Steampunk Motorcycle Runs On Compressed Air, Is Pure Hacking Art

Sometimes it’s ok to sacrifice some practicality for aesthetics, especially for passion projects. Falling solidly in this category is [Peter Forsberg]’s beautiful, barely functional steam punk motorcycle. If this isn’t hacker art, then we don’t know what is.

The most eye-catching part of the motorcycle is the engine and drive train, with most of the mechanical components visible. The cylinders are clear glass tubes with custom pistons, seals, valves and push rods. The crank mechanism is from an old Harley and is mounted inside a piece of stainless steel pipe. Because it runs on compressed air it cools down instead of heating up, so an oil system is not needed.

For steering, the entire front of the bike swings side to side on hinges in the middle of the frame, which is quite tricky to ride with a top speed that’s just above walking speed. It can run for about 3-5 minutes on a tank, so the [Peter] mounted a big three-minute hour glass in the frame. The engine is fed from an external air tank, which he wears on his back; he admits it’s borderline torture to carry the thing for any length of time. He plans to build a side-car to house a much larger tank to extend range and improve riding comfort.

[Peter] admits that it isn’t very good as a motorcycle, but the amount of creativity and resourcefulness required to make it functional at all is the mark of a true mechanical hacker. We look forward to seeing it in its final form.

For more inspiration check out the DIY electric motorcycle, and the flying motorcycle that the Dubai police is testing.

27 thoughts on “Steampunk Motorcycle Runs On Compressed Air, Is Pure Hacking Art

  1. A truly interesting project, and rather impressive craftsmanship too.

    Though, the engine could need some minor improvements, it seems a bit jerky to say the least…
    Adding a flywheel could smooth things out a bit, but I guess a lot will be needed, so might not be worth while… (Unless one use a planetary gear set to have the flywheel spinning at a much greater RPM, as to give it far more inertia to work with.)

    Then there is also the question if acrylic or polycarbonate would have made for better cylinders, since glass isn’t particularly known for its tensile strength. Could allow the thing to run at a bit higher pressure, and together with a good flywheel, could make the experience a bit more pleasant.

    Though, the steering system seems rather laughable, almost to the point where a solid frame would have just made more sense. Maybe give some assistance by adding some springs to help on keep it straight could be a decent upgrade, not to mention adding to the steampunk look of it all.

    1. The glass is fine for the 5 bar pressure he’s using. It’s rather that the cylinders are too large and the piston stroke is too long, so it uses up too much air just to keep moving. It’s basically running on “idle” at max RPM for the engine at any speed and he’s basically just blowing air through it. The efficiency of the engine is approximately zero.

      That’s also the reason why the engine freezes over. When you empty a scuba tank in 3 minutes, you’ve got about 7 kW of power at tap but most of it is lost in the rapid expansion of the gas – it’s working to make the engine cold instead of driving the bike. If you release it much slower, say in 30 minutes, you’ve got about 700 Watts (1 HP) of power and the cooling isn’t so dramatic so the pressure stays up and the air can actually do some work, but the engine must be sized for that throughput. Now he’s got a bucket a bang engine that’s way oversized for the amount of power he should be using.

      1. If he had something smaller, like a 100 cc engine running at 5 bars and something like 200 RPM, he would be consuming 1.67 litres of air from the tank per second, so a regular 10 litre 230 bar scuba tank would last him close to 40 minutes. Given a reasonably poor efficiency due to the expansion cooling, about 5-10 minutes in practice, but he would actually be going somewhere with the machine at some speed.

      2. 5 bar, he states in the video that he uses 8 bar, and the glass is rated to 10.

        Now, I do understand that watching a 20 minute long video in a weird language with frankly lackluster sub titles can make things a bit hard for most people to follow along. Though personally, it is my native language….

        But I do have to ask where you got “That’s also the reason why the engine freezes over.” from, he never states such in the video. He does however state that the air tank/regulator freezed over, but not the engine. And that isn’t surprising, since the pressure regulator is where we have the largest pressure change that the gas gets to see in the whole contraption.

        He does state that an air engine does get cooled by the expanding gas. But since the engine isn’t having a lot of power to begin with, it isn’t cooling down all that much. And he also talks about how some people think he is only joking about it running on compressed air, even how some people ask where the petrol tank is.

        Though, the engine does have the design flaw that it fills the cylinder with high pressure air until it practically bottoms out. This is good from a power standpoint, but it also means it consumes a lot more air and generally doesn’t get much energy out from the air. This means that the cylinder runs at full power, so rather far from “Idle” if I have to say so myself.

        Cutting off the inlet on the down stroke and letting the air expand to full cylinder volume by itself is generally going to increase fuel efficiency rather dramatically. Though at the expense of torque and power.

        But the engine also exhausts the air when the piston is about half way up. So it kinda works more like a compressor up to that point. (Now it needs to work as an engine for the majority of the time, otherwise it wouldn’t run, but the performance seems to be held back due to poor valve timing, something that shouldn’t be too hard to fix. A sign of this being the case is that he needs to push the thing into motion. A proper valve timing should be fully able to start by itself. (unless it is a single cylinder engine, there one do have an edge case where the timing system can park in the dead zone…))

        So with some timing adjustments, the engine will likely consume a fair bit less air, and run better overall.

