How Pneumatic Valve Systems Work

Renault first raced a Formula One car with pneumatic valve springs in 1986, and their use solved so many difficult problems that all other constructors quickly adopted it.

At the time, the idea of “distribution pneumatique” sounded so exotic that it was easy to get the wrong idea about it – that somehow, gas pressure was being used in mysterious ways to “blow” valves open and closed.

The truth is more ordinary. The only difference between a conventional valve drive system, with camshafts and valve springs, and a pneumatic system is that helical steel coil springs have been replaced by compressed air. In both cases, what the spring does is force the valve and its cam follower to remain in contact with the cam lobe profile at all times. With air springs as with steel coil springs before, the spring just keeps the valve train following the path dictated by the cam lobe.

In a metal spring system, a pair of tapered collets engage grooves near the upper end of the valve stem, and their taper in turn locks them to a conical hole in a disc-like valve spring retainer. Under that retainer, and pressing up against it to pull the valve shut against its seat, are one or two steel coil springs under compression. Bearing against the tip of the valve stem is the end of a short, light pivoted finger follower. Its upper surface forms a radiused pad against which the cam lobe acts. Because rubbing forces between cam lobe and this pad on the finger follower are large, the pad is now often coated with super-hard diamond-like coating (DLC).

Everything is the same as with metal springs except that the metal spring is gone and the spring retainer has become a gas piston whose outer edge seals against a cylindrical bore. Under this piston a moderate gas pressure of 90-150 psi is maintained. The valve stem is also sealed to its guide.

The pressure under the valve piston is maintained by a “lecture bottle” of nitrogen gas, acting through a pressure-reducing regulator and pressurizing the volumes under all the valve pistons through one-way valves.

On a MotoGP bike–all of which, save for Ducati, employ pneumatic springs–the bottle is charged from a self-contained compressor/storage unit a little bigger than a mechanic’s stool, mounted on casters for convenience. Once the bottle is charged, the engine can be operated for approximately one hour without needing to be recharged. In the case of racing cars competing in longer races a small engine-driven or electric compressor maintains system pressure.

Engine horsepower is (given best-practice engineering) proportional to cylinder displacement, engine rpm, and stroke averaged net combustion pressure. As displacement is nearly always limited by racing class rules, and because stroke averaged net combustion pressure long ago reached its limit in non supercharged engines, that leaves mainly rpm as the tool by which to seek higher power.

As we increase rpm, we must also accelerate and decelerate its valves faster. One way to soften the problems this brings is to open the valves less far and keep them open longer – both of which reduce valve acceleration. But reducing valve lift throttles the flow, and the longer we keep the valves open, the more mid-range and bottom-end we sacrifice. Why? Because at lower rpm, the intake flow moves more slowly, so it becomes easier for the piston, rising on its compression stroke, to stop and reverse the flow. This, by back-pumping some of the intake charge, reduces engine torque. This is the origin of the “light-switch” powerbands of engines that get their power from extended cam timings.

As race teams worked their steel valve springs harder, asking of them higher valve accelerations, and operating the material at higher and higher percentages of its yield strength, the processes of load loss and metal fatigue steadily shortened the lives of metal springs until the teams were having to change them every day to avoid engine wrecks from broken springs, uncontrolled valve motion, and piston-to-valve collisions.

One evening at the Valencia Spanish MotoGP, I was walking along behind the garages when a door opened, and Stu Shenton of Suzuki stuck his head out. Possibly feeling sorry for me, carrying my heavy load of ignorance, he asked if I’d like to have a look inside. What I saw was the entire team, engaged in the work-a-day task of changing valve springs in every engine they’d run that day. To avoid having to remove the cylinder heads (Suzuki’s engine in those days was a V4) they were using air pressure to hold the valves closed as their spring retainers were depressed, the collets extracted, and the springs exchanged for new ones.

In Pro Stock Auto drag racing it is even worse: springs must be changed after every run. In a recent season of Pro Stock Motorcycle Byron Hines revealed that they were getting “great life” from their springs – eight runs.

Why such short life? Two reasons. First, steel springs even if “pre-set” to limit load loss, steadily lose their pressure as metal yields locally. Second, as a fast-acceleration cam lobe rapidly accelerates the valve and spring, the spring’s coils, which have mass and inertia, tend to pile up against the retainer, then bounce away as soon as acceleration (which takes place in the first quarter of valve lift) turns to the deceleration required to stop the valve at the top of its lift without its continuing to rise in the phenomenon of valve float. That coil pile-up and then expansion sends a stress wave bouncing back and forth from one end of the spring to the other, several times per valve event. This greatly accelerates the rate at which the metal in the spring accumulates damaging stress cycles.

How much do air springs increase possible maximum valve acceleration over metal springs? I visited Del West—a high-end precision parts manufacturer in Valencia, California—some 20 years ago, and was told the new technology had made it possible to double peak valve acceleration.

Mentioned often as the greatest benefit of air springs is that being highly progressive (their pressure rises faster than in direct proportion to valve lift) they can provide the over-the-nose pressure to prevent valve float without needing to have very high seat pressure. High seat pressure leads to valve stretch, leakage, and failure.

A great deal of fine technology has been applied to metal valve springs to improve their performance and durability–use of ultra-clean vacuum-remelted steel, shot-peening of spring wire surfaces to place them in compression, thereby postponing the appearance of spring-destroying tension. Springs may be progressively wound or be given a “bee-hive” shape to outfox standing waves in the coils. Flat wire spring friction dampers may be used, or oppositely wound pairs of springs nested together to kill spring wave action.

Ultimately, however, it was clear that steel springs were setting hard limits to engine performance. Jean-Pierre Boudry ‘s group at Renault F1 developed air springs as a substitute for steel coils, and first raced their system in 1986. Other teams quickly followed, making their use 100% in that series.

It is commonly assumed that since pneumatic springs were first used to allow large increases in F1 engine safe rpm, that must be their only application. Not so! In any situation in which steel springs are keeping valve motion from becoming “what the engine wants”, the superior acceleration tolerance of pneumatic springs becomes useful. Has your team resorted to performance-sapping reduced valve lift and extended open duration to make steel springs reliable in long endurance events? Give them a rest! Send in air springs to do what they do so well! Did you ever see a fatigue crack in air?

Del West claims on its website delwestengineering.com that, “Today we supply air springs and regulators to virtually all the F1 and MotoGP teams,” and that it is the “only commercially available“ design and manufacturing team for pneumatic springs.

Yes, there are hydraulic or hydro-pneumatic or solenoid-electric means of opening and closing valves, but so far, none has proved equal to the task of accurate valve control in very high performance racing engines. Most difficult of all is to progressively slow the valve as it approaches its seat. It cannot be allowed to snap shut because just a very few cycles of such violence destroy the valve’s seal and then break the head off the valve, resulting in an engine wreck. Accurately-profiled cam systems continue to do the best job in this area.

Sign up here to receive our newsletters. Get the latest in motorcycle reviews, tests, and industry news, subscribe here for our YouTube channel.