Editor’s note: This is the first of a three-part series on the topic. Parts two and three will appear in the following weeks. Enjoy!
The obvious hot parts in an internal combustion motorcycle engine are those exposed to the source of all its power—hot combustion gas. These are the cylinder head, piston crown, exhaust valves, and the center wire of each spark plug. Less obvious but still capable of generating heat are connecting-rod bearings (both ends!), main bearings, and the very high pressure between cam lobes and cam followers. Although gears are highly efficient at transmitting power, their friction does generate minor heating.
Back in the 1920s British IC engines pioneer Harry Ricardo learned something interesting: Half of the heat flowing into an engine’s cylinder head enters through the wall of the exhaust port. Combustion inside the cylinder is much hotter than already-expanded exhaust gas, but what makes the exhaust port pick up so much heat is its large surface area and the velocity of the flow through it, which is mostly sonic. Fast-moving, turbulent flows are ideal conditions for rapid heat transfer.
When the ignition spark lights the fuel-air mixture compressed between piston crown and combustion chamber, the fast conversion of the fuel’s chemical energy into heat shoots the gas temperature up 4,800 degrees Fahrenheit. If metal parts were exposed to that for more than an instant, both the head and the piston would quickly melt—aluminum’s melting point is below 1,200 degrees Fahrenheit.
Two things save these parts from otherwise certain destruction: the briefness of combustion (roughly 0.005 second at 6,000 rpm) and aluminum’s ability to rapidly conduct heat away from hot regions. Before the engine fires again, the short shot of heat the parts received will have 540 degrees of crank rotation (720-180) in which to transmit that heat to nearby cooling fins or to liquid coolant—water, antifreeze, or oil.
The above explains two strong trends in modern engine design—the desirability of fast combustion (to shorten the period of intense heat loss from combustion gas into piston and head), and the need to reduce the internal surface area of the exhaust port as much as possible. Engineers have learned to combine high intake velocity with the direction of flow into the cylinder as a means of creating turbulence that will in turn speed up combustion. The reverse is also interesting—as you make intake ports bigger in hope of boosting peak power, midrange suffers. This is because those bigger ports slow midrange flow such that combustion also slows and heat loss increases.
A big change in the cylinder head of Harley’s Evo Big Twin (which entered service in 1984) was its very short and fairly small exhaust port. Both changes were consciously made to try to keep exhaust heat out of the heads. Engines designed to be turbocharged often use ceramic exhaust-port liners to insulate the head from this heating. MV Agusta, as its classic period of European dominance neared its end, provided some of its engines with stainless exhaust-port liners that provided an insulating air gap—47 years ago. Some of the worst-ever exhaust ports have been given to engines that were shoehorned into chassis that had no room for short, straight exhaust ports. The heads on Honda’s beloved Ascot V-twins of the 1980s were particularly bad in this respect. Getting its exhaust ports past frame members required curving them sharply and making them longer—both of which increased surface area, driving excess heat into the heads.
Designers also work to reduce both cylinder-head and piston surface area for the same reason—to keep the valuable heat energy in the combustion gas, rather than letting it get lost in heating up (and potentially weakening) engine parts. When I was first reading about IC engines it was fashionable to praise the hemispherical combustion chamber, or “hemi head.” And while hemi heads definitely have some valuable qualities, low surface area is not one of them. In fact, the classic full hemi, with its two valve stems splayed at 90 degrees (as in vintage Triumph twins) has twice as much surface area as a disc of bore diameter. That’s why today’s combustion chambers are nearly flat.
The tall piston dome necessary to reach high compression ratios in a hemi head also increases piston-crown surface area. Engineers think about these things as they try to stay awake in the endless meetings they are required to attend (so often, most of the meeting is about setting the agenda for the next meeting). English engineer Bert Hopwood, during his very short stay at Norton just after World War II, did something about this excess area. For the export twin Norton needed to harvest US sales, Hopwood made the hemi head much shallower, bringing each cylinder’s two valve stems closer together at a 58-degree included angle. Result? Much less chamber and piston crown surface area to pick up heat from combustion gas. The less heat that flows into the head, the cooler it runs. The cooler combustion chamber surfaces are, the higher the compression ratio the engine can tolerate without detonation (engine knock), and the more torque that engine produces.
Exhaust valves have a difficult life because they are exposed to hot combustion gases on both surfaces. How are they cooled? Most of the heat flow out of an exhaust valve takes the direct route, through the narrow strip of contact between the valve’s 45-degree seating surface and the valve-seat ring in the cylinder head. Some additional heat travels down the valve stem to lose itself through the valve guide. If you pull intake and exhaust valves out of a high-output four-stroke engine, you’ll notice that the intake valve’s seating surface is narrower than that of the exhaust valve’s. This is because 1) intake flow increases as flow across the seat is made smoother (the narrower the seat contact, the better), and 2) the exhaust valve needs that wider seat contact to rid itself of its heat.
As observed by the late F1 engine designer Keith Duckworth, everything improves when you put four valves in place of two. Think of the effect on the distance from the hottest part of the valve—its center—to the valve seat through which most heat flows out of the valve. That distance becomes much shorter for two small exhaust valves than for a single larger one of the same area.
Intake valves run cooler than exhausts despite their greater area because 1) they are heated by exhaust only on one surface, and 2) their other surface is cooled every cycle by the rapid inflow of cool fresh charge. Interesting to note that in a 1936 paper, NACA (the National Advisory Committee for Aeronautics—it became NASA in 1957) measured a 75-degree Fahrenheit rise in intake temperature as the flow passed through the hot intake port and valve, into the cylinder of an engine operating at full power. That’s where a lot of intake valve heat goes—into the airflow.
Parts Two and Three will address piston cooling, iron and aluminum cylinders and heads, liquid-cooling, and more.