The Allison

by William Levenor

After weathering years of alternate praise and condemnation, the Allison liquid-cooled engine today is proving itself in action.

The question of the relative merits of liquid-cooled and air-cooled aircraft engines has been for years an excuse for the bandying of words in the public print.

Inquire closely, and it becomes evident that the most eminent aeronautical engineers are as yet unable to reach any substantial agreement on the subject — a fact that fails entirely to check the dogmatic assertions of the enthusiasts on both sides.

The feud has meant in this country frequent criticism of the Allison engine, the 12-cylinder Vee-type high-horsepower plant that powers most of the Army's standard fighters. This criticism, moreover, is by no means confined to the air-cooled versus liquid-cooled controversy, but extends to the Allison versus other liquid-cooled engines — Rolls-Royce, Daimler Benz, Junkers Jumo, Hispano-Suiza — with the all-too-frequent result that the writer concludes that, of all the world's liquid-cooled aircraft engines, the Allison is incomparably the worst.

One "military analyst" went so far as to assert that the "inferiority" of American fighter planes was due to the fact that the Allison could not be supercharged! That opinion could have been based only on (a) some startling oversights, plus (b) a rather silly assumption. For (a) two examples will suffice — the Lockheed Lightning had been designed around two turbosupercharged Allisons several years before the "military analyst" decided it couldn't be done; the shift from Allison to Rolls-Royce in the Warhawk, latest of the Curtiss P-40s, produced no extraordinary changes in altitude characteristics of the plane. This brings us to (b) — the assumption, not only silly but quite prevalent, that unless a fighter is a high-altitude plane, it is "inferior."

On the other hand, the liquid-cooled point of view has been set forth with equal enthusiasm and equal disregard for accuracy and restraint.

The development of commercial aviation in the United States, a development greater than in any other nation, has as everyone knows been accompanied by the almost universal use of the radial air-cooled engine.

A number of generally accepted facts account in large part for this trend. The radial air-cooled engine has been considered simpler in construction. In practice, the air-cooled was supposed to operate about twice as long between major overhauls as did the in-line liquid-cooled engine. Its output of power per pound of weight had been greater in the past — on the average the air-cooled weighed about 1.6 pounds, the liquid- cooled about two pounds, per horsepower. [The Allison today is down to 1 pound per horsepower. —ED] All these are factors making for economy of operation, which obviously is a matter of prime importance in commercial aircraft.

On the other hand, the liquid-cooled possesses its own inherent advantages over the air-cooled; advantages given close attention in the design of military planes in which speed and power for combat come ahead of economical operation. Among the advantages generally claimed for the liquid-cooled are these:

  1. The engine can be operated at a higher power output per cubic inch of piston displacement due to the type and uniformity of cooling;
  2. because of the heat capacity of the liquid coolant, the engine is less sensitive to temporary overloads, and therefore more reliable in the stresses of combat;
  3. the frontal area is smaller, with consequent reduction in drag;
  4. there is greater flexibility of engine location. The latter point is based on the fact that the cooling element is independent of the engine, so far as dimensions and location are concerned, which frees the designer of many restrictions imposed by air-cooled engines.

It should be pointed out here that the argument about frontal area and its drag (a point giving rise to the liquid-cooled enthusiasts' charge that radials are "built-in headwinds") is by no means ended.

It's axiomatic that the horsepower required to overcome wind resistance increases as the cube of the speed; double the speed of a plane, and the power must be increased eight times. Now the frontal area of a 1,200-hp air-cooled engine is about 1,800 square inches, that of a 1,200-hp liquid-cooled engine about 900 square inches — a comparison that ex plains the opinion of many engineers that at speeds over 400 mph, even with optimum cowling for the radial, the smaller frontal area may give the liquid-cooled an advantage of 20 mph.

On the other hand, equally competent and learned engineers point out that as the power goes up, so does the size of the propeller and of the spinner. This, they say, soon obviates the streamlining advantage of the in-line engine, since the radial can be hidden behind the large spinner. Some go on to the assertion that the radial's drag actually can become less than that of the in-line engine, because there is no exposed radiator.

