The P-38 Lightning
Evolution of Speed and Climb Performance

By Wayne Cunningham


In June of 1937 the Lockheed Aircraft Corporation of Burbank, California began work on a new fighter aircraft. In-house they called it the Model 22; the United States Army Air Corps called it the XP-38.

The specifications for the new fighter included the following:

1.A speed of at least 360 mph at 20,000 feet. (The desired maximum speed was 400 mph).
2.Climb to 20,000 feet in six minutes or less.
3.Fuel for one hour at operating speed.
4.Armament was to be designed around a Rapid-fire cannon, as destruction of enemy bombers was considered the primary function.

To encourage the builders to present new ideas and innovative solutions the requirements were intentionally obscure and beyond the reach of what was considered possible.

Although there was considerable resistance to a twin-engine fighter concept, it was considered necessary because of the power plants available at the time. As speed was the primary issue a liquid-cooled engine appeared appropriate, and the only available American liquid-cooled engine in the 1000 hp. class was the General-Motors Allison V-1710. Lt. Benjamin Kelsey- head of the Fighter Projects Office at Wright-Field- wanted a 1500 horsepower single-engine fighter, but an American engine of this type was not even on the drawing board in 1936. The V-1710 was rated at 1150 horsepower at military power, and it was reasoned that it would probably take two of them, mated with General-Electric turbosuperchargers, to reach 400 mph at 20,000 feet.

Lockheed’s best designer was put in charge of the fighter project. Clarence “Kelly” Johnson would go on to become one of the world’s great airplane designers, and his P-38 was the first of several “fast” aircraft, which would include the Mach-three SR-71 Blackbird. Johnson was determined to exceed the requirements of the specification; thus his concept for the XP-38 was very advanced for 1937, and included the following considerations.

The primary disadvantage of any large fighter is lack of agility and maneuverability, and a twin-engine aircraft is, of course, more complex than the single-engine type and takes longer for both pilots and ground crew to master. It also cost more and takes longer to build. One of the most important issues from a performance standpoint, however, was drag.

By dividing thrust horsepower (bhp x propeller efficiency) by parasite area (parasite drag coefficient x wing area), we can estimate the relationship between thrust and drag for any airplane. Aircraft with similar values will have similar maximum speeds.

Maximum speed is ultimately dependent on thrust and drag. Two engines essentially double the thrust, but the added drag can be a major problem. Compared to a single engine aircraft the P-38 had essentially three fuselages- two booms and one cockpit "gondola". It also had a complex assortment of scoops and vents for the radiators and oil coolers, and two induction scoops. As the airplane also had a 52-foot wingspan there was potential for a considerable amount of drag. The above chart, however, shows the P-38 well up in its class.

Of course, a tremendous effort was necessary to produce the cleanest, smoothest airplane possible. The most unusual feature in this respect was the design of the intercoolers. Engines with two stages of compression (two supercharger impellers) almost always require an intercooler to reduce induction air temperature that could lead to detonation. Intercoolers usually require an air-scoop and are sure to cause additional drag. Johnson used an unusual type, which utilized the space in the leading edge of the wings. Hot air from the turbosuperchargers was forced through spanwise passages in the wings and cooled by a small amount of outside air, which circulated through the center of the unit. This device produced very little drag, but would turn out to be a major problem in the development of the aircraft… more about this later.

The XP-38 did not live long enough to find its way into the full-scale Langley wind tunnel, but some performance data survived and it indicates a basic, zero-lift drag coefficient of just over .020, which would be good for even a single engine airplane. The mass produced, combat-ready models would generate more drag.

Johnson saw the considerable increase in power as the primary advantage of a twin-engine fighter, thus overcoming his major obstacle. The XP-38 also had a number of advanced features. For example, the propellers turned in opposite directions, thus canceling engine torque forces. The aircraft was designed with a tricycle landing gear, which made takeoff and landing much easier and safer than the “conventional” gear arrangement. The gun installation was also very efficient, with all guns mounted in the nose. All skin sections were butt-joined with flush riveting throughout, and all flight controls were metal covered. Add to this an unusually long range for a fighter aircraft, which would become one of its most important assets.

