Hot Metal

Hot Metal

KEVIN CAMERON

TDC

THINK ABOUT THE HEAT FLOWING into and out of an engine's cylinder head. During combustion, turbulent glowing gases drive heat into the piston crown, cylinder head and valves. During exhaust, the head of the exhaust valve is surrounded on all sides by hot gas, but with a difference. Now, the gas is flowing at high speed—more than 2500 feet per second. This very turbulent flow produces ideal conditions for rapid heat transfer because rotating turbulence cells continually bring fresh, hot gas from within the flow into direct contact with the valve. The same conditions prevail within the exhaust port itself.

Near the end of the exhaust event, as the exhaust valve approaches its seat, wisps of fresh charge from the nowlifting intake valve pass out the exhaust with some cooling effect on the valve. During the 1950s, when Alfa Romeo raced a supercharged 1.5-liter Grand Prix car, its engineers called this period of exhaust-valve cooling by intake overlap flow "the fifth cycle." To make its exhaust valves survive in what would be otherwise unsurvivable conditions, Alfa used very long overlap timing so the hot valve was cooled by this flow. In the process, the car's mileage went down to 1.5 mpg.

As the exhaust valve re-seats, heat from its now very hot head begins to flow into the valve seat, which is itself being cooled by the flow of heat from it into the surrounding aluminum of the head. If this is an air-cooled engine, the head material around the exhaust valve becomes very hot in racing operation— hot enough to slowly yield to the distorting forces of the head bolts and the shrink-fitting of the exhaust valve's seat insert. In a liquid-cooled engine, the material around the exhaust valve(s) is much cooler because nearby water passages, filled with high-speed turbulent flow of their own, rapidly pick up heat and carry it away.

If this is a four-valve engine, the material between the two exhaust-valve seats, called "the exhaust bridge," experiences special distress because it is being heated from the combustion side in the normal manner and by heat flowing toward it from both exhaust ports on the downstream side. If the bridge becomes too hot, its thermal expansion pushes hard enough on surrounding metal to force it back a bit, making it yield permanently. When parts cool after shutdown, the material in the bridge comes into tension. Given enough of these cycles, the bridge cracks. In its large, four-valve, horizontally opposedTwins, BMW circulates oil through these bridges, keeping them from getting hot enough to crack in this way.

Even in two-valve engines, this expansion/yield/contraction cracking mechanism sometimes produces cracking between the exhaust valve seat and a sparkplug hole.

Because so many motorcycle engines are now liquid-cooled, exhaust-valve problems are rare, but the classic failure was to break right at the point where the stem flares out to form the head. This part of the valve is right in the middle of the hot exhaust flow, and in an aircooled engine, special valve materials taken from the gas-turbine industry may be necessary to achieve reliability.

In its XR1200R, Harley-Davidson circulates oil behind the exhaust-valve seats. If you look at how air-cooled heads are made, you can see that getting heat from the exhaust seats to cooling fins is not easy, especially from the region between the bottom of the exhaust port and head gasket.

It has been noted that about half the heat picked up in a cylinder head enters the metal through the walls of the exhaust port(s). This makes perfect sense because a) the surface area of the port is comparable to that of the chamber; and b) although both the head and piston crown pick up heat, the heat transfer in the exhaust port is extremely rapid.

It is therefore just common sense to make the surface area of the exhaust port as small as possible by both making the port short and of minimum diameter. Despite this, you will see exhaust ports on some engines that are excessively long and, in some cases, have 90-degree bends inside the head, usually to move the exhaust pipe to someplace where it doesn't conflict with a frame member. One example is the rear head on the V-Twin Honda Super Hawk.

One argument is, "Well, this engine is liquid-cooled, and carrying away a bit of extra heat is no big deal." Okay, point noted. But even so, more heat means a bigger radiator, which means more aero drag and more fuel consumption. Better to have that energy stay in the exhaust gas, where its energetic wave action can contribute to power. And if the engine is air-cooled, extra heat entering the head from a long and/or over-large exhaust port just pushes up head temperature, further limiting the compression ratio we can get away with.

Finally, there's the argument that many heads are "cooled backward." If air or liquid cools the exhaust side first and then flows to the intake side, we are in effect heating the intake side with exhaust heat. That extra heat is expanding the fresh charge as it rushes through the intake port(s), reducing its density and cutting torque a bit. In that view, cooling the intake side first, and then the exhaust side, makes better sense. Some liquid-cooled engines now use this sequence, and a classic air-cooled example is Pratt & Whitney's 28-cylinder R-4360 radial aircraft engine, whose design began in 1940.

Finally, there's the piston. The cooler we can make it run, the higher the compression ratio we can get away with. In many early engines, it was assumed that only the top of the cylinder really needed cooling, so either the cooling fins were tapered (largest just under the head) or the water jacket came only halfway down the cylinder. But if the piston is too hot, isn't it too hot at the bottom of its stroke as well as at the top? The counter argument is that often the lower part of the cylinder projects into the crankcase, where it is cooled by oil splash. In modern large-bore designs, in which heat has a long way to go from bore center to cylinder wall, it may need help from piston-cooling oil jets.

Heat in the combustion gas produces power. Heat anywhere else produces problems, so we devise systems to carry it away. □