Is this the next great leap for internal combustion?

HVD456 Flow Full
Vaztec

We may have been doing this wrong all along. With cylinder heads, that is.

Back when it all began, steam was the start—how we first captured heat energy from fuel and put it to work. More than a century ago, steam powerplants formed the basis for the design of the first internal-combustion (IC) engines. Moving the combustion from an external source (a steam boiler) to the inside of the drive cylinder was a major step forward in power and efficiency.

Since steam engines were laying around all over the place, many parts were carried over. Components like pistons, connecting rods, and cranks were near-enough correct to make a gasoline engine sort of run, and things evolved from there. The sliding sleeves and popular barrel valves used to control the flow of high-pressure steam, however, did not make the cut. They leaked.

Leaking a little high-pressure steam here and there is not of major consequence in the grand scheme of a locomotive. Because cylinder leakage carries real drawbacks with an IC engine, sleeve valves and barrel valves were abandoned, replaced by a kind of valve called the poppet. The type of valve now found in almost every four-stroke IC engine on earth.

Poppet valves were and are a decent fit for a four-stroke’s high internal pressures and combustion temperatures. Mated with a countersunk seat, the poppet’s chamfered face forms a positive seal, and that seal improves as pressure on the valve face increases. Although they require complex drive and support systems (cams, keepers, springs, lubricated guides, etc.) poppets have been developed into a reliable solution.

Poppet Valve Engineering Drawing Patent Diagram
A patent drawing for a novel poppet-valve system filed in the early 1920s. The specifics of this particular layout proposal aren’t important, but note the tulip-shaped valve at right. A century later, poppets don’t look much different. USPTO

What we gained in reliable sealing, though, we lost in flow. A poppet valve’s head is always in the way. On the intake stroke, a poppet valve’s head blocks about 20 percent of the air trying to fill the cylinder. With the exhaust stroke, the valve is essentially attempting to evacuate a crowded theater through a door that won’t fully open. The piston is rising, pushing combustion byproducts out the exhaust port, but the valve’s head is in the way.

In the case of the intake valve, we’ve evolved to a sweet spot where a poppet can flow more than 100-percent capacity—more air can be moved past the valve and into the cylinder than the dimensions of the intake path and cylinder would suggest. But that’s due to the velocity of the incoming airstream, the air “ramming” itself past the valve’s head. (For an exaggerated illustration of this effect, see the now-outdated Japanese “subway stuffers”—they were literally ram effect personified.

This is no small issue. Every engine is, in essence, a dumb air pump, limited in its ability to bring in air. Burning gasoline and exhausting high-pressure gases are pretty easy tasks by comparison.

Cylinder filling has been the problem for two centuries, since we began moving away from steam. The column of air sitting over the earth weighs only enough to create around 14.5 psi of ambient pressure at sea level. When it comes to pushing air into a vacuum—like the one created by a piston moving down a cylinder—that pressure isn’t much. It’s even less at altitude, like on Pikes Peak, where the mountain’s height leaves the air column above short enough to give less than 9 psi. (Ever wondered why turbo- and superchargers predominate on race cars run at altitude? Now you know.)

After 100 years of being fussed with, the poppet valve is, in a sense, grandfathered in. It is now highly functional but not optimal. We have band-aided the poppet’s poor performance partly by increasing the number of valves—some cylinder heads have four or more per piston. Over the last century, however, the inventive human mind has tried to revive certain steam-engine concepts, to take advantage of their superior flow ability.

During World War II, when air power was the ultimate arms race, engine designers on both sides dusted off the concept of sleeve valves. By configuring those valves around pistons, they found tremendous power improvements. (While also making massive progress in the development of things like high-octane fuels, superchargers, turbochargers, and nitrous-oxide injection—all over the span of about five summers.) The power increases came mostly from the airflow made possible by those sleeve valves, a clean and unobstructed path from intake port to cylinder, no poppet head in the way.

Bristol Aircraft sleeve valve engine
The operation of the sleeve-valve assembly used on Bristol airplane engines during World War II. The sleeve rose and fell within the cylinder, around the moving piston; the holes in the sleeve’s side served as ports for intake and exhaust. Bristol Aeroplane Company

This improved air flow was put to good use: A fighter plane that can out-climb or outrun an opponent can easily turn and shoot them down. For all the benefits brought by their improved flow, though, sleeve valves leaked oil, they wore out too quickly, and they required monkey-motion levers in the crankcase in order to work correctly. The technology never made it to automotive production in any significant form.