The XP-38 made its first flight on 27January 1939, with Lt. Ben Kelsey at the controls. Only five more test flights were carried out in the local area around March Field, California, and then the story takes an unusual twist.

The depression had reduced funding for military aircraft, especially fighters, to almost nothing. Nevertheless, by early 1939 America was beginning to become concerned about the re-arming of Nazi Germany and the probability of war in Europe and Asia. Consequently, General Henry “Hap” Arnold, Commanding General of the Army Air Corps, was beginning to feel high level pressure concerning the poor state of America’s defensive fighter force. The German Messerschmitt Bf-109 and the British Spitfire both had a top speed of around 350-360 mph with a ceiling of well over 30,000 feet. The US would soon put the P-40 into production, which was relatively fast below 15,000 feet but had poor climb and altitude performance. What the General wanted was a 400mph fighter…something to put America back in the game. The XP-38 design looked promising, but the next hurtle was to get Congress to fund the project. What he needed was something to get their attention.

In 1939, Howard Hughes and his H-1 racer held the coast-to-coast speed record. At 7 hours and 28 minutes his average speed from Burbank to Newark was 327mph. Lt. Kelsey was sure the XP-38 could easily set a new record. His plan was to “deliver” the airplane from Southern California to Wright Field in Dayton, Ohio, and If all went according to plan he would continue on to Mitchell Field on Long Island, New York. General Arnold approved the plan and all did go well until on final approach to Mitchell Field. The XP-38 was the only model of the type equipped with float-type carburetors. All other P-38s used pressure carburetors of one type or another. Even though the aircraft had turbosuperchargers (which compressed and heated the induction air) it was apparently possible, under certain conditions of low power, to develop carburetor ice, and that was shown as the official cause of the crash.

After putting gear and flaps down Kelsey pushed up the throttles to add power… and got nothing, the engines remained at idle. As it turned out he was directly over a golf course at the time and made a desperate attempt to land the airplane there. Lt. Kelsey walked away from the crash, but the XP-38 was a write-off.

The XP-38 was destroyed, but the project was very much alive. General Arnold now had the data he was looking for…a very advanced 400-mph American “interceptor”. Yes, they even used the word interceptor to imply a defensive aircraft, something to help defend America from foreign invaders. Whatever it took to satisfy the strong isolationist Congress in charge at the time…and it worked. A contract for thirteen YP-38 aircraft soon followed.

What was the actual maximum speed of the XP-38? To begin, it is almost certain that no attempt was made to operate the aircraft at maximum speed. The first flight almost ended in disaster when the flap operating rods broke on takeoff. There were only five other test flights used to correct the flap problem and other known defects, after which the aircraft was sent on its cross-country flight. Kelsey was a trained test pilot and the flight plan required strict adherence to specific power settings calculated to balance speed with fuel consumption. In fact, he makes a similar statement in an interview (see below).

Warren Bodie’s definitive book on the P-38 has most of the answers concerning this question. The best data comes from an interview that took place several years after Kelsey’s retirement. It appears that he took two sets of performance readings during the flight while at high-speed cruise. He used this to calculate what the maximum speed would be at 20,000 feet at rated power. Remember that Kelsey was an MIT engineering graduate who spent much of his time estimating the performance of new airplanes.

One of the calculations showed a maximum speed of 394 mph at 20,000 feet on 1150 hp/ engine. The other showed 399 mph at 20,000 feet if 1250 hp/ engine was used. His data and calculations were given to Kelly Johnson who came up with 403 mph at critical altitude (around 20,000 feet) on 1150 hp/ engine. Johnson also had plans to alter the design of the airplane and expected to improve the speed by around 10-mph, giving it a top speed of 413 mph. This is often quoted as the top speed of the XP-38, but as flown, it would appear to be between 394 to 403 mph. Nonetheless, this makes the P-38 the first 400-mph fighter in history.

Maximum Speed and Climb

There were approximately ten-thousand P-38s built during World War II, and like all fighters of that era, it went through numerous modifications and model changes.