        Adding a flywheel will also help to even out the speed of the engine, after all, it kinda only outputs power in one half or so of a revolution. Though, he doesn’t really show it running at speed, so then it might be a different story.

        1. >”“That’s also the reason why the engine freezes over.” from, he never states such in the video.”

          Yes he does. He said that when he ran the scuba tank empty in 3 minutes, the cylinders were freezing over.

    2. Yeah the steering system especially belies that he hasn’t really had much experience or understanding of the basics of how a two wheeled vehicle works. Or perhaps he’s never even ridden a motorbike. Which should be some kind of prerequisite you’d think.

      Which is a shame cause he definitely could make an actual running gasoline engine with this level of skill, should just do that and then get someone who knows to design the frame so he doesn’t die. Cause it looks great.

  2. Very nice build ! The prizes were well deserved.

    This reminds me of an Uruguayan engineer, who around the 80’s developed some prototypes of compressed air driven vehicles (motorcycles and cars) and even an “Hydrosvol” (driven by a flywheel).

    His prototypes weren’t so beautiful but they worked OK and he was showing them in action, in various uruguayan cities and highways. You can find more info and photos at:

    You can Google for: ingeniero regusci and ingeniero regusci moto to find more info (mostly in spanish).

    He was also in a project in North Texas University to develop his vehicles but I can’t find more info about it (one of the photos at is described as: “Ing Mitty Plummer and Ing Armando Regusci in North Texas University”).

    (Prof. Mitty Plummer appears for example in: ).

    Best regards,
    A/P Daniel F. Larrosa

  3. “Because it runs on compressed air it cools down instead of heating up, so an oil system is not needed.”

    Pretty sure the oil lubrication on most ICEs is for lubrication (thus wear reduction) reasons, not heat management.

    1. System is the word. The heat needs taking out of the oil to preserve the oil too. In other words without the oil being at risk of boiling up, burning, or getting full of soot or loess sucked out of the air, you don’t have to have a system to cool and filter it, and it will stay put, requiring periodic lubrication but not constant.

    2. Forgot to add: Yes, I know there is such a thing as oil cooled engines, on motorcycles this mainly serves to cool the heads usually

      I guess if you’re only running 3-5 minutes at a time it doesn’t really matter. You wouldn’t be running a SaddleSore 1000 on this thing anyway.

  4. The jerkiness exists for the same reasons a Harley engine is jerky. It’s a 2 cylinder with a 45 degree cylinder offset that uses a common crank pin for both cylinders and fires them both on the same rotation. (Harley originally did it for simplicity’s sake, and keeps doing it because that’s what people pay for)

    It doesn’t po-ta-to like a Harley because it’s essentially a 2 stroke so there’s no long pause between the second cylinder firing and the first one firing again. It doesn’t really po-tate either but that’s probably from using low pressure air instead of explosions to power it.

    Either way, the jerkiness is most likely intentional.

    1. A Harley V-Twin could have a different camshaft made to change the valve timing so the strokes are 180 degrees apart rather than one right after the other so that instead of going I-I, C-C, P-P, E-E, they would be P-I, E-C, I-P, C-E.

      Do such camshafts (and distributors or electronic ignition to match) exist for Harleys?

      Harley’s method is essentially to run the engine as a split single cylinder. The other timing would space the two power strokes out for smoother running. To get even smoother would require a crankshaft with two throws, which wouldn’t work on a Harley V-Twin due to how the cylinders are in the same plane and the big ends of the connecting rods are made with one split and partially wrapped around the other. For an example, look for photos of a 1939 Harley crankshaft.

      A dual throw crank could be made, but it would take some really funky shaped connecting rods and very likely specially designed pistons so the rods’ big ends could be offset and clear the center web of the crank.

      All that would be a lot of effort to go to, to make a Harley run smooth and quieter. Judging by the many and varied aftermarket Harley crankshafts with super heavy, full circle counterweights, making a Harley run smooth as possible is a popular thing.

      Harley once tried to trademark or copyright the sound of their engines in order to block other manufacturers from using a single throw crank with the same one after the other stroke order. They got denied.

      1. I get the feeling that for the size of firebox and boiler that would be practical on a sidecar it wouldn’t run a heck of a lot longer than tanks of similar dimensions anyway.

        Best improvement would be heat exchangers on the air lines so at least it gets close to atmospheric temperatures. It’ll always run like crap in winter though.

  5. A random obscure fact – back in the early days of the last century electrical motors still hadn’t really caught on in industrial applications and many factories still used the model of one giant steam engine turning a big line shaft that ran down the building, and individual machines would tap their drive off the shaft with belts.

    Of course, this implied stationary machines, so what do you do when you want some degree of portability?

    Well, a plant of any size would already have copious amounts of compressed air, and the available air was commonly used instead of that newfangled electricity stuff that would be an expensive retrofit.

    For the first part of the century little air motors were a stock item in any good machinery catalog, and you could order machine tools right from the factory with the air motors installed.

    Even once electricity took over, air as a power source stuck around surprisingly long in some specialty fields where equipment had to be portable and weight was an issue. For example, some of the early movie cameras had air-powered options for “action work”. A short roll of film only lasted 4 minutes and a quick-change air cylinder was much smaller than the bulky batteries of the day.

  6. this could be better riding with just some adjustments to where the hinges on the steering are. a more upright rake and less bent over riding posture, would work wonders without really messing with the overall aesthetic

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.