Is this opinion based on an assumption that the size of the radial engine does not increase with the rise in power output? That would seem to be hardly a safe assumption in the light of experience to date. And while no successful in-line aircraft engine of 2,000 hp or better has yet proved itself, Allison some time ago upped the rating of its standard engine by about 300 hp, and at the same time reduced the frontal area by 10 per cent as well as achieving cuts in other dimensions and in weight.

Settlement of the controversy will come, if ever, out of wind-tunnel experiments and flying tests. And for the present, the in-line liquid-cooled engine powers most of the world's top-notch fighter planes. The Allison is not the only American engine of this type, but it is the only one now in general use. What is its background?

Back in 1915, James A Allison, one of the founders of the Indianapolis Speedway, opened the Allison Speedway Team Company, a 20-man shop, to build and service racing cars for his three teams.

Two years later the shop's precision facilities were placed at Government disposal, and Allison entered the aircraft industry for the first time, building tools, jigs, fixtures and experimental models for the production of Liberty engines. Orders for production models of superchargers, whippet tanks, trucks and high speed tractors expanded the shop to 100 mechanics and 150 designers and engineers, and it emerged from the war as the Allison Experimental Company.

After an Allison car won the 500 mile international sweepstakes in 1919, the Allison Company abandoned motor car racing to devote full time to experimental work, keeping its organization together despite the disappearance of Government orders after the war ended.

In 1927, the Army and Navy asked Allison to rebuild 2,000 of the 10,000 Liberty engines produced during World War I. One assignment was to rebuild a Liberty to use 20 per cent benzol, 80 per cent aviation fuel, in an attempt to increase horsepower from 400 to 570. The attempt was not a success — connecting rods in the rebuilt engine broke down; but the project aroused the interest of Allison and his shop superintendent, Norman H Gilman, in the development of a high-power liquid-cooled engine at a time when the attention of American aviation engineers was being devoted largely to radials, which had powered the sensational transoceanic flights of the 1920's and established themselves as the engines for commercial transports. Allison was almost the only American liquid-cooled laboratory to pace the European liquid-cooled aviation development.

Allison died in 1928, and his precision engineering company was acquired by General Motors the following year. Gilman remained head of the organization and obtained General Motors financial and research backing for his pet project — the development of a 1,000-hp liquid- cooled engine.

Before undertaking this work, Gilman tackled a preliminary job — finding a substitute for the bronze-backed babbitt bearings which had left much to be desired in the Liberty engine. Allison engineers perfected steel-back, leaded bronze bearings that could withstand the greater loads required for more power output, and these "copper lead" bearings later became standard in most high-powered aircraft engines.

This accomplished, the problem of the engine itself was approached. The earliest efforts were devoted strictly to laboratory research, and the General Motors Research Laboratories in Detroit, headed by C F Kettering, contributed considerable assistance. By 1933, the Army Air Corps had become interested in the development. Glycol ethylene (base chemical of Prestone) with its boiling point of 357°, presented possibilities as a coolant that would permit running an engine at 250° to give higher power output than a 160° water-cooled engine and control advantages over the 400° air-cooled engine. An unsuccessful attempt to substitute glycol for water in a Curtiss V-12 Conqueror engine demonstrated the leakage problem involved in the use of the oily, fast-running chemical as a coolant.

The answer to this problem was an Allison-designed coolant unit that could be tested for pressure and temperature prior to assembly. This unit, which pumped glycol six times as fast as water, could cool an engine effectively with one-third the radiator size of a water coolant unit. Its development forecast a lighter engine and the possibility of realizing the goal of one pound of engine weight per horsepower. The new coolant unit was a feature in the first Allison V-1710-A, built in 1931, for the US Navy. It was found necessary to modify several parts after half of a 50-hour development test. The following year, the modified engine rated 750 hp at 2,400 rpm in a successful development test. Redesigned engines to serve as reversible unsupercharged engines were ordered to replace German liquid-cooled engines in the dirigibles Akron and Macon. The shifting of camshafts, magneto and distributor fingers enabled this "B" engine to reverse from full power in one direction to full power in the opposite direction in eight seconds. Model tests were completed in 1935, and the Navy ordered 22 engines, but only two were shipped before the dirigibles were tragically lost and the Navy abandoned lighter-than-air development.

