The Wright Brothers' Engines and Their Design Part 4

The 4-cylinder vertical engine was a considerable improvement over the previous designs. They had obtained a power increase of about 40 percent, with a weight decrease of 10 percent, and now had an engine whose design was almost standard form for good internal combustion engines for years to come. In fact, had they split the crankcase at the crankshaft center line and operated the inlet valves mechanically, they would have had what could be termed a truly modern design. They needed more cylinder cooling, both barrel and head, particularly the latter, and an opened-up induction system for maximum power output, but this was not what they were yet striving for. They had directly stated that they were much more interested in reliability than light weight.

This engine model was the only one of the Wright designs to be licensed and produced abroad, being manufactured in Germany by the Neue Automobil-Gesellschaft and by Bariquand et Marre in France. The latter was much more prominent and their engines were used in several early European airplanes.

[Ill.u.s.tration: _Figure 11._--4-Cylinder vertical engine a.s.sembly, Bariquand et Marre version. (Drawing courtesy Bristol Siddeley Engines, Ltd.)]

[Ill.u.s.tration: THE WRIGHT BROTHERS AERO ENGINE]

The French manufacturer, without altering the basic design, made a number of changes of detail which seem to have greatly annoyed Wilbur Wright, although some of them could probably be listed as improvements, based on several features of later standard design. One consisted of an alteration in the position of the fuel and oil pumps, the latter being lowered to the level of the sump. The crankcase was drilled to provide forced-feed lubrication to the connecting rod big end and crankshaft main bearings. Strengthening ribs were added to the pistons running from the upper side of the pin bosses to the piston wall, and the crankcase studs holding down the cylinders were replaced with bolts having their heads inside the case. The hinged cam follower was omitted and the pushrod bore directly on the cam through a roller in its end. The magneto was moved toward the rear of the engine a considerable distance and an ignition timing control device was introduced between it and its driving gear. Instead of the magneto being mounted directly on the special bracket integral with the crankcase, a wooden board running from front to rear of the engine was used and this was fastened to the two engine support pads, the magneto bracket being omitted entirely.

Despite his criticism of the French motor and the quality of its manufacture, Wilbur was compelled to install one in his own exhibition airplane during his early French demonstrations at Le Mans after rod failure had broken his spare crankcase, and much of his subsequent demonstration flying was made with the French product.

The Eight-Cylinder Racing Engine

By 1909 regular and special air meets and races were being held and various compet.i.tions for trophies conducted. Among these the Gordon Bennett Cup Race for many years was considered a major event. For the 1910 compet.i.tion it was decided to enter a Wright machine and, since this was a race with speed the sole objective, the available 4-cylinder engine, even in a version pushed to its maximum output, was deemed too small. They built for it a special 8-cylinder unit in a 90V form. They were thus resorting to one of their 1904 concepts--modifying and enlarging a version known and proved in use--as the proper method of most quickly increasing output.

Unfortunately again, there are essentially no detailed drawings available, so that the design cannot be studied.[16]

[Footnote 16: A drawing of the camshaft is held by The Franklin Inst.i.tute.]

Only one engine is historically recorded as having been built, although in view of the Wrights' record of foresight and preparation it is almost certain that at least one spare unit, a.s.sembled or in parts, was provided. In any case, the airplane--it was called the _Baby Grand Racer_--and engine were wrecked just before the race, and no physical parts were retained, so that the sole descriptions come from external photographs, memory, and hearsay. McFarland thinks that possibly Orville Wright, particularly, was somewhat discomfited over the accident that eliminated the machine, as he had previously flown it quite successfully at a speed substantially higher than that of the ultimate winner, and he wanted to get it out of sight and mind as quickly as possible. The Air Force Museum at Wright Field, Dayton, Ohio, has an incomplete set of drawings of a 90V, 8-cylinder Wright engine, but it is quite obvious from the basic design and individual features, as well as from at least one date on the drawings, that this conception is of a considerably later vintage than that of the _Baby Grand Racer_.

