Spring Fever

Spring fever

TDC

Kevin Cameron

NOW THAT FOUR-STROKES HAVE BEEN put back into Grand Prix roadracing, the subject of pneumatic valves keeps popping up. People automatically expect that MotoGP four-strokes must adopt Formula One car technologies. Yet, only Aprilia has employed pneumatic valves, so far without visible advantage over those who remain with metal springs.

It was therefore with great interest that I recently read a brochure from Integral Power Train, an English consulting firm formed by breakaway Cosworth engineers. They describe the effects of a switch from metal springs to an “air spring solution” with great clarity.

As an engine’s operating rpm is raised in an attempt to produce more power, the rate at which its valves must be accelerated and decelerated naturally increases. But metal springs are made of dense, heavy material-steel or titanium—and so wave motions propagate in them during these motions. Metal springs are wound to have very high natural frequencies, for if the natural frequency is greatly higher than the engine’s rpm, the spring is less likely to be excited into large-amplitude motion. But there are limits: When you raise the spring’s frequency, you also raise the stress in the wire. At some point, even with use of highest-purity wire, its fatigue properties enhanced by shot-peening and/or surface acid etching, spring failure becomes too frequent.

Further rpm increases are possible, but only if valve acceleration does not also increase. This condition can be met in two ways: 1) By reducing valve lift or 2) by extending valve-open duration. Either way, you are in effect making the “hill” that the valve must climb less steep.

The airflow people would like to get the valve up and out of the way of the flow quickly, and doing this well requires that the valve be lifted to about 40 percent of its head diameter. But such high lift means high acceleration, so that’s out, and limited lifts must be tolerated.

Extending valve-open duration also allows higher rpm at a limited valve acceleration, but the longer the valves are open, the more overlap lift there must be. Overlap is that period near TDC at the end of the exhaust stroke when the exhaust valves have not yet closed, but the intakes have begun to open. The higher overlap lift is, the deeper the milled valve clearance cutouts in the piston must be, and the lower the compression ratio that can be achieved. Lower compression equals reduced engine torque.

So which do we prefer, poor airflow or limited compression ratio? Because we have to run the engine to high rpm, which requires generous airflow, we choose to keep the valve lift and accept the drop in compression ratio caused by high overlap lift and deep valve cutouts. But we don’t like either choice. That dislike caused engineers to dream up pneumatic closers.

Using gas pressure in place of a metal spring allows a great increase in maximum valve acceleration/deceleration rates because the air spring’s natural frequency is so high it’s out of sight. It is, in fact, the speed of sound divided by twice the length of the air spring cavity. If that length is 2 inches, then the frequency will be on the order of 3500 cycles per second, or more than 10 times the engine rpm. Further, because the gas has almost no mass as compared to a metal spring, its oscillation contains little energy. The result is practically zero vibratory “spring” pressure variation. The gas spring just pushes. It doesn’t gallop and oscillate like a metal spring.

With spring wave action eliminated, maximum valve acceleration can be increased. In the “air spring solution” it rises by a conservative 20-25 percent, but others speak of larger increases, as high as 100 percent. In any case, Integral Power Train’s graph shows that with air springs, lift could be increased from 11 to 14mm (27 percent), which makes the airflow people happy (lift as percent of valve head diameter rises from 28 to 35). By applying part of the benefit of the air springs’ higher acceleration tolerance to duration, the former long 320degree duration is shortened to only 300 degrees-bringing valuable reduction in overlap lift. That allows the depth of the valve-clearance cutouts in the pistons to be reduced, thereby increasing compression ratio. This, in turn, increases engine torque. Integral Power Train calls this a “virtuous circle” of benefit.

Of course, nothing is ever solved in engineering-the problems just morph and move, reappearing somewhere else. In this case, as cam duration is made shorter, the nose radius of the cam lobes becomes smaller. This increases the so-called “Hertz stress,” which reaches maximum values just below the surface of the lobe. Because we can’t have cam lobes breaking up in operation, the properties of the cam material have to be improved. Does weld application of hard-facing rod to the lobes solve this? Does the cam itself have to be made of super-clean, highly fatigue-resistant steel, the kind that revolutionized the fatigue lives of rolling element bearings after 1964?

If valve accelerations increase, so do the torque spikes that must be transmitted through cam drives, whether they be gear, belt or chain. Torque spikes mean that gear teeth must bend more, or belts stretch, or chains whip back and forth harder against their dampers. Drives that had previously been developed for stable operation with metal valve springs may display some new kind of damaging dynamics. If accessories are driven from cam or cam drives, these dynamics may damage the accessories even if they do not break gear teeth or belts, or cause rapid chain stretch.

If air springs do find a useful place in motorcycle racing, it is likely that their benefit will be applied to making valve duration shorter, which smoothes partthrottle operation, enabling machines to get off turns faster. That point has yet to be reached, because Honda dominated the 2002 championship with metal springs.