Meanwhile, the Army had ordered a redesign of the first V-1710, and this model, known as the V-1710-C, weighed 1,122 pounds, 100 more than the original "A." A satisfactory 50-hour development test was completed at a rating of 800 hp at 2,400 rpm. In 1935, an improved model, on a second development test, was rated at 1,000 hp at 2,650 rpm.

The attainment of 1,000 hp was hitting close to the Allison goal. A few items required redesign, but general performance warranted the undertaking of the 150-hour type approval test necessary for final Army acceptance of the Allison V-1710 as an approved power plant for new-design military planes.

At the Allison plant, the engine was run at increasingly higher horsepower outputs in preparation for the test — and things went "haywire." Vibrations twisted off the crankshaft, crankcases cracked, valves burned out, bearings failed.

Gilman decided on a major redesign of the engine. Ronald M Hazen, who had been working on Allison engineering problems for four years at the General Motors Research Laboratories in Detroit, was given the difficult assignment. Hazen had designed the Ranger aircraft engine before coming to Allison, and was Gilman's choice as chief engineer.

In exactly three months and three days, every part of the engine except the connecting rod was redesigned under Hazen's direction, and the model was built and delivered to the Army June 13, 1936. Shortly afterward the grueling 150-hour test began at Wright Field.

With each succeeding report from the Field, confidence rose at the Allison laboratory. When the news came that the engine had run sweet and true for 140½ hours, Gilman was sure that 9½ more hours would see the realization of his dream. He decided that the time had come to retire, and he dictated his resignation against the following day when the triumph of the Allison engine. would be final. But he had to tear up his resignation the next morning.

There was still work to be done. The engine, started again, developed a disqualifying crack in the head of one bank of cylinders. With a new block of the same design installed, the engine operated on a penalty run for 245 hours when the cylinder head in the other bank failed in a similar fashion.

This was conclusive proof of a defect in one particular, but the design as a whole was pronounced fundamentally sound. The Air Corps, enthusiastic, cheered up the heartbroken engineers by ordering several engines built for experimental airplane installation.

Meanwhile, Allison engineers started to iron out the last defect in their design. They found that the cylinder head cracked for want of several ounces of aluminum — all that was necessary to strengthen the head between No 3 and No 4 cylinders and reduce deflection one-fifth. However, the interrelation of all parts of an aviation engine is close. and the new stiffened cylinder head led to vibrations of hold down studs in the manifold that made a redesign of this part necessary.

The Allison V-1710 C-8, with redesigned cylinder blocks and manifolds, passed the 150-hour type approval test with flying colors March 23, 1937, and became the first aviation engine to establish itself at a normal rating of 1,000 hp on a military test.

The engine, however, could not be put into production immediately. Two years were taken for the intermediate steps of flight-testing, detail improvement of the engine, and design of aircraft before a model of the Allison could be frozen into production.

In 1937 and 1938, with war in the air, aircraft designers were bending every ef fort to accelerate aircraft evolution. As the Army-approved power plant for single-engined pursuit planes, the Allison had to set a new epoch of design. In the past, the development of a new power plant always had evoked scores of new designs; but before new pursuit planes could be designed around the Allison engine it had to be perfected as the best of its type.

The trend was toward more horsepower and altitude, as designers in every country were drawing plans for air supremacy.

In 1937, an Allison engine was installed in the Curtiss XP-37. This plane, originally designed to take a radial engine, was changed in the nose for an improvised installation of an in-line type, and the aerodynamic features of the Allison, therefore, did not show up to full advantage. For the first time, in this experimental plane, a high-powered engine in combination with a turbosupercharger was flown.

The "D" engine, with a five-foot extended drive shaft, was developed for pusher-propulsion of the two-engined Airacuda, designed by the Bell Aircraft Corporation.

These early flight installations gave Allison engineers and the Air Corps valuable advanced data. Although the turbosupercharged Allison "C" engine combination was an important step forward, it was decided to take an intermediate step — the building in of internal first-stage supercharging in the engine.

