Technobabble: Spetember 2003

Our readers kick ass. After my June '03 "Technobabble" about Garrett's innovative dual-compressor turbo, I was flooded with geekmail from technodorks across the country. They wanted to tell me about everything from tractor pulls to arcane facts about German aircraft engines at the end of World War II. It wasn't new news, necessarily (my uncle had a series- turbocharged pulling tractor, for the record), but it warms a geek's heart to know he isn't alone in the world.

One of these letters stuck in my head. It was from Dwight Looi, who, as far as I can tell, is a bigger geek than me, since he writes for a PC magazine. His idea was for a turbocharged engine that used VTEC-style variable valve timing to switch between Otto-cycle operation (the Otto cycle is what normal four-stroke gas engines run on) and using the more efficient Miller cycle. Technically, the idea is fascinating for its simplicity.

In the normal suck, squish, bang, blow of the four-stroke engine, there's quite a bit of wasted bang flying out with the blow. But what if you could switch from the normal four-stroke cycle to something that took more of the bang and turned it into zoom?

This something is called the Miller cycle, and it's alarmingly simple. The idea is to have a power stroke (bang) that's longer than the compression stroke (squish) to give more time to convert bang into zoom. Mechanically, banging more than you squish seems impossible. Squish is when the piston goes up, and bang is when it goes down, so a surplus of bang should, it seems, send the piston farther downward with each engine cycle, quickly punching through the oil pan, into the ground, and down toward the center of the Earth as you drive.

Turns out there's an easier solution. Jump back, for a second, to a normal four-stroke Otto cycle engine and realize there are mechanical limitations to how long a stroke you can have before an engine ends up with an intolerably low redline. In other words, you can't really lengthen the stroke that much. The only practical way to have a longer power stroke than compression stroke is to shorten the compression stroke. And given the need to keep the piston going up just as much as it goes down, the only practical way to do this is to not start the compression stroke until the piston is already part way up the bore.

And how exactly do you convince the fuel and air trapped in a cylinder that despite the fact the piston is coming up, there's no need to start compressing just yet? Simply leave the door open. All you have to do is change the shape of the intake cam so the intake valves stay open for a ridiculously long time, even as the piston is coming back up on what would be the compression stroke. Relatively little energy is used pushing air back out the intake valve compared to compressing it in the cylinder, so the amount of work expended compressing air is reduced.

Depending on how late the intake valve closes, you can end up with a compression ratio of, say 9.5:1, but an expansion ratio of 11 or 12:1. The expansion ratio is exactly what it sounds like: how much volume you have at the top of the power stoke vs. how much you have at the bottom.

Now, there are some obvious problems with the Miller cycle. First, there are all those nasty pulses in the intake manifold from the air getting shoved back out the intake valves when the piston starts coming up what it thinks should be the compression stroke.

Then there's the size problem. Shortening the compression stroke effectively reduces the engine's displacement. The air forced back out the intake valves doesn't do anything, so you measure Miller cycle displacement as the volume displaced during the compression stroke after the intake valve has closed. This is why you effectively end up with a smaller engine that takes up the same amount of space and weighs as much as a larger displacement one.

Then there's the power problem. A Miller cycle engine makes better use of the intake charge it allows to stay in the cylinder, but it still doesn't manage to make as much power as it would if the valve just closed at the normal time and the cylinder was allowed to swallow more fuel and air. Think about it: the power stroke is the same in both configurations, but the Otto cycle power stroke has more fuel and air in it. Which would you choose?

This makes the VTEC idea appealing, since it would allow you to switch from a Miller cycle cam profile back to an Otto cycle cam at wide-open throttle. The idea is extra tempting since it could theoretically be done almost entirely with off-the-shelf parts. Take an H22 Prelude engine, for example, and grind the low-rpm lobes with a Miller cycle grind. Then slap on an off-the-shelf turbo or a supercharger and retune the thing. The Honda ECU already has separate fuel maps for high- and low-rpm cams, so you can tune the Miller cycle and Otto cycle modes separately, then play around on the dyno until you figure out when you should switch between them. All this could be done with an off-the-shelf Hondata ECU.

But there's a complication.
Say you set up the Miller cycle with a boost-friendly 8:1 compression ratio and an 11:1 expansion ratio. Guess what happens when you switch to the Otto cycle? You get 11:1 compression, too. Oops. Now you have to cut boost, and suddenly

you aren't making power anymore. Now it's all pointless. Even without the compression problem, the only point to a Miller cycle mode would be to save gas, which ultimately is a way to save money. You'd have to save a lot of gas to pay for all the time you put into figuring out the right cam profiles, juggling compression ratios and tuning this beast.

Now who could afford to spend tons of money to improve gas mileage and has an engine that could handle huge changes in compression while it's making boost? The diesel truck industry. Diesel engines can handle silly things like 18:1 compression with huge piles of boost. And not only does the Miller cycle give them the fuel economy truckers want, but it lowers NOx emissions. Oxides of nitrogen (NOx) come from high combustion temperatures, and those high temperatures come from high compression. Make the turbo do a big portion of the compression (in the form of boost), cool it off before it reaches the cylinder (big intercooler) and run lower compression, and you can end up with the same combustion pressure at a lower temperature.

So once again, what started out as the hot performance engine of the future ended up with the trucks. I hate when that happens.