The racing engine was in essence a combination of two of the standard 4s on a redesigned crankcase utilizing as many of the 4-cylinder engine parts as possible. The rods were reported to have been placed side by side, and the regular 4-cylinder crankshaft, with alterations to accommodate the rods, was utilized. A single cam operated all the exhaust valves. It was compact and light, its only fundamental disadvantage being the inherent unbalance of the 90V-8. The arrangement provided a much higher powered unit in the cheapest and quickest manner, and one that could be expected to operate satisfactorily with the least development.

The Six-Cylinder Vertical Engines

Shortly after the construction of the 8-cylinder engine the Wrights were again faced with the ever-recurrent problem of providing a higher powered standard production engine for their airplanes, which were now being produced in some numbers. By this time, 1911, there had been a relatively tremendous growth in both flying and automotive use of the internal combustion engine and as a result many kinds and sizes had been produced and utilized, so that numerous choices were presented to them. But if they were both to make use of their past experience and retain the simplicity they had always striven for, the more practical possibilities narrowed down to three: they could increase the cylinder size in the 4-cylinder combination, or they could go either to 6 or 8 cylinders in the approximate size they had previously used.

[Ill.u.s.tration: _Figure 12._--Original 6-cylinder engine: a, push-rod side; b, valve-port side; c, crankcase with sump removed. (Photos: Smithsonian A-3773A, 45598; Pratt & Whitney D-15015, respectively.)]

The 4-in. cylinder in combination with a 5-in. stroke would provide in four cylinders about the displacement they wanted. Strokes of 6 in.

were not uncommon and cylinders of 6-in. bore had been very successfully utilized in high-output automobile racing engines many years before this, so there was seemingly no reason to doubt that the 5-in. cylinder could be made to operate satisfactorily, but it is not difficult to imagine the Wrights' thoughts concerning the roughness of an engine with cylinders of this diameter. The question of the grade of available fuel may possibly have entered into their decision to some extent, but it seems far more likely that roughness, their perennial concern, was the predominant reason for not staying with the more simple 4-cylinder form (as we have seen, roughness to them meant the effect of the cylinder explosion forces). Actually, of course, they never went larger than a 4-3/8-in. cylinder bore, and later aircraft engine experience would seem generally to confirm their judgment, for with the piston engine it has always been much more difficult to make the larger bores operate satisfactorily at any given specific output.

While the 90V, 8-cylinder arrangement would have enabled them to utilize a great number of the 4-cylinder-engine parts, it would have given them a somewhat larger engine than was their apparent desire, unless they reduced the cylinder size. And while they had had some limited experience in building and operating this kind of engine, and twice had chosen it when seeking more power, both of these choices were greatly influenced by the desire to obtain quickly an engine of higher power. It is also possible that something in their experience with the V-8 moved them away from it; the unbalanced shaking force inherent in the arrangement may well have become evident to them. What probably also helped them to their final conclusion was the fundamental consideration that the V-8 provided two extra cylinders which were not really needed.

The eventual selection of the 6-cylinder was a slight compromise. In order to get the desired output the cylinder displacement was increased, but this was done by lengthening the stroke--the first time this had been altered since the original design. The increase (from 4 to 4-1/2 in.) was only 1/2 in., and the bore, the more important influence on fuel performance, was kept the same. Overall, the choice was quite logical. They were utilizing the in-line construction upon which almost all of their now considerable experience had been based, and the sizes of and requirements for parts also conformed to this experience. They could, in fact, use many of the same parts. The natural balance of the 6-cylinder arrangement gave them a very smooth engine, and had they stiffened the shaft and counter-weighted the cranks, they would have produced the smoothest engine that could have been built at that time.