A conversion of the sea level "C" engine to an altitude rated model, known as the V-1710-C13, was installed in the Curtiss XP-40 in the fall of 1938. During flight tests, internal first-stage super-charger improvements were made. A combination was found that was very effective. The XP-40 won the Air Corps pursuit competition in the spring of 1939. This event was open to all aircraft developers, and the Curtiss XP-40 was the first liquid-cooled entry in years. Significantly, it tripled the annual speed increase of 10 to 15 mph, set by a succession of previous air-cooled winners.

The Allison V-1710-C15, an altitude rated 1,090-hp engine, was the production model when the Army in May, 1939, issued its first quantity order. It was the 14th model of 20 engines, custom-built by hand, part by part, during the eight-year development period.

One month before receiving the Army quantity order, two months before the declaration of war in Europe, General Motors had broken ground for 390,000 square feet of factory, office and testing space. The concept of the Allison production schedule, with each succeeding Axis aggression in Europe, was revised upward by the Army until now there are several million square feet of plant space devoted to Allison engine production. The new Allison plant was planned without windows and with other features which later came to be called "blackout."

The problem of putting a custom-built precision engine into mass production presented many challenges: The Allison contains roughly 7,000 pieces; rough castings and blanks were to be analyzed before acceptance, inspected individually after each machining process, and after test of the engine (one-fifth of all Allison employees are classified as inspectors); all highly-stressed steel parts (like crankshafts and rods) and all rotating and reciprocating parts were to be magnafluxed both before and after machining; the engine called for 62 different specifications of metal, all of which would soon be high on the OPM list of critical metals; many parts required precision machining of the highest standard known — involute gear profiles and some splines were to be machined to .0001 inch tolerance, crankshaft parts and main journals to .0005 inch.

F C Kroeger, general manager of the Delco-Remy division of General Motors, became general manager of the Allison plant, and tooled it to build aircraft engines on something not much like an automobile assembly line, yet with the same principle of "flow" toward completion. Training personnel was the job of O T Kreusser, who had succeeded Gilman as head of the Allison organization. His "faculty" was the nucleus of 400 or more oldtime Allison master mechanics who had custom-built the 20 different engines of the eight-year development period.

The gravest Allison personnel problem — and other manufacturers have suffered similarly — arises from the excellence of the plant training and the defense migratory worker situation. Many of the employees have lived in Indianapolis "on the hook." As soon as they heard of a plant opening nearer their home towns, they had been prone to leave. At an 11,000-worker level, Allison had a turnover of 170 men a month.

As Allison tooled up and trained personnel, management and workers developed new manufacturing techniques. The hand micrometer, long regarded as the symbol of precision work, was not sensitive enough to check the closest limits to which parts had to be machined. New automatic gauges had to be devised. Available materials were adapted wherever possible, but recent metallurgical research was heavily drawn upon for new methods of processing and heat treatment. Shot blasting was adopted to toughen the skin of stressed steel parts. Nitriding, it was discovered, gave greater strength to the crankshaft, which with the same weight had to turn over 10 times the power of a fine auto crankshaft.

Subcontracting was a natural automobile solution to the problem of obtaining a flow of raw, semi-finished and finished parts. GM managers knew, from past experience, the manufacturing facilities of hundreds of potential suppliers. Other divisions of General Motors took assignments from Allison.

The Cadillac and Delco-Remy divisions made the major contribution. Cadillac makes 250 Allison parts, including crankshafts, connecting rods and gear-reduction assemblies. Delco-Remy furnishes aluminum and magnesium castings, in addition to 75 different machined parts. For Allison production, Delco-Remy has added two specially constructed plants.

Other GM divisions subcontracting Allison parts are Chevrolet, New Departure, Hyatt Bearing, Delco Products, Packard Electric, AC Spark Plug, Antioch Foundry, Harrison and Inland. Outside the corporation, 93 suppliers of raw materials, semi-finished and finished parts were enrolled.

