AT an early stage in the development of i.c. engines, it was realized that to increase their efficiency to the maximum, it would be necessary to run them much faster than was common in contemporary steam engine practice, where speeds rarely exceeded about 100 r.p.m. To do this successfully, it was necessary to improve the design of moving parts so as to reduce weight without sacrificing strength, and also give careful consideration to the matter of balance. One reason why the great majority of i.c. engines have always favored the truck piston, in which the crosshead and piston rod are eliminated, is that it enables the mass of the reciprocating parts to be considerably reduced.
Correct dynamic balance of a reciprocating mass, however, can only be attained by arranging an equal mass to move in the opposite direction in the same plane, at the same speed. This is a tough proposition, which is but rarely approached in practice; most balancing systems in orthodox engines are at best, a fairly satisfactory working compromise between the ideal and the practical expedient.
It is obvious, therefore, that the primary object in using two oppositely moving pistons in one cylinder was to obtain perfect balance; but this system also enables certain incidental advantages to be obtained, which in practice may be just as important, or even more so. On the other hand, it also introduces constructional difficulties and complications, which account for its comparative lack of popularity. Nevertheless, several very successful opposed-piston engines, both of the four-stroke and two-stroke type, have been and still are being produced for stationary, marine, automobile and aircraft duties.
The earliest example of an opposed piston gas engine of which any record can be found is that produced by Sturgeon in 1885. The diagram shown here has been constructed from the incomplete information available, but it is of great interest, as the design seems to be well ahead of its time, and it embodies features which forecast later developments.
A two-throw crankshaft was employed, connected through articulated rods and rocking beams to pistons in the ends of a long cylinder, so that they moved in opposite phase. As the engine operated on the two-stroke principle, a charging pump was necessary. This consisted of a cylinder at right angles to the first, with a piston connected to one of the throws of the crankshaft, thus moving at right angles to the phase of the main pistons.
No details are shown of the fuel supply or carburetion arrangements, but these are presumed to be normal, the combustible mixture was admitted to the pump cylinder by a port which, if located as shown, must have been fitted with some form of non-return valve. By shifting this port towards the outward end of the cylinder, so that it was only uncovered by the piston at the end of its stroke, it would be possible to make this engine completely "valveless".
The mixture was transferred to the main cylinder by a transfer port controlled by one of the main pistons, while the exhaust port was simultaneously opened by the opposite piston. Ignition was effected by means of an incandescent tube, though electric or compression ignition could also be applied.
This design embodies all the essentials of a practical and efficient engine; it is not necessary to explain its functional system in detail, as it is the same as all other two-stroke engines. It should be noted that although dynamic balance of the main pistons is achieved, the pump piston introduces an unbalanced mass, but it can be made comparatively light, and partially cancelled by a crankshaft counterweight. The charging system of such engines has been a thorn in the side of many later designers, and in recent years, rotary pumps or blowers have largely superseded piston pumps for this purpose.
A well known example of an early opposed-piston gas engine was the Oechelhauser, which was of the horizontal type, and made in various sizes with one or more cylinders up to powers of several hundred h.p. This engine employed a crankshaft with three throws per cylinder, arranged crosswise relatively to the cylinder, and with one piston connected to the centre crankpin in the normal way, while the other had a crosshead and yoke, operating through long connecting-rods on to the two side crank pins. The charging pump was usually arranged in tandem with the latter piston.
A considerable amount of research work on opposed piston engines for all purposes was carried out in Germany by Professor Junkers, who is often quoted as the inventor of this type of engine, though this is open to doubt. He produced a variety of designs having different mechanical arrangements, including the Jumo aircraft diesel engine, in which two crankshafts synchronized by a train of gears were employed. This engine was manufactured under license in Britain by D. Napier and Co., under the name of the Culverin engine, but the low power/weight ratio of diesel engines compared with petrol engines limited their practical application in aircraft.
The most successful opposed-piston engines in marine practice are those produced by Fullagar and Doxford, both of which employ a cylinder and crank arrangement similar to that of the Oechelhauser, but disposed vertically instead of horizontally. This makes them tall compared to orthodox types of engines, but they take up little if any more floor space, and the extra height is not necessarily a disadvantage.