In the literature are two references to a Wright 6-cylinder engine constructed around the cylinders of the vertical 4. One of these is in Angle's _Airplane Engine Encyclopedia_, published in 1921, and the other is in _Aerosphere 1939_, published in 1940. The wording of the latter is essentially identical with that of the former; it seems a reasonable conclusion that it is a copy. Although it is possible that such an engine was built at some time, just as the 8-cylinder racing engine was cobbled up out of parts from the 4-cylinder vertical, no other record, no engines, and no ill.u.s.trations have been found. It is thus quite certain that no significant quant.i.ty was ever manufactured or utilized.

The crankcase was considerably changed from that of the vertical 4, and was now in two pieces, with the split on the crankshaft center line. However, the shaft was not supported by the lower half of the case, as eventually became standard practice, but by bearing caps bolted to the ends of the upper case and, in between, to heavy ribs running across the upper case between the cylinders. The lower half of the case thus received none of the dynamic or explosion loads, and, serving only to support the engine and to provide for its mounting, was lightly ribbed. In it were incorporated integral-boss standpipe oil drains which discharged into a bolted-on sump. The upper half of the case was again left open on one side, giving the desired access to the interior, and, additionally, the design was altered to provide a method of camshaft a.s.sembly that was much simpler than that of the vertical 4 (see p. 42).

The cylinder was also greatly altered from that of the vertical 4. It was made in three parts, a piece of seamless steel tubing being shrunk on a cast-iron barrel to form the water jacket, with a cast-iron cylinder head shrunk on the upper end of the barrel. This construction compelled the use of long studs running from the cylinder head to the case for fastening down the cylinder (see Figures 12a-c). For the first time the cylinder heads were water-cooled, cored pa.s.sages being provided, and more barrel surface was jacketed than previously, although a considerable area at the bottom was still left uncooled, obviously by direct intent, as the ported exhaust arrangement was no longer employed.

Also for the first time one-piece forged valves were used, but just when these were incorporated is not certain and, surprisingly, they were applied to the inlet only, the exhaust valve being continued in the previous two-piece screwed and riveted construction. The reasoning behind this is not evident. If a satisfactory two-piece exhaust valve had finally been developed it would be logical to carry it over to the new design; but exhaust valves normally being much more troublesome, it would seem that a good exhaust valve would make an even better inlet valve and, in the quant.i.ties utilized, the two-piece design should have been much cheaper. In the original 6-cylinder engine the inlet valves operated automatically as in all previous models, but at the time of a later extensive redesign (1913) this was changed to mechanical actuation, and the succeeding engines incorporated this feature. All the valve-actuating mechanism was similar to that of the vertical 4, and the engine had the usual compression-release mechanism, the detail design being carried over directly from the 4-cylinder.

Design of the piston followed their previous practice, with wide rings above the pin and shallow grooves below the pin on the thrust face, and with the pin fastened in the piston by a set screw. The piston had four ribs underneath the head (see Figure 13b) radiating from the center and with the two over the pin bosses incorporating strengthening webs running down and joining the bosses. The piston length was reduced by 1 in., thus giving a much less clumsy appearance and, with other minor alterations, a weight saving of 40 percent (see Figures 13b and c). The rods were for the first time made of I-section forgings, a major departure, machined on the sides and hand finished at the ends, with a babbit lining in the big end, the piston pin bearing remaining steel on steel.

[Ill.u.s.tration: _Figure 13._--Original 6-cylinder engine: a, cylinder a.s.sembly and valve parts; b, bottom side of piston; c, piston, piston pin and connecting rod; d, valve mechanism; e, crankshaft and flywheel. (Pratt & Whitney photos D-15012, 15017, 15013, 15018, respectively.)]

At least two different general carburetion and induction systems were utilized, possibly three. One, and most probably the original, consisted of a duplicate of the injection pump of the 4-cylinder fitted to a manifold which ran the length of the engine, with three takeoffs, each of which then split into two, one for each cylinder. Of this arrangement they tried at least two variations involving changes in the location and method of injecting the fuel into the manifold; and there seems to have been an intermediate manifold arrangement, using fuel-pump injection at the middle of the straight side, or gallery, manifold, which was fed additional air at both ends through short auxiliary inlet pipes. This would indicate that with the original arrangement, the end cylinders were receiving too rich a mixture, when the fuel in the manifold was not properly vaporized.