Today only 20 per cent of parts (the most intricate) have to be machined at the Allison plant. The remaining 80 per cent flow into the plant from 60 different cities, scattered from Iowa to Connecticut. By May, 1940, the new Allison plant began turning out engines on a production basis. During the latter part of 1939, the laboratory, in three months, custom built almost twice as many engines as had been built during the entire eight-year development period. Production, during 1940 and 1941, was stepped up as the Allison was specified as power for new Army pursuit planes. The goal of peak production, scheduled just 12 months before, was reached in December, 1941.

The attainment would have been impossible without the closest cooperation between the aircraft and automobile industries and the Air Forces. The mass production ideas of the automobile industry were reconciled to the precision requirements of the aircraft industry, applying the American mechanical principle of simplification, which made the sewing machine and the automobile mass products, to the intricate mechanism of the Allison engine.

What makes the Allison engine as American as ham and eggs is one important point of difference with comparable European liquid-cooled engines. The 7,000 parts of the Allison engine comprise only 700 "piece parts" — or separate production problems — as against 2,300 "piece parts" in the most widely-known European rival engine. For example, piece parts in one small sub-assembly were reduced from 38 to 3 simply by casting the part whole rather than bolting it together. Incidentally, this resulted in a part that could be machined with greater accuracy for durability.

The implications of Allison's record 700 piece parts are important — the Allison is easier to maintain than a European engine with three times as many parts. Reduction of overhaul periods means more fighting hours for a squadron of fighting ships, more economy in maintenance. The number of actual fighting hours that can be wrung out of a squadron may well overcome a numerical disadvantage.

Putting the Allison engine into production did not mean that it was frozen there. Power output of the engine was first increased from 1,090 to 1,150 hp early in 1940. Usually model change means complete shutdown of production for necessary retooling, but Allison managed to maintain at least 60 per cent of its productive output during the switch-over.

The "C" engine powered the Curtiss P-40, and the "E" engine was developed for the Bell Airacobra. Design called for development of a 10-foot extension drive shaft. Engineers in Europe had tried and dropped the idea, and Allison engineers approached the problem with trepidation. Torsional analysis gave the extension shaft good characteristics through the range normally checked, but at low speed terrific vibrations developed that broke down a flexible shaft at the rear of the engine. The solution, which might have occurred to any automobile engineer, was a simple hydraulic damper application — in effect, a shock absorber that cushioned rebound in any direction.

Another Allison feature is the symmetrical crankshaft, which makes right- or left-hand rotation optional on current models. This gives the twin-engined Lockheed Lightning, in the American version , counter-rotating propellers. Equipped with turbosuperchargers, the Lightning, with its present total of 2,300 hp, climbs rapidly to high altitude.

The "F" model recently was designed to give Allison higher horsepower output. It is 30 pounds lighter, with 10 per cent less frontal area, and is 10 inches shorter than the "C" model, due to a redesigned reduction gear. In June, 1941, the "F" passed a successful 150-hour type test at a 1,325 hp altitude rating. This represents an increase in power of 175 hp, the equivalent to the power of two small motor cars, attained without major redesign of the engine.

Thus, since the 1,090-hp engine passed the 150-hour type approval test Allison engineers have increased power output by .more than 20 per cent and they confidently expect this process to continue.

The "F" engine now is furnished in several models, right- or left-hand rotation, sea level or altitude supercharging, and various power ratings from 1,150 to 1,325 hp Production several months ago was changed over from one model of the "C" type to four models of the "E" and "F." The first month of the changeover, "E" and "F" production fell only one-third below peak "C" production, and the second month exceeded that level. Production since has risen steadily.

That, considerably telescoped, is the chronology of Allison engine development and its adoption by the Army for pursuit-plane power. Obviously, those pursuit planes are not identical, or even similar. What are some of the matters affecting decisions as to what type of engine — what type of Allison engine — to use in fighter aircraft?

First, there is the point that by this time should be obvious to everyone, but is still too frequently ignored in loosely-phrased comparisons of fighter types: fighter aircraft have various missions to perform, some requiring planes that can fight well at high altitudes, some at medium altitudes and others close to the ground.

It is doubtful that designers ever will be able to turn out a completely satisfactory all-purpose fighter. Certainly at the present stage of aeronautical development it is impossible to design one plane that will have the maximum of all requirements for superiority in every mission. Although it is quite possible to supply an engine that is effective at all altitudes, the size, weight and complications necessary to accomplish this offset the advantages in horsepower gain when the aircraft operates in the lower altitudes.

