Rotary Valve Engine Gets A Second Chance, Smokes The Competition

It’s a dedicated hacker who has the patience to build an engine from scratch. And it’s a borderline obsessed hacker who does it twice. [Meanwhile In the Garage] is of the second ilk, and in the video below the break, he takes a failed engine design and musters up the oomph to get it running.

The whole build began with an idea for a different kind of intake and exhaust valve. [Meanwhile In the Garage] dreamed up a design that does away with the traditional poppet valve. Instead of valves that open by being pushed away from their seat by a camshaft, this design uses a cylinder that is scooped so that as it rotates, its ports are exposed to either the intake or the exhaust.

Four Stroke Cycle with Poppet valves. Courtesy Wikipedia, CC BY-SA 3.0

During the compression stroke, the valve cylinder becomes part of the combustion chamber, with both ports facing away from the piston. If you read the comments, you’ll find that multiple people have come up with the idea through the years. With his mill, lathe, and know-how, [Meanwhile In the Garage] made it happen. But not without some trouble.

The first iteration resisted all valiant attempts at getting it started. The hour-long video preceding this one ended up in a no-start. Despite his beautiful machine work and a well thought out design, it wasn’t to be. Fire came from the engine either through the exhaust or the carburetor, but it never ran. In this version, several parts have been re-worked and the effect is immediate! The engine fired up nicely and even seems to rev up pretty well. Being a first-generation prototype, it lacks seals and other fancy parts to keep oil out of the combustion chamber. Normal engine oil has been added to the fuel as a precaution as well. The fact that it smokes quite badly isn’t a surprise and only proves that the design will benefit from another iteration. Isn’t that true for most prototypes, though?

Home-grown engines aren’t a new thing at Hackaday, and one of This Author’s favorite jet turbines used a toilet paper holder. Yes, really.  Thanks to [Keith] for the Tip!

42 thoughts on “Rotary Valve Engine Gets A Second Chance, Smokes The Competition

      1. There isn’t really a clear diagram in either the article (which shows a conventional 4t engine) or video but this still works on a 4 stroke concept – an independent valve system with states for inlet open, exhaust open and both closed -but now using ports in a rotating shaft instead of poppet valves. In a two stroke the “valves” are just ports opened by the motion of piston. I think the innovation for this project is in trying to drive the valves directly from the timing belt /cam shaft rotary motion and cut out the cams, springs and linear motion of conventional valves.

      1. I researched this in the 60s, I think Norton did this for one year of racing with the Norton Manx. At the time, I had to go to a large library to find a record of the work. According to what I read back then, there was no advantage over the poppet valve design in the 500cc Norton, so it was abandoned.

    1. One of the really cool things about some rotary sleeve valve engines is they ran the sleeve valves with an epicyclic gear train to get non-uniform motion, so the valve could dwell at not open and then quite suddenly run to fully open and dwell there, and then close quickly. That made timing more precise and gave them a lot of advantages in flow efficiency.
      There were several reasons that the largest and most powerful aircraft engines in WWII were mostly rotary sleeve valve.

          1. Actually a Detroit 2 stroke will not run without a blower. A gas 2 stroke uses crankcase pressure to force the air/ fuel mixture into the cylinder thru the cylinder wall ports but that’s not an option with the Detroit. It is still basically breathing at atmospheric pressure. There were Detroit’s with turbos getting pressurized air to the blower, and those were technically “supercharged” (air entering the cylinder at higher than atmospheric pressure). So your basic 2 stroke Detroit with just a blower is in fact not supercharged.

  1. There were engines years ago that used a rotary disk for a valve and a similar concept. The lack of valve motion aids in power and raises RPM limits since there are no springs or valve mass in alternating directions. It also greatly simplifies the engine components with the removal of 2-4 (or more) valves, springs, retainers, dampeners, rockers, lash buckets, pushrods and other parts per cylinder). This reduces the number of moving parts in a 4 cylinder short block engine to 20 or less including the oil and water pump. I had an idea for a rotary valve system that only has 1 port per cylinder and the rotary valve gate would alternate between the intake and exhaust side by grinding the port gate at an angle into the valve rotor and have it sweep over both intake and exhaust manifold joints. It would be good for low RPM, but at high RPM the lack of valve overlap time would affect VE. It would be interesting to build at about 6:1 CR, direct injection, plasma ignition, and run a turbo at about 3-4 bar with boost starting at idle making a high torque low RPM engine. Guessing 1 litre 4 cylinder could make about 200+ HP under these conditions easily, from an engine that could possibly weigh about 40kg and be the size of a large toaster.

  2. It seems to me an obvious flaw will be the build-up of combustion products on the rotating valve. You need to seal the compression around the valve and those compression products will ruin any possibility of sealing for long

  3. There have been several rotary valve designs through the years, but poppet valves have continued to reign that whole time. If any of the rotary valve designs were clearly superior, they would have supplanted them.

  4. For too long we are stuck with old engine design and all we do is try to make them more efficient so I am delighted to see this revision of the internal combustion engine. Quick question, is this format more efficient or powerful than a standard engine. What about its durability and maintenance. It seems more compact.
    It’s a beauty to watch it working. All the best

  5. If the cam barrel had a 2nd spring loaded and centrifugally controlled concentric internal barrel mechanism to adjust port size and placement the mechanism would allow for variable valve timing and duration according to RPM. (asymmetrical port alignment overlap between the inner and outer barrel). These types of mechanisms can produce volumetric efficiencies in the 130% range through the optimization of valve events. Modern materials allow for about 26.9 M/s max piston speed (F1 and other very high end engines) If an engine has a 2″ stroke and 4″ bore displacing about 25 cu in per cylinder, at 18000 RPM (where the 26.9 m/s speed becomes the limit) and 130% VE each cylinder will consume 170 CFM or enough air to produce 245 HP per cylinder. In an engine with no possibility of valve bounce the only other limiting factors are parts and assembly stress failure limits, ability to maintain sufficient oil film on the friction surfaces, and ability to cool the parts.

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.