Little development of opposed piston engines had been attempted in automobile practice until recent years; the one early example of such an engine was that fitted to the Gobron Brillie car in the first decade of this century. This had a single crankshaft below the cylinders, with long connecting-rods to the upper pistons. Unlike the previous engines mentioned, it was of the four-stroke type. The only other example of a four-stroke I have been able to trace was the curious Sautter-Harle horizontal gas engine. In this, the crankshaft actually passed through a housing in the cylinder itself, and had four throws per cylinder with crossheads and yokes connecting both pistons through side connecting-rods.
Modern development of opposed piston diesel engines for transport has been carried out by Sulzer in Switzerland, and Rootes in this country. In both cases horizontal arrangement of the cylinders is adopted, with the crankshaft disposed centrally underneath, connected through short rods and rocking beams. The Rootes engine has been successfully employed in Commer commercial vehicles, showing high efficiency and fuel economy. In this engine, a rotary displacement blower is employed for charging, but in other respects the mechanical arrangement of both this and the Sulzer engine follows that of the Sturgeon engine.
Both in marine, and locomotive practice, the Napier Deltic engine has opened up new possibilities in packing an enormous power output in this smallest possible space. This is done by arranging banks of cylinders in the form of a triangle, with crankshafts at the corners, each connected to pistons in two adjacent cylinders. The space inside the triangle is occupied by gearing, pumps and other auxiliaries.
Other engines employing the triangular arrangement of cylinders have been designed, in some cases operating on a single central crankshaft through bell crank levers and connecting-rods. Nearly all possible arrangements of the mechanical system of engines using opposed pistons have been tried, or at least suggested, but those illustrated have proved to be the most successful in practice.
The opposed-piston principle is particularly well suited to two-stroke engines, because in addition to improved balance it also enables the ports leading into and out of the cylinder to be located at positions remotely separated from each other, and thus facilitates scavenging by end-to-end flow of the gases. In modem engines, the design of the ports and passages is carefully arranged to produce the right amount of turbulence, so that the exhaust products are driven out without mixing with and polluting the fresh charge. The heat gradient in the cylinder is also unidirectional, i.e. one end hot and the other cool, which not only helps to avoid distortion, but also promotes thermal efficiency.
In diesel engines, where the fuel is not mixed with the charge of air, it is possible to "super-scavenge" the cylinder by increasing the volume of the charging pump over and above the cylinder capacity; the air wasted in this way does not affect fuel economy, as it would in fuel-mixture engines. Supercharging is also possible by timing the port openings so that charging continues after the exhaust ports are closed. This can be done by offsetting the cranks so that piston movements are not exactly synchronous, and though this affects dynamic balance, it is not a serious disadvantage in practice.
Crankcase compression would be difficult to carry out in opposed-piston engines, owing to the necessarily large space required to accommodate internal moving parts, so that only a low charging pressure would be available. It is also impossible to supercharge or super-scavenge the engine unless some form of separate charging pump is employed.
In the early days of the Diesel engine, many breakdowns through cracking of cylinder heads and valve fractures were encountered, and the opposed-piston engine, by the elimination of both cylinder heads and valves, offered a cure for these troubles. Though both the design and metallurgy of engines has vastly improved, the cylinder heads and their attachments still present many problems which have not yet been completely overcome. The cylinders of opposed-piston engines are completely free of end stresses and only have to withstand bursting stress. If they are made in the form of liners, located in the region of the injection nozzle or ignition plug, and allowance made for expansion at either end, they are comparatively free from trouble.
Readers have asked me about the possibilities of opposed-piston engines on a small scale. In my opinion the only serious objection to them is that imposed by their rather awkward shape, which may limit their installation facilities. Some years ago I suggested the use of an opposed piston engine for model aircraft, installed horizontally in the wing, and driving from separate crankshafts, two airscrews synchronized by a geared cross-shaft; but nobody was sufficiently interested to try out the idea.
For propelling a boat or traction vehicle, the Rootes type of engine would be well suited, as it is fairly compact, and the low crankshaft position would be just right for direct coupling to the propeller or transmission shaft. The Deltic system would also offer interesting possibilities.
The Rootes type engine by Mr C. C. Brinton, which was shown at the National Models Exhibition has worked quite successfully, but it is of rather larger capacity than would be required (or most model purposes. It is hoped that a detailed description of this interesting experimental engine will be available in due course.