Although the exhaust was on the same side of the engine as the inlet system, no attempt was made to heat the incoming charge at any point in its travel. An entirely different system adopted at the time of the complete redesign in 1913 consisted of two float-feed Zenith carburetors each feeding a conventional three-outlet manifold. This carburetor was one of the first of the plain-tube type, that is, with the airflow through a straight venturi without any spring-loaded or auxiliary air valves, and was the simplest that could be devised. When properly fitted to the engine, it gave a quite good approximation of the correct fuel and air mixture ratio over the speed-load running range, although it is considerably more than doubtful that this was maintained at alt.i.tude, as is stated in one of the best descriptions of the engine published at the time the carburetors were applied.

The compression ratio of this engine was lowered by almost 20 percent from that of the vertical 4. This, in combination with the low bore-to-stroke ratio, the unheated charge, and the later mechanically operated inlet valve, indicates that the Wrights were now attempting for the first time to secure from an engine something approaching the maximum output of which it was capable.

As the engine originally came out, it continued to utilize only one spark plug in each cylinder. The high-tension magneto had a wide range of spark advance adjustment, which again provided the only control of the engine when equipped with the original fuel pump injection.

The location of the valves and pushrods was similar to that in the 4, so that the cams were immediately adjacent to the camshaft bearings, which were carried in the crankcase ends and in the heavy webs. The camshaft was gear-driven and the cam shape was similar to that of the last 4s, with a quite rapid opening and closing and a long dwell, leaving the valve opening accelerations and seating velocities still quite high.

The crankshaft was a continuation of their basic design of rather light construction, particularly in the webs. The cheeks were even thinner (by 1/4 in.) than those of the 4 although the width was increased by 1/8 in. (see Figure 13e). For the first time they went to a forging, the rough contour type of the time, and utilized a chrome-nickel alloy steel.

Lubrication was by means of the usual gear pump, and the piston and rod bearings continued to be splash-fed. The rod big-end bearing carried a small sharp undrilled boss at the point where, on the other engines, had been located scuppers whose purpose was apparently still to throw lubricating oil on the cylinder wall carrying the more highly loaded side of the piston. The rod big-end bearing was lubricated by a hole on the top of the big-end boss catching some of the crankcase splash, which was then carried to the bearing by a groove.

When the 6-cylinder engine was completely redesigned in 1913 this led to the introduction in late fall of that year of a new model called the 6-60, the 60 designating the rating in horsepower. There is little in the Wright records to show why such a radical revision was thought necessary, but the general history of the period gives a rather clear indication. The compet.i.tion had caught up to the Wrights in powerplants. Other engines were being installed in Wright airplanes, and Navy log books show these other engines being used interchangeably with those of the Wrights.

Most of the descriptions of the new model published at the time it was introduced concentrate on the addition of the two carburetors and the mechanical operation of the inlet valves, but these were only two of many major changes. The cylinder was completely revised, the intake being moved to the camshaft side of the engine from its position adjacent to the exhaust, so that the two ports were now on opposite sides of the cylinder. By proper positioning of the rocker-arm supports and choice of their length and angles, all valves were made operable from a single camshaft. The shrunk-on steel water jacket cylinder was retained, but the water connections were repositioned so that the water entered at the bottom and came out at the top of the cylinder. Over the life of the 6-cylinder engine several different valve types were used but the published specifications for the model 6-60 called for "cast iron heads"--the old two-piece construction. The piston pins were case hardened and ground and the crankshaft pins and journals were heat treated and ground.