Still, the siren song of full flow capacity is hard to resist. And like the sleeve valve, the barrel valve—a type of rotary valve—has long captured the fascination of engine designers.

Vaztec Single Barrel Rotary Valve Exposed in 45cc Prototype Cylinder Head
The barrel rotary valve and cylinder head of a 45-cc single-cylinder prototype engine developed by American firm Vaztec. As the barrel spins, the rectangular holes allow air-fuel mixture and exhaust gas to reach the engine’s intake and exhaust ports, respectively. Vaztec

A shaft with a slot cut into it, rotating in seals, is so simple and capable that it has been tried dozens and dozens of times. Where a standard poppet valve may peak at flowing 85 percent of its intake area (and then only temporarily, when fully open), a rotary valve opens faster and can flow 100 percent of its intake area more of the time (meaning, over more of the valve’s opening duration). If you have a 5.0-liter engine that can only breathe at 85 percent of capacity, you don’t really have a 5.0-liter engine. One-hundred-percent intake-valve flow makes more power every time.

With the barrel valve, the gremlin has always been sealing. For a long time, any seal material that pressed against a barrel valve’s shaft enough to contain combustion pressure—upwards of 1500 psi at 2500 degrees Fahrenheit—would also give high friction and wear. If you loosened up the seal’s tolerance on the shaft to reduce that friction, it would leak. Either way, the high temperatures ate away at any sealing material tried.

The industry birthed prototypes, test engines, and a few production attempts, but no rotary-valve design or layout has been fit for real commercial success. Felix Wankel, the inventor of the rotary engine, started his career developing disc rotary valves that were eventually used in German torpedoes. In 2004, the Mercedes-Ilmor Formula 1 team got as close to the summit as anyone, with an adaptation of a rotary valve patented by a man named Bishop. But alas, F1’s (power-hating?) sanctioning body responded by outlawing any valve not a poppet. So much for racing promoting innovation.

Cross Rotary Valve and the 500cc Cross-Rudge Engine
The Cross rotary valve and the 500-cc circa 1935. The arrangement is similar to the now-outlawed Bishop design developed by the Mercedes-Ilmor Formula 1 team for 2004. Crossley Bros./oldbikemag.com.au

All hope is not lost. As part of my work as a professor of IC Engine Design at UNC-Charlotte, I was called to consult in “Race City, USA” (a.k.a. Mooresville, North Carolina, home of NASCAR), to assist a company that had found a way to seal a rotary valve and take advantage of the potential.

That company, Vaztec, was founded by a team of engine developers who had supported GM and Ford racing for decades, and who had grown disillusioned with how badly a poppet valve actually flows. Now six years in, they have just returned from a Society of Automotive Engineers conference in Japan, where their white-paper presentation on a successful rotary-valve design won an award.

Maybe time is nigh for a change.

Single Intake Rotary Valve Detail View
Vaztec

Vaztec has developed and patented a dynamic, high-temperature sealing system that adapts to the shape of a rotary valve as it expands. That expansion is the true Achilles heel of a rotary-valve system. Almost all metals expand as they heat up. If your valve seal relies on a thousandth of an inch of clearance to work, as many automotive seals do, having that valve grow three times that amount as the engine warms is a recipe for failure.

As the Vaztec project advances, some interesting side benefits have appeared. One is a cooler-running valve. The poppet is a sitting duck in the combustion chamber, exposed to 2500-degree combustion gases during the piston’s power stroke. Then, during the exhaust stroke, the valve’s head gets stuck in a jet blast of 1400-degree exhaust. Milliseconds later, that head carries leftover heat into the engine’s next power cycle. Because a barrel valve is constantly turning, heat transfers more evenly into the cylinder head.

HVD456 Flow Full
Piston up and down, barrel spins, fuel-air mix moves in, exhaust gas moves out. Full flow, no poppet stem or face in the way. Vaztec

That heat transfer is helpful. At peak power, the head of a poppet exhaust valve can serve as a “glow plug,” growing so hot as to spontaneously ignite fuel and air. Lowering an engine’s static compression ratio helps alleviate this; by the same token, if you eliminate the glow-plug effect, you can make that ratio higher. All else being equal, higher compression ratios improve power and efficiency—a strong step toward keeping the IC engine alive and relevant in years to come.