Global warfare involves operations in a wide variety of climatic and tactical conditions, and the United States must have a complete line of planes — low altitude fighters for attack, high altitude for defense and escort. And they must fly in heat or cold, dust or damp.

The basic engines can be the same in all these planes, but the complete power plants will differ because it is necessary to vary such auxiliaries as supercharging to make the power plant most effective at the altitude for which the plane was designed. The engine is the power-package delivered to the plane designer. He can use it to fly high or low, to carry more weight, to gain speed, increase the rate of climb, or do other things. But the designer has to make a choice, to reach a compromise. Using one advantage means sacrificing another — even an airplane designer cannot "eat his cake and have it, too."

Military planes are highly specialized aircraft, and the specifications for their design are based on the kind of work they are expected to do. This sets up marked differences, not only the obvious ones such as between bombers and fighters, but as between models within the same general category. For instance, Britain produces heavy bombers operating at night at low or medium altitudes for large-scale area-bombing, while the United States builds high-altitude heavy bombers for daylight precision bombing of smaller but specific objectives. Similarly, there are many different kinds of fighters — more or less heavily armed, with more or less armor, more or less cruising range, with optimum performance in speed, climb and maneuverability at low levels or high altitudes.

Planes of many different types and having varying aerodynamic characteristics may be built around a given basic aircraft engine and will have widely varying performance characteristics. So it becomes obvious that the engine alone does not determine the performance characteristics of any given plane.

Basic aircraft engines are evaluated largely on the following factors: (1) horsepower/weight ratio — the weight of the engine in relation to its horsepower capacity; (2) horsepower per cubic inch of displacement — a secondary measure of the power/weight ratio, but commonly discussed in evaluating engine design since the size of an engine, or the dimensions within which it may be installed, is important in the plane design; (3) frontal area — the area within which an engine may be streamlined in its installation (affects wind resistance or drag; the effect of engine auxiliaries on drag must also be included in this evaluation); (4) fuel consumption — measured in pounds of fuel consumed per horsepower per hour, and varying from 0.4 pounds to 1.2 pounds per horsepower per hour. F

ollowing this evaluation of the basic engine comes a similar consideration of auxiliary equipment, the principal auxiliary being the supercharger.

It takes about 14 pounds of air mixed with a pound of gasoline for so-called complete combustion of aircraft engine fuel, and the amount of power which can be obtained is directly proportional to the amount of air it is possible to get into the cylinders. The conventional unsupercharged engine depends upon the atmospheric pressure to push the air into the engine, but air pressure decreases as the plane climbs higher, and the pressure available to push the air into the engine drops correspondingly. Furthermore, as the atmospheric pressure drops, so does the weight of air in a given volume.

With the effect of temperature change and back pressure the engine could deliver more power to the crankshaft at increasing altitude if the same weight of air could be gotten into the cylinders, mixed with the proper proportion of fuel. To do this, a supercharger is used. This is simply a centrifugal fan type of air pump running at very high speed, which must take enough air at the lower density of the higher altitude and force it into the engine at the same rate (measured in weight) as the engine receives air at sea level. Aircraft power plant designers in this country can make available many variations in methods of supercharging, each with its particular advantages. Of these, eight stand out as having important differences:

  1. sea level;
  2. single speed, single stage;
  3. two speed-single stage, mechanical clutch;
  4. variable speed, single stage, hydraulic clutch;
  5. two stage with mechanical clutch;
  6. two stage with hydraulic clutch;
  7. two stage with hydraulic clutch and intercooling, and
  8. turbosupercharger with intercooling.
    (For a more complete discussion of supercharging, see "Engines and Altitude," by Rex Sydney, in Flying, March 1943 —ED)

Knowledge of these various types of supercharging make it obvious how many different possibilities there are in the same basic engine. The sea-level engine, useful in a ground attack plane with no altitude performance, when fitted with the turbosupercharger and intercoolers becomes a first-rate high altitude engine. With careful attention to interchangeability, the engine for every one of these combinations can be made to come off the same production line with only a few modifications at the rear of the engine.

