So much has changed in motorcycle GP racing since the high water-mark of the two-strokes, the 1990s.

Tires for two-strokes had to be designed with abuse-tolerance—to survive their explosive torque delivery. As the rider throttles-up to begin acceleration off a corner, a small volume of fresh air-fuel mixture is being blown through the transfer ports into cylinders still containing left-over exhaust gas. So diluted, there is too little air-fuel mixture to fire at all.

As the carburetor slides lift more, the engine begins to fire irregularly, but only after one or more misfires have added enough more mixture to result in something that can be ignited. It does this irregularly—it eight-strokes, then four-strokes, sending yanks down the chain to the rear wheel's drive sprocket.

When the engine finally chimes-in, it does so at perhaps 30% throttle—which is quite a handful. I used to hear the hot disc-valve 250 Aprilias and the Honda NSR500s make a couple of really loud pops, then accelerate.

Not smoothly, either, because as engine revs rise, the pipes begin to pump mixture with steeply increasing force. Riders perceived this as a bang – torque close to doubling as engine rpm came quickly into step with pipe frequency.

Freddie Spencer’s Honda triple of 1983 (his first 500 title) came on the power at 9800 and signed-off at 11,000. Twelve hundred revs to work with.

Yes, but didn’t it have reed intake? Doesn’t that smooth everything out? No, it can’t, because the pipe is what’s doing the pumping. The torque band is the pipe.



The whole history of MotoGP since 2002, on the other hand, has focused on achieving a close match between the shape of engine delivery and the grip available as the rider lifts the bike during corner exit, setting free grip for acceleration as less tire is need for turning. Electronics have been important here, but four-strokes are also fundamentally different. As the rider begins throttle-up, the fresh charge is entering cylinders containing very little exhaust product – because the entire previous stroke – exhaust –has been used to pump out the exhaust. This means that the engine can begin firing at once – maybe not strongly, because here too, pipe design is important – but the rider can feed power from zero – not from 30% throttle. This gets the bike under way sooner, and gets more from the tire because everything has been engineered to match engine torque to tire capability.

Two-strokes could have changed. Around the year 2000 I went to Noale to visit Aprilia. Something that impressed me was their 50 cc two-stroke direct injection scooter engine. It idled like a four-stroke and it delivered torque in a fairly four-stroke-ish manner as well. To make the engine fire from low throttle it was probably necessary to establish a stratified charge – rich enough around the spark plug to fire, with less fuel in the main charge at part-throttle.

This raises another difference. At present, the 1000 cc V4s are limited to 22 liters of fuel for Sunday’s race. A 1990s GP two-stroke 500 could never complete a race on that little fuel.

The old estimate for four-strokes is that they need half-a-pound of fuel per horsepower, per hour of operation. Two-strokes I have seen on the dyno, with fuel-flow instrumentation, need more fuel because so much is lost out the exhaust port during transfer flow. This is inevitable if you are letting gas out on one side of the cylinder, and blowing more in from the other. Mixing is inevitable! The result was that the simple two-strokes of the 1975-1990s period needed from 0.6 to 0.65 pounds per hp per hour - thirty percent more than a four-stroke making the same power. A MotoGP today is roughly 75 miles, but traditional two-strokes would run dry at 55 miles.

You and I know why sanctioning bodies set rules as if trying to turn racing into a Mobil Economy Run: they need to look at least a little bit green – fuel limits, and from 2027, renewable fuels, made by “unburning” atmospheric carbon dioxide to get carbon to combine with hydrogen (electrolyzed from water using wind or solar electricity) to make the very same hydrocarbon species we now refine from fossil fuels.
Aprilia’s little scooter, with direct cylinder injection, flowed no fuel until the exhaust port had closed, so it was much more economical.

Bombardier simplified this idea by injecting the fuel into parts of the transfer flows that would not reach the exhaust port before it closed (transfer port injection).

Some quite decent low-emissions engines were built this way. NOx emissions are low because of exhaust dilution, and CO (carbon monoxide) could be burned to carbon dioxide by a simple catalyst. From the mid-1980s to 1993, I counted 26 automakers with R&D programs to evaluate two-strokes, which are mechanically simpler and cheaper, and occupy 40% less volume than equivalent four-stroke auto engines. I visited GM’s two-stroke project. One by one they gave up – it was cheaper and safer to buy off-the-shelf four-stroke emissions tech than to gamble on being able to push their new two-strokes through emissions. And nobody knew what the emissions regulators might want next.

Too risky.

The motorcycle manufacturers, with far less $$ to commit to two-stroke R&D, made the same decision.

Yet it might have been possible to smooth out two-stroke throttle-up by a combination of stratified charge and direct or transfer port fuel injection, to come up with an engine whose torque delivery could be smoothly controlled.

DFI (direct fuel injection) simple two-strokes (simple in that the underside of the piston is used as the scavenge pump – pushing air through the transfer ports) had much higher crankcase temperatures that reduced power by reducing charge density. But with carburetor fueling, fuel evaporation in the crankcase prevented that. OK, I was pleased to see in the Motegi Museum that late-model NSR500 Honda GP bikes had partially water-cooled crankcases.

Know any bored billionaires, yearning for a fascinating problem on which to spend vast R&D budgets?

The original reason for two-strokes was to provide the very cheapest of motor vehicles – crank, rod, and piston were the only moving parts. Sure, they used more fuel, but first cost was a hurdle for lots of people during the 1930s.

When the FIM set the 1949-2001 basic GP rules, a 125 two-stroke was a 4 hp ring-ding. But by about 1958 they had improved to equal four-stroke power. More power came quickly, so that in 1975 every FIM GP class, including sidecar, was won with two-stroke power.