, or 120 degrees, of cam rotation (240 degrees of crank rotation) between the firing of each cylinder. 90 + 30 + 90 + 30 + 90 = 330 degrees total between the first and last cam on the shaft. 30 degrees more and we reach 360 degrees, or one full revolution, and the cycle starts again with the first valve.
I chose three cylinders partly because the spacing of the cylinder firing gives a nice highly-varying pattern of movement where each cylinder is doing a different thing at a different time. In 4-cylinder engines, by contrast, the cylinders are typically moving up and down in pairs, which is less visually engaging (though better from the point of view of cancelling out vibration).
Valve timing can lead to quite lovely rhythms, especially in old engines with huge cylinders running at very low speeds. This antique tractor I filmed at a show in Rantoul, Illinois is a joy to listen to, don’t you think? (Play the video and turn up the volume.) You can see the rocker arms operating the valves: in this style of engine the cams are close to the crankshaft and push rods transfer the cam action to the other side of the cylinders where the valves are.
There is one odd fact that results from the way I made the connection between the pistons and connecting rods. It’s not visible in the side view above, but if you look at the engine from an angle, you see that the connecting rods have an odd bulge.
Why? Well, imagine that the rods were just straight bars, rattling around as they swing back and forth. The only things holding them in place in the left-to-right direction are the slots in the front and back plates. If they were too narrow, they would come completely out of the slot for some or all of their length in some positions. This would lead to them potentially bumping into one side of the slot as they come back. Maybe it’s hard to visualize…but the shape is designed to be the minimum curve that keeps each rod entirely within both slots, front and back, throughout its whole range of motion. This avoids any potential jamming, at the expense of looking a little odd from the side.
Now for the hard question: between the older dual overhead and the newer single overhead cam designs, which is the better model? It’s tempting to say each has its own merits. And, yes, if you’re seeking to demonstrate the difference between these two designs (both of which are widely used in real engines), it’s great to have both available. But to be perfectly honest, if I’d thought of the newer connection idea sooner, I would never have made the first engine. The new one satisfies the number one criterion I had from the start, and which I didn’t achieve with the first design: everything visible from one viewpoint. That’s what I need for use in my upcoming book about engines, and that’s what makes this new design even more satisfying to watch.
From an educational point of view, it’s particularly useful that both intake and exhaust valves are visible at the same time, which is not really the case for the old design, except when you’re looking at it from the back, in which case you’re only really seeing one cylinder in action.
So what about all the people who ordered the old design? I know! I feel terrible that I’ve made a better one so soon! Well, if you’re one of those people, you should have gotten an email from me offering you this new one at half price, which hopefully makes up for it. Going forward I do intend to keep offering both designs, but in each listing I describe the pros and cons and encourage people to consider the new one unless they specifically want a dual overhead cam model.
Here are links to the two models: