The higher the speed, the higher will be the output of an engine. The formula for the output also contains the factor n, the rotational speed of the engine. This theoretical consideration cannot be fully translated into practical terms. With increasing engine speed, the piston speed increases and the fractional losses become higher. At the same time, the mean effective pressure diminishes because of the higher resistance encountered in the inlet and exhaust system i.e. throttling effect on the gas flow.

This in turn reduces the volumetric efficiency. Besides, the inertial forces developed by the reciprocating parts of the crank and valve mechanisms are not allowed to exceed certain values, otherwise damage is liable to occur. When the cubic capacity for a new engine design has been determined, the influences of high speed that adversely affect power output and engine life can be largely obviated by a suitable choice of the number of cylinders, the stroke-bore ratio, and the piston speed.

The engines used in ordinary present-day cars have rotational speeds of between 5000 and 6000 rpm a range that only a few years ago was reserved for sports-car engines. Racing engines have meanwhile moved up into the 11,000-14,000 rpm range, though this result has been achieved only with considerable effort and cost.

The total cubic capacity i.e., the total piston-swept working volume VH of an engine should be divided over the largest possible number of cylinders, to ensure that the reciprocating masses of the individual pistons and connecting rods will be small. The lighter these components are, the easier and less power-consuming will be their acceleration and deceleration at the ends of the piston stroke. For a given cubic capacity, the capacity of the individual cylinder is reduced, the bore and stroke are likewise reduced, and the piston speed is lower.

However, an increase in the number of cylinders also has its drawbacks. For one thing, there are now more bearings in which friction occurs. In addition, the cost of manufacture goes up because of the more numerous components that have to be made, machined and assembled. For reasons of economy, the cubic capacity of a cylinder of an ordinary car engine is normally between 250 and 500 cc. A racing car engine usually has many relatively small cylinders ranging from, for e.g. 62cc (Honda) to about 200cc.

In addition to dividing the total cubic capacity among a large number of cylinders each of relatively small capacity, other measures to reduce the reciprocating masses of the pistons and crank mechanism consist in the use of light-alloy pistons and connecting rods made from titanium, a metal not unlike steel, but lighter.

When the capacity of the individual cylinder has been determined, the stroke s and the bore d can be determined from the stroke-bore ratio (s/d) that has been chosen. As a rule this ratio is somewhere between 0.7 and 1.0. It should be as low as possible for high-speed engines, so that the cylinder bore is larger than the stroke; i.e., the cylinder is relatively wide, making possible the use of large valves. Besides, the piston speed is then also lower, so that the frictional and throttling losses during the suction stroke are less.

At high speeds the crankshaft functions under severe stress conditions because at each power stroke it is subjected to sudden impactlike torsional loading. The crankshaft must therefore be of very rigid construction; it must not deflect. Better resistance to deflection is obtained by closer positioning of the crankshaft bearings (usually called the main bearings).

Efficient design of the valve mechanism is of major importance in high-speed engines because accurate valve timing at all rotational speeds is essential. This calls for rigid and vibration-free construction. The valve is opened against the closing action of a spring; the force developed by the spring should be sufficiently powerful to ensure that all speeds the valve motion accurately conform to the shape of the cam.

At high speeds there is only very little time available in which closure of the valve can be affected, a mere fraction of a second. To keep the spring force needed for this within reasonable limits, the weight of the reciprocating valve parts in a high-speed engine should be reduced to a minimum. There are various methods of achieving this. Dividing the total cubic capacity among a large number of cylinders permits the use of correspondingly smaller and lighter valves.

The high speeds of present-day engines have been attained partly as a result of using overhead camshafts, thereby eliminating transmission elements which make the valve mechanism slower and more cumbersome. For high-speed engines the arrangement in Fig.1a is preferable to that in Fig.1b because the moving masses in the former are smaller. Sports-car and racing-car engines have hemispherical combustion chambers, so that the valves have to be inclined.

For this reason each row of valves is provided with its own camshaft. This solution is too expensive for the engines of ordinary cars. Alternatively, two rows of inclined valves can be actuated by one camshaft (Fig.2), though in this arrangement the rockers constitute a larger moving mass. Fig.3 shows a different overhead camshaft arrangement embodying a tappet.

In ordinary car engines, the overhead camshaft is usually driven by a chain from the crankshaft and at half the speed of the latter. To avoid objectionable noise arising from wear and thermal expansion, the chain is kept under uniform tension by a tensioning device. In some instances a silent valve drive in the form of a toothed plastic belt (reinforced with steel wire) is used instead of a metal chain. The camshaft drive systems illustrated in Figs.1 and 2 are suitable for engine speeds up to 7000rpm.

In racing engines which operate at considerably higher speeds, the overhead camshafts are driven though the agency of gear systems or bevel-geared shafts. Such systems are preferable to chain drives because they are free of vibration and backlash effects. They are of course, also more expensive.

Another means of reducing the weight of the valves consists in using valves with hollow stems. To improve the heat conduction and cooling of the exhaust valves, which becomes very hot, their stems are partly filled with sodium. At the high working temperatures the sodium is molten and its movements help to conduct heat from the valve head to the cooler parts of the stem, thus cooling the head (Fig.4).

As an alternative to one large and heavy valve it is possible to employ two smaller, lighter valves. Thus the cylinders of some racing engines are each provided with two inlet valves and one exhaust valve. This is a very expensive form of construction and therefore unsuitable for ordinary engines. Various types of valve embodying positive actuation, as distinct from the spring-controlled reciprocating action of the usual poppet (or mushroom) valve envisaged here, have been devised, including more particularly the rotary valve, but have never achieved much practical significance.