Variations in horsepower-altitude performance are obtainable, then, by various methods of supercharging — variations which are partly engine modifications and partly items entirely separate from the engine. Moreover, the weight, size and characteristics of the plane actually affect the overall airplane performance to at least as great a degree as power plant modifications. And in every case, while more supercharging results in better high altitude performance by the engine it also results in increasing complication, weight and frequently drag.

Dive bombers, torpedo bombers and aircraft specifically designed for cooperation with ground troops do not require engines supercharged for operation at high altitudes and are more efficient without such equipment.

High-altitude bombers, bomber escorts and interceptors are types of ships requiring maximum performance at high altitudes. High altitude supercharging must be provided regardless of the detrimental effect on low altitude performance.

Intercooling or aftercooling — lowering the temperature of the air-fuel mixture — introduces a new set of problems in power plant and plane design almost equal to those created by the addition of supercharging itself. Then why any cooling?

Compression raises the temperature of air. Supercharging is compression of air, and a higher degree of compression is necessary as air density thins out with rise in altitude, but — the more compression the higher the temperature of the compressed air. When a given grade of fuel is used, the higher the temperature the nearer the approach of either pre-ignition or detonation.

There is a point in supercharging where this danger of pre-ignition or detonation, or both, occurs. When the requirements of supercharging to get high horsepower at high altitude reach this point it becomes economical to introduce intercooling despite the added weight and size of equipment and the drag of the added cooling radiators.

Looking farther into this matter of comparing fighter planes, another stumbling block attracts attention: failure to understand that critical altitude and service ceiling are two distinctly different things. Critical altitude of a power plant, determined by the engine and its supercharging equipment, is the altitude at which maximum power output of the engine begins to fall off directly with altitude. Critical altitude of an airplane is the altitude at which maximum speed in level flight is obtained. With a given basic engine the designers can, within certain limits, fix this critical altitude wherever they think it ought to be, all factors considered. Moreover, although the critical altitude of the plane is the important factor, it is not necessarily the same as the critical altitude of the power plant, and may be appreciably higher.

The service ceiling of an airplane, which varies with the weight of the fully equipped plane and its wing loading, as well as power, is the highest altitude at which it is practical to fly. Specifically, it usually is defined at the maximum altitude at which the plane can still maintain a rate of climb of 100 feet a minute.

There is a wide difference between the critical altitude and the service ceiling of any given plane, and the measure of superiority is neither speed nor altitude alone, but the performance at any altitude compared with the enemy's performance at the same altitude.

In summation, the limitations upon power plant design, particularly as to the installation of supercharging and intercooling, created by the requirements of plane design, are of high importance. And they are imposed, not so much by the abilities of the engine as by the tactical needs, the combat mission. All these considerations — and of course many others — influenced the design of the American fighter planes using the Allison engine — Tomahawk, Kittyhawk, Airacobra, Mustang, Lightning. And just as their successes are not due solely to the Allison engine, neither are their deficiencies.

The principal deficiency of the Tomahawk, for instance, was inadequate firepower. The result was that comparatively few were produced, and the Kittyhawk came along quickly — basically the same, but with enough firepower to make it one of the deadliest fighters in the world. It is, to be sure, a medium altitude plane — and so is the Warhawk, latest of the P-40s, with a Ro1ls-Royce Merlin engine in it. It is interesting to note in passing that the P-40, once most criticized of American fighters, is the only one which appears to have performed equally well in all combat theaters, under all conditions of climate. Such versatility, of course, could be obtained only at the sacrifice of various specialties which might have given it brilliant distinction in restricted zones. Compare the P-40's record with that of the Spitfire, for instance. The latter, incontestably the greatest of all fighters over Britain, was just a mediocre airplane in Burma, and definitely inferior to the P-40 in the peculiar conditions of the air fighting over Egypt and Libya.

The Lightning has been from its inception a high-altitude fighter, an interceptor of tremendous speed and extraordinary range as current production-model fighters go. It was a long time coming into quantity output, but the difficulties were not in its twin Allison engines. Now, in increasing numbers in Africa and the Pacific, it is proving itself one of the finest fighters in use.