Those benefits were somewhat expected. An unexpected byproduct of unobstructed intake flow was the increased, and now very high, velocity of intake air as it rushes into the combustion chamber.

Intake air gets up to crazy speed in high-load conditions—Mach 0.55, or around 450 mph. That velocity is what drives the aforementioned “ram effect,” the jamming of a combustible air-fuel mixture into a cylinder. That velocity generally produces turbulence, which stirs up and quickens the combustion process, aiding power.

Computation Fluid Model of Intake Flow through Rotary Valve
An animation of the path taken by the intake charge in Vaztec’s barrel-valve system. The gradient strip at the bottom reflects charge velocity. Vaztec

Interesting side note: A few decades ago, I had a fascinating conversation about intake velocity with former Car and Driver editor Gordon Jennings. He proposed Mach 0.9 as the actual goal—near-sonic airflow. Even then, however, things would still be greatly disturbed by the poppet valve’s head, which hangs out at the end of the intake port like a bouncer at the entrance to a party. With the bouncer gone, things change. Rather than an orderly “deflagration” (picture of a sheet of paper lit by a match at one corner, the flame moving slowly to engulf the page), we get flame-folding and combustion that takes place much more quickly.

This is a different type of cylinder process than we are used to, one highly efficient, very rapid, and very useful. A telltale is how Vaztec’s development engines have required roughly half as much ignition-timing advance as they would with poppet valves. Because the fuel-air mixture burns quicker, these engines can fire their spark plugs later in the compression stroke, which lets them make more power—the force of the cylinder’s burn is no longer trying to shove the rising piston back down. (With slower combustion rates, this is a very real “taxation” cost of advanced ignition timing.) The greater turbulence also sweeps lazy end-gases into the combustion party, giving less time for detonation sites to emerge in the chamber.

Rotary Valve vs Poppet Valve intake diagrams
A high-velocity intake stream slows greatly when forced to work around the stem and head of a poppet valve—note the relative lack of yellow and red in the combustion chamber (the rectangle on the right) of the upper image. Vaztec

Designing a new engine system is one thing. Actually getting it to run is another. Six years ago, Vaztec began developing its rotary valve inside, of all things, a diesel engine. Diesels use extremely high combustion pressures, up to 2500 psi, so this was a tall bar, but also a fine way to prove the idea. Next, the company set its sights on the power kings of the engine world: two-strokes.

A two-stroke cylinder has a firing cycle every crankshaft rotation. A four-stroke fires only every second rotation, half as many power pulses for a given rpm. Although two-strokes excel in power output per cubic inch, they offer relatively poor control of incoming and outgoing gases, and that serves as an upper limit for output. (My ’72 Bultaco 250-cc two-stroke motocross bike made 144 hp per liter when I raced it 45 years ago, and Honda’s two-stroke, 500-cc NSR 500 grand-prix bike made 400 hp per liter in the 1990s. Compare these to the laudable 5.5-liter V-8 in the current Corvette Z06, which makes 122 hp per liter. What two-strokes don’t do well is run cleanly, which is why federal emissions regulators are eyeing them for eventual banishment.)

HONDA NSR 500 Motorcycle front three quarter
Honda’s NSR 500 motorcycle. Honda

As a test, Vaztec set its sights on making a high-power, clean-emissions replacement for a 50-cc two-stroke, in the form of a rotary-valve four-stroke. This was the topic of the award-winning SAE presentation mentioned earlier. Vaztec’s rotary-valve, 45-cc single matches the power of a production 45-cc two-stroke chainsaw/weed-trimmer engine, and it can rev just as high (more than 12,000 rpm). It also made 50 percent more power than a similar, 48-cc four-stroke poppet engine currently on the market.

Power Graph for the Rotary Valve gas string trimmer engine vs Poppet Valve 48cc engine
Vaztec

About that rpm peak: Poppet valves give an engine an upper speed limit by nature. At extremely high rpm, a spring-closed poppet valve will float—hover off its seat, unable to fully close and seal before the cam that controls the valve forces it open again. Since a rotary valve spins like a shaft, supported by bearings (imagine a camshaft without lobes or loads), it lacks that traditional limit. That Mercedes-Ilmor F1 engine easily spun above 20,000 rpm, limited primarily by the strength of its crankshaft and connecting rods.