The fuel and oil pumps were removed from the side of the crankcase and a different ignition system was applied, although still of the high-tension spark-plug type which by this time had become general practice on all so-called high-speed internal-combustion engines. A second threaded spark-plug hole was provided in the cylinder head and despite its more common use for other purposes, it is evident that the intention was to provide two-plug ignition. It is doubtful that at the specific output of this engine any power difference would be found between one-and two-plug operation, so that the objective was clearly to provide a reserve unit in case of plug failure. However, it was also used for the installation of a priming c.o.c.k for starting and because of the prevalence of single-wire ignition systems on existing and ill.u.s.trated engines, it seems to have been used mostly in this manner, even though dual-ignition systems later became an unvarying standard for aircraft engines.

Viewed externally, the only part of the engine that appears the same as the original 6 is the small lower portion of the crankcase; but what is more visually striking is the beauty of the new lines and extreme cleanness of the exterior design (see Figures 14 and 15). Many of their individual parts had shown the beauty of the spa.r.s.e design of pure utility but it was now in evidence in the whole. Despite the proven practical value of their other models, this is the only one that can be called a good-looking engine, instantly appealing to the aesthetic sense, even though the vertical 4 is not an ugly engine. The appearance of their final effort, in a field they were originally reluctant to enter and concerning which they always deprecated the results of their own work, was a thing of which a technically trained professional engine designer could be proud.

The 6-60 was continued in production and development until it became the 6-70, and indications are that it eventually approached an output of 80 horsepower.

[Ill.u.s.tration: _Figure 14._--6-Cylinder 6-60 and 6-70 engine, right rear intake side. (Pratt & Whitney photo.)]

[Ill.u.s.tration: _Figure 15._--6-Cylinder 6-70 engine, incorporating flexible flywheel drive, exhaust side. (Smithsonian photo A-54381.)]

Minor Design Details and Performance of the Wright Engines

In the Wright brothers' various models were many minor design items which altogether required a great deal of consideration, but which did not materially affect overall engine performance. The results generally could all be cla.s.sed as good practice; however, one of these utilized in the 4-cylinder vertical engine was rather unorthodox and consisted of offsetting the cylinders with relation to the crankshaft.

This arrangement, which can be seen in the drawing (Figure 11) was apparently an attempt to reduce the maximum side load on the piston during the power stroke, but since the peak gas loading usually occurs at about 10 to 15 percent of the power stroke, this probably did not have much effect, and it was not carried over to the 6-cylinder design.

All engine bearings were of the plain sleeve type and, except for the bronze and steel bearings in the connecting rod, were of babbit. The advantages of babbit for bearings were discovered very early in the development of the mechanical arts, and apparently the Wrights never encountered a bearing loading sufficiently high to cause a structural breakdown in this relatively weak material.

Valve openings show no variation through the successive production engines, although the Wrights most probably experimented with different amounts. The 1903 engine and the vertical 4-and 6-cylinder all had lifts of 5/16 in., but the valve-seat angles varied somewhat; the records show included angles of 110 to 90--not a large difference.

The valve-operating mechanism was the same from the first vertical 4 onward. The high side thrust caused by the cam shape required for the very rapid valve opening they chose was, no doubt, the reason for the use of the hinged cam follower, and since the same general cam design was used in their last engine, the 6-cylinder, the same method of operation which had apparently proved very serviceable was continued.

How satisfactory was the considerably simpler subst.i.tute used in the Bariquand et Marre version of the 4-cylinder engine is not known.

Possibly it was one of the alterations in the Wrights' design that Wilbur Wright objected to, although in principle it more closely conforms to the later fairly standard combination valve tappet and roller construction: The available drawings do indicate, however, that the cam of the Bariquand et Marre engine was also altered to give a considerably less abrupt valve opening than the Wright design, so that there was less side thrust. For the Wright 6-cylinder engine their 4-cylinder cam was slightly altered to provide a rounding off near the top of the lobe, thus providing some reduction in the velocity before maximum opening was reached. All their cam designs indicate a somewhat greater fear of the effect of seating velocities than of opening accelerations.