The latest model of the Mustang also marks a shift from Allison to Rolls-Royce, and in this case the shift is accompanied by high-altitude performance. This North American plane was designed originally, however, as a ground support and attack fighter, and its Allison engine was part of the design which put it in the 400 mph class.

The Airacobra, with its Allison engine placed behind the pilot and at the center of gravity of the plane, is one of the most maneuverable of fighters. It is not as fast as the Mustang or the Lightning — but few planes are.

The matter of supercharging Allison engines has been widely discussed in public, with a multiplicity of conclusions. Well, they have been supercharged one way or another right along, but always with the idea that they could use any supercharger, depending only on the operational altitude desired.

Power output also has been a subject of considerable yammering in print, with the Allison generally low-rated as the least of the in-line engines. That simply is not true. The military rating of the Allison V-1710-C15 (later Allison model ratings are available, but not those of all the others) was 1,090 hp. Military rating of other liquid-cooled engines of the corresponding model was: 1,000 hp for the Daimler Benz DB-601-A, 1,025 hp for the Rolls-Royce Merlin X, 1,100 for the Hispano-Suiza 12Y-51, and 975 hp for the Junkers Jumo 211.

As to the "built-in headwind" argument, the liquid-cooled protagonists pointed with glee to the fact that the Navy, with its Corsair, and the Army, with its Republic Thunderbolt, both powered with 2,000-hp radial air-cooled engines, had passed 400 mph with some to spare, it's true, but had been forced to add about 35 per cent more power to gain only about 10 per cent in speed. Superficially it looks like the liquid-cooled advocates had a point. Actually, of course, so many more factors than a snub-nose entered into the designs that, while the point might be indicative, it certainly was not conclusive. There are designers who insist that, with increasing size of planes, the streamlining advantage of in-line engines disappears at speeds above 400 mph. But at higher speeds, a lot of other things begin to happen, too, such as the unhappy tendency of air to "shatter" instead of flowing past the streamlined form. Engineers are still seeking the answers to flight problems at speeds above 450 mph, and the answer involves a lot more than settling the argument between air-cooled and liquid-cooled engine enthusiasts.

The Navy has stuck with the radial for many reasons, most of them having nothing to do with streamlining for additional speed. Stubnose ships have more lifting power, get off the carriers with shorter runs and land more slowly than do extremely streamlined types. The elaborate plumbing system of most liquid-cooled power plants is vulnerable in combat, entailing greater repairs — and a carrier must go to sea cluttered with the least possible amount of space-using machine shop.

There has seemed to be a tendency in some parts of the Army to design new planes around air-cooled radials rather than liquid-cooled in-line engines, possibly because the new planes are heavier and Allison and other manufacturers of liquid-cooled engines had not yet announced development of engines of 2,000 hp or better, as the manufacturers of air-cooled engines have done.

But Allison in a few short years of experiment and manufacture has upped the rating of its 12-cylinder V-1710 from 1,090 to 1,325 (it actually puts out quite a little more, and has an edge over the Merlin), and it can be flatly stated that the ultimate in liquid-cooled engines has not been achieved. As in the case of the British Napier Sabre, so in the case of Allison's proposed 2,000-hp engine, the scheme has not quite proved itself. The Napier Sabre is ahead — the new Hawker Typhoon is powered with it, and credited with a speed in excess of 400 mph. The Typhoon first saw action over the Dieppe commando raid last fall, but it was mid-January before its was announced publicly, and one of the reasons for holding up the news was that all the "bugs" had not been worked out — had not been, in fact, when the announcement finally was made.

Meantime, Allison engines have performed superbly in various types of planes on every front of the war, and will go right on powering fighter aircraft until the war is won — even if there were no further improvement at all.

This article was originally published in the May, 1943, issue of Flying magazine, vol 32, no 5, pp 28-31, 84, 86, 89-90, 92.
The original article includes photos of the engine, P-40, P-38, P-51, P-39 and engines on the assembly line.
The P-40 photo is credited to Rudy Arnold; the other photos are not credited.