For a more automotive-scale test, Vaztec has developed a one-cylinder engine with two rotary valves, one each for intake and exhaust. This engine uses a 90-mm bore and a 70-mm stroke, which makes it comparable in size to one cylinder of a 1.8-liter in-line four. It currently makes 50 percent more power at 4000 rpm than the similarly sized poppet-valve engine that Vaztec chose as a baseline. Very promising.

On the endurance front, Vaztec has built ten prototype engines over the past six years, from a 28-cc single to a 5.3-liter version of GM’s LS1 V-8. The company’s engineers have hundreds of hours of durability tests showing scalability and strong results. The valve itself is simple, with very little to deteriorate. Even at these early stages of development, wear rates of these prototypes extrapolate to an engine life exceeding 5000 hours.

At the moment, Vaztec is working with engine manufacturers to scale the technology into powersports, and eventually automotive-production applications, including hybrids. The greater power density, improved combustion, and increased efficiency are coupled with reduced noise, reduced vibration, and reduced wear. On top of that, those benefits are brought by a system that uses 33 percent fewer parts than a traditional poppet arrangement, that requires no special materials or processes to create, and that costs less to make and assemble.

An added bonus? No valve adjustments for the life of the engine. And we get to keep our pistons and crankshafts.

While most of the advances in engines in the past three decades have related to controls like direct fuel-injection and variable valve timing, it seems we’ve found a way to improve the fundamentals once more.

Could the poppet valve go the way of the choke lever and dashboard cassette player, into the dustbin of history? Perhaps, just as carburetors have been largely replaced by fuel injection, rotary valves will take over. We shall see.

***

Full Disclosure: Norman Garrett is an engineering consultant for Vaztec and has helped develop the company’s rotary-valve technology. He is a freelance contributor to Hagerty and was paid for this story.

Check out the Hagerty Media homepage so you don’t miss a single story, or better yet, bookmark it.

Click below for more about
Read next Up next: Mustang Dark Horse boasts wild blue paint, Tremec’s new electric motor, Mazda prices the CX-90

Comments

    VVT (variable valve timing) is now becoming common place. Does this rotary valve allow that function. VVT helps increase efficiency throughout the rpm and load range. This is important for emissions reduction and milage increase. Also, hemispherical heads increase h. p. Do the rotary valves eliminate the need for those?
    Perhaps these valves are only used in specific application motors, like high rpm only racing and utility machines.

    It depends. A dual valve per cylinder setup can allow for variable valve timing like any dual cam head out. The single rotary valve per cylinder configurations have a fixed timing between intake and exhaust but could use the variable valve duration setup that Rover used on their K series engines in the 90s. The most difficult thing to do with a rotary valve is change the “lift” like Honda’s VTEC does. It is technically feasible but ads a level of complexity that would require a valve within a sleeve or variable valve duration on each rotary valve.

    So they really didn’t disclose how they overcame the sealing issue. Would like to know what novel approach was used to make this happen.

    I don’t know if I missed it, but is the fuel efficiency of this system increased over the poppet valve Life’s? Love the concept of this!

    Makes sense if the seals will tolerate the combustion temperature and wear from the rotary shaft My next question would be; how wide is the power band?

    I’ve built a handful of rotary valve engines. Most single cylinder but also a 6 cylinder BMW e36. The article is certainly correct that sealing the rotating valve is the biggest challenge. The most common successful way is to utilize a floating seal under the valve that uses combustion pressure to push the seal against the valve. My preference is to integrate whole top of the combination chamber into the seal. The main mistake seems to be that most people want to use small seals made of exotic materials. I’ve found a large seal made of bronze works quite well. These guys at Vaztec sound like they are on a good path. Their cross drilled valve is not a efficient for flow compared the axial flow rotary valves that Bishop did in the 1990s or Roland Cross did in the 1930s. Still the cross flow valve is more efficient than a poppet.

    This was a fun article, I like seeing that there are more people out there working on Rotary Valves.

    I learned the advantages of using a Turbocharger at altitude. My Corvair Spyder ran just as well at the top of Pikes Peak as at the bottom. On a warm August afternoon in 1967 we decided to take an afternoon ride up the mountain. We went by many cars which would not run by halfway up the road and had to park on the side. The thin air actually caused the turbo to turn easier and therefore faster, generating more boost and pumping more air.

Leave a Reply

Your email address will not be published. Required fields are marked *