Simanaitis Says

On cars, old, new and future; science & technology; vintage airplanes, computer flight simulation of them; Sherlockiana; our English language; travel; and other stuff


THE INTAKE and exhaust valves of an automotive engine have busy lives, operating each time the engine fires. Actuation of these depends on (at least) one crankshaft-driven camshaft, with a multiplicity of ways to achieve this. I’ve already shared three of my favorites, from the dawn of internal combustion, 1876, to the 1930s. Here are three more interesting aspects of valve actuation from the 1950s to the present. One requires no valve springs, another could make or break your car’s engine and a third is exotically pneumatic.

Mercedes-Benz desmodromic valves, 1954. The engines of mid-1950s Formula One cars were probing a limit of revs dictated by their valve springs. The reason for this was part metallurgical and part physics.

Compressing and expanding 30 times each second, a valve spring might float (resonating in unspring-like behavior) or bounce (thus compromising the seal of valve into its seat) or simply fail through fatigue. Improvements in metallurgy mitigated outright failure, but even the highly competitive Maserati 250F ran with a 7200-rpm limit.

In its 1954 W196 Grand Prix car, Mercedes-Benz eliminated float and bounce by doing away with valve springs through desmodromic valves opening and closing by cam. Desmodromic comes from the Greek, δεσμός, desmos (bond) and δρόμος, dromos (track), the idea being a valve motion bonded to cam profile.


Mercedes-Benz W196 engine, double-overhead camshafts, desmodromic operation of its valves, 1954. Image from Daimler-Benz AG.

A conventional cam lobe opened each valve; a dual-finger rocker arm returned the valve to its closed position. Revs as high as 10,000 rpm were targeted. In fact, auto authority Leo Levine shared a tale of his 1960 drive in the ex-Fangio W196. Advised to keep the revs less than 7000, he inadvertently selected a wrong gear and buzzed the tell-tale tachometer to 10,200.

The chief mechanic asked, “How long?”

“About a second,” Levine said.

The mechanic shrugged. “Five minutes, it doesn’t like.”

Interference or no? Belts or chains? Durability of modern road cars is measured in tens of thousands of miles, not minutes. Here too there’s a valve-actuation tale to be told: whether your car’s engine is an interference or a non-interference design; whether its valve gear is driven by belt or chain.

As background, pistons and valves occupy similar real estate in an engine’s combustion chamber. With many designs, were it not for the precise orchestration of pistons and valves, these components would collide with disastrous results. Such engines are aptly named interference designs. By contrast, if a piston/valve collision is geometrically impossible, the engine is a non-interference (or free-running) design.


Image from FreeASEStudyGuides.

Why not make them all free-running? Because combustion efficiency comes with higher compression ratio, and such an engine profits from interference geometry. There’s no problem, as long as the piston/valve orchestration works.

This orchestration depends on the precision and integrity of the camshaft drive. Classic engines often had gears for this: precise, robust, costly—and whiney. Modern engines have either rubber-composite belts or metallic chains.

Chains display greater durability, but are more costly and potentially more noisy than belts. Lots of cars use the latter, and their owners’ manuals recommend periodic replacement, say every 60,000 miles.

Belt replacement comes at not-inconsiderable expense, as it requires non-trivial teardown of an engine. (Don’t confuse this camshaft drive with the belt operating a car’s alternator, cooling and other external components.)

Maintenance becomes more than a matter of luck. If the belt fails on a non-interference engine, the engine stops. If the belt fails on an interference engine, the engine self-destructs.

Moral: Follow the specific owners’ manual recommendations. A Google search on “interference vs non-interference engines” yields information, but not always the most up-to-date.

Formula One pneumatic valve return, 1986 to present. Even the best of computer-optimized valve springs have limits around 10,000 rpm. Desmodromic systems might be capable of 15,000. If cost is no object (read: in Formula One), revs can soar beyond 20,000 if valves are closed not mechanically, but by pressurized gas.

The word pneumatics comes from the Greek, πνευματικός (breath), implying technology of pressurized air or gas. The valves of F1 engines are opened by conventional camshafts, but their closure depends on pneumatic pressure.


Pneumatic valve return of the RA122E/B V-12 Formula 1 engine. Image and the following image from A Decade of Continuous Challenges: A Record of Honda’s Formula One Racing Activities,Honda Motor Company, 1993.

F1 generally doesn’t share much technical information. Details of the Renault EF15B’s pneumatic valve return remained sketchy after its 1986 debut on the John Player Special Team Lotus. However, in 1993, Honda published A Decade of Continuous Challenges in which details were offered of its RA122E/B V-12 Grand Prix powerplant.


Cross section of Honda RA122E/B.

Honda’s Pneumatic Valve Return System had inverted pressurized cups attached to the valve stems. The pressure came from nitrogen gas delivered at 6 – 8 bar (87 – 116 psi), equivalent to spring pressure of the replaced components. At the start of a race, the PVRS storage cylinder held approximately 150 bar (2175 psi). Replacing conventional springs reduced engine weight by 20 percent.


Honda’s RA122E/B powered the McLaren MP4/7A in the 1992 F1 season. Image from Honda of Ayrton Senna at 1992 Hungaroring.

The Honda-McLaren MP4/7A’s powerplant had a bore/stroke ratio of 88.0 mm/47.9 mm, a compression ratio of 12.9:1, displaced 3496 cc, exceeded 800 hp at 14,000 rpm—and produced glorious sounds of internal combustion.

With all its technical innovation, modern F1 pales by comparison. ds

© Dennis Simanaitis,, 2014


  1. Mike B
    May 13, 2015

    Another odd one was the Opel 1900 engine (and possibly other Opel models, too) of the late 1960s-early 1970s. It had a cam in the head – so SOHC – but used short lifters and rocker arms as if it were a cam-in-block OHV. In my 1971, the lifters were hydraulic, which were had issues when the engine was revved hard or when it was very cold.

  2. Victor Ceicys
    May 14, 2015

    In 1965 the Honda RC148 racing 125cc 5 cylinder had a redline of 22,000 using valve springs. Honda’s 1960’s racing motorcycles had outstanding engineering in their slide rule era four stroke engines.

  3. Jeff Wick
    May 17, 2015

    “Why not make them all free-running? Because combustion efficiency comes with higher compression ratio…”

    I’ve wondered for some years: Isn’t efficiency more a matter of expansion ratio than compression ratio? Isn’t this the theory behind the modern version of the Atkinson Cycle engine; design in a long mechanical stroke, shorten the effective compression stroke and lower the compression ratio by leaving the intake valve open well past bottom-dead-center, then maintain a long and efficient power stroke by keeping the valves closed until very close to BDC?

    If one were willing to accept low displacement-specific output, would it be possible to operate a four-cycle engine efficiently at an even lower compression ratio than today’s Atkinson Cycle engines, as long as the amount of fuel injected resulted in a stoichiometric mixture, and the power stroke were long enough to extract most of the energy?

    Finally, if we can widely vary valve timing (and of course with direct injection, we can vary the fuel injected), when greater power is required, close the intake valve sooner, inject more fuel to maintain the proper mixture, and we get more power, still with the efficiency of stoichiometric mixture and a long power stroke. Surely not 100bhp/liter, but perhaps enough.

    Best regards,


    Jeff Wick
    Black Earth, WI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

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

%d bloggers like this: