Engine Braking
ENGINE BRAKING
What happens when you roll off the throttle? It depends on what you're riding.
STEVE ANDERSON
MOTORCYCLE HANDLING ISN'T JUST ABOUT CHASSIS and tire design. As Don Canet, fresh off his "Master Bike" multi-model supersport shootout can tell you, a critical part of the machine-control equation is the way an engine behaves both on and off the throttle.
I’ve long hypothesized that one reason some riders prefer classic sporty Twins over four-cylinder machines is that the Twins have more flywheel-and thus less engine braking when you chop the throttle. I realize this is counter-intuitive for many people-shouldn’t Twins have morel-but in fact a Twin’s heavy flywheels mean less engine braking, less unexpected chasis pitch, and thus more predictability and a lower rider workload.
To test at least part of this hypothesis, we took six bikes of varying sportiness to the test track. The machines were chosen for diversity. A 2002 Yamaha YZF-R1 represented Open-class supersport Fours, while an ’02 R6 was used for middleweight Fours. A Buell XB9R had the relatively heavy flywheels typical of classic sporting Twins (think Guzzi and two-valve Ducati), while a Suzuki TL1000R was the extreme version of the quick-revving, light-fly wheeled Open-class Twin, more typical of the Ducati 916/999 or Honda RC51-type Superbike. Another extreme was a 2003 Harley-Davidson Dyna Sport. While the crankshafts of Harley’s Twin Cam models don’t have the sheer mass of Evo-model cranks, they have much more inertia than any sportbike crankshaft, which made the comparison interesting. As a Final envelope stretcher, we added an older Yamaha RZ350 two-stroke.
But before we get to the results, a little explanation on underlying engine physics is in order. Engine braking results from two major causes: 1 ) Engine friction and 2) the pumping losses caused by trying to pull air past a carb(s) closed to idle position. In general, engine friction increases more rapidly than does rpm; the higher you rev, the more friction. Fourstrokes have both many more parts and more friction than do two-strokes-operating all that valvegear eats energy. In addition, the two-stroke cycle enforces such substantial overlap of the intake and exhaust processes that pumping losses are diminished at closed throttle. Typically, two-strokes have had so little engine braking that it’s almost unnoticeable.
What counters engine braking (an energy loss) in fourstrokes is flywheel mass (an energy storage mechanism). Flywheel mass has always been a mixed blessing; it smooths engine power delivery, but the energy stored in the flywheel doesn’t go into acceleration. In the quarter-mile drag, and on the straights of a roadrace course, flywheel mass costs time. For high-performance machines, then, the desirable amount of flywheel has almost always been the minimum required to sustain a smooth idle, and for competition machines, even that degree of “smooth” has been subject to a loose interpretation.
But minimum flywheel mass depends on engine configuration and cylinder count as well. Four-cylinder engines, with a power impulse every 180 degrees of crank rotation, have been able to idle with minimal flywheel and a very light crankshaft. Singles and Twins, with fewer power pulses, need more flywheel at low speed simply to keep the engine ticking over halfway smoothly. The flywheel required for low-speed running in a Twin is far in excess of that required at high rpm, where it acts to moderate engine response-for better and for worse.
Modem four-valve engines have a relatively small amount of valve overlap, and inherently run more smoothly at idle than an equivalent high-performance two-valve engine. Fuel-injection has aided low-speed running, as well. Consequently, current high-performance Fours have their cranks whittled away to a minimum. Such an engine signals this to the rider when he blips the throttle at a stoplight with a sharp zinging Whaaap! Whaaap! Whaaap! and a rapidly dancing tach needle. Try the same thing on a Harley Twin Cam, and the revs barely change. Repeat the performance on a Suzuki TL1000R, and the response is almost the same as a Four.
But enough on theory-back to our testing. At the track, the first thing we did was to measure coast-down performance in neutral for each bike, so we could determine the combined aerodynamic and rolling drag of each bike. From 70 mph, our Stalker radar gun gave us data that allowed a linear deceleration rate to be calculated. Five of the bikes clustered in a tight -0.11 to -0.13 g range (one g represents a change of speed of about 22 mph per second), with only the RZ350 an outlier at -0.16 g-not too surprising, as deceleration is proportional to drag divided by machine weight, and the RZ is by far the lightest of the bunch.
After that, each machine was accelerated to redline in first gear, held at that speed momentarily, and then had its throttle chopped. Because of the different gearing on each bike, the speed at this point was very different, varying from 39 mph on the RZ to more than 70 mph on the R1. The average deceleration over the first 2 seconds was then calculated.
The three Japanese sportbikes, which each reached speeds at least in the 60s, clustered in a narrow' range, decelerating at -0.34 g (R6 and TL) to -0.33 g (Rl). The Harley-Davidson, with its very low first gear, slowed down from 43 mph at -0.22 g, and the Buell from 49 mph at the same -0.22 g. The RZ, as predicted, barely slowed at all. With the diminished aerodynamic drag at 39 mph, it decelerated at only -0.12 g, less than 2.5 mph for each second the throttle was closed.
We then conducted a test that reflected actual riding conditions, with each bike slowing from 70 mph in the lowest gear that would allow it to reach that speed. The R1 had the same result as before, -0.33 g in first gear at 70 mph. The R6 and TL-R each had to be shifted to second to hit 70, and slowed less quickly at -0.24 and -0.27, respectively. The Buell, in second, split the difference between the supersports and the Harley, slowing at -0.23 g. The Dyna and the Yamaha RZ were both in third gear, and slowed at -0.18 g and -0.16 g. respectively.
The results matched subjective impressions closely. When you rolled off the throttle on the Harley or the RZ at redline, nothing dramatic happened. The bikes just rolled forward, barely slowing. Chopping the throttle on the Buell led to a very smooth response, with more slowing than on the above machines, but noticeably less than on the Japanese bikes. For the Rl, rolling off at redline in first was almost violent in comparison, with the -0.33 g slowing just about 10 percent less than would be achieved with a locked rear tire and no front wheel braking. Similarly, the TL-R behaved more like a Four than a Twin, with very strong engine braking (and delightfully quick engine response). The R6 performed almost identically.
The TL-R is fitted with a back-torque limiter in the clutch that is designed to minimize the worst engine braking effects. It did not engage in these relatively smooth, singlegear tests, though it would allow clutch slip and reduced engine braking from an abrupt downshift. The necessity of such devices in roadracing applications was pointed out by this testing; chopping throttle on a high-performance Four or light-flywheclcd Twin in first gear at redline creates sufficient engine braking to take the rear tire reasonably close to lock-up. Throw in increased weight transfer from frontwheel braking, and engine braking alone could use up all rear-tire traction!
The Buell demonstrates both the benefits and costs of increased flywheel. Performance-simulation software indicates it might be another 0.1-second quicker in the quartermile if it had the light crankshaft of the TL-R, and it would rev quicker as well, making matching revs on downshifts that much easier to do. But at its current level of flywheel mass, it has roughly half the engine braking (factoring out rolling resistance and aerodynamics) of an Rl, a very noticeable difference. Almost every road test of the Firebolt has pointed out how smoothly the bike responds to throttle-on/off transitions, and its reduced engine braking is one reason why. (It also helps that Buell has tailored fuel management specifically to complement the effect.)
In the end, our testing left us convinced that engine characteristics are an integral part of motorcycle handling, and that classic sporting Twins are different than supersport Fours and some late Japanese Twins. As for which is superior, that surely depends on where you’re riding and what you want to accomplish. E