   ModelEngineMax. Takeoff
TurboInternal FuelGross Wt.
   XP-38   V-1710-10/111150B-241013700
   YP-38   V-1710-27/291150B-241013500
   P-322   V-1710-151090B-2N/A14500
   P-38LO   V-1710-27/291150B-230014200
   P-38D   V-1710-27/291150B-230014500
   P-38E   V-1710-27/291150B-230014400
   P-38F   V-1710-49/531325B-230015900
   P-38G   V-1710-51/551325B-1330015800
   P-38H   V-1710-89/911425B-13/3330016300
   P-38J   V-1710-89/911425B-3341017500
   P-38L   V-1710-111/1131425B-3341017500
   P-38M   V-1710-111/1131425B-3341018000

The maximum speed and climb performance of the various P-38 models is, however, something of a mystery. Although rated takeoff power from the XP-38 through the P-38H increased from 1150 bhp to 1425 bhp the Tactical Planning Charts issued by Wright Field show all variants with a similar speed. Closer examination of the charts will reveal that all P-38 powerplants, through the H model, are limited to 1150-1240 bhp, due to "inadequate cooling."

To be more specific, the foremost problem was the temperature of the compressed air from the turbosupercharger entering the carburetor. High carburetor air temperature (CAT for short) can cause all kinds of engine problems including detonation, which can lead to catastrophic engine failure. Allison recommended a CAT of no more than 45 degrees C.

As it turned out high CAT was one of the major problems limiting P-38 performance through the P-38H. The root cause was, of course, the limited cooling ability of the wing leading edge intercoolers found in all early P-38s. They were a very clever design, inducing almost no aerodynamic drag, but they were designed for the 1000 hp Allisons of the late 1930s. By 1943 Military power was up to 1425 bhp and War Emergency Power was 1600 bhp. The increased power required higher induction pressure, which through compression by the superchargers heated the air by several hundred degrees. There is no way that the simple intercoolers could keep CAT below 45 degrees C. when operating at high power at altitude.

On the other hand, Ben Kelsey (now a Colonel) and Colonel Cass Hough, of the Eighth Air Force Technical Section, had different ideas about CAT. They were of the opinion (as were many of the fighter pilots) that American fighter engines were still being used at conservative, peacetime power settings, which were inadequate for combat. With this in mind they set out to determine just how much abuse a P-38 engine could take, and what they found surprised everyone.

The first P-38s to arrive in England were rated at 42" up to about 20,000 feet, 40" at 25,000 feet, with further reductions above that. Colonel Hough decided the best way to find out how much power was actually available without blowing up the engines was to remove the throttle stops and find out for himself. This made full throttle available at any altitude. Operating like this, Hough spent two weeks "abusing the engines", searching for their maximum limit. "We found that below 25,000 feet we could pull up to 60" of manifold pressure without material harm, and we could run as high as 40" at 40,000 feet (60" would yield around 1600 + bhp/ engine). He did warn that this kind of abuse should be of short duration. Col. Kelsey was busy doing the same thing at the Lockheed plant in California.

In a February 1943, P-38 Progress Report, Kelsey described how he had been "beating engines unmercifully". The F-10 engines in the P-38G had been run at 51" (1440 bhp) or more for periods of 7 and 8 minutes. "A series of climbs have been made at this power from takeoff to 22,000 feet…" "From our best previous estimates of limiting carburetor air temperature to 45 degrees, 51" could not be pulled above 15,000 feet." "Actually, 70 degrees C. has been run satisfactorily". "We have not yet established actual limits".

In March Kelsey reports: "I finally succeeded in reaching limiting carburetor air temperature at altitude. I got excessive roughness, cutting out, and backfires at 190 and 200 degrees F [88 and 93 degrees C]. at about 25,000 feet"… one intercooler was actually blown up". "We very evidently have much larger tolerances in temperature, back pressure and carburetor air pressure than we anticipate".

Kelsey and Hough were looking for a compromise…they wanted the most power available without engine damage. Kelsey recommended a combat rating of 47" at 3000 rpm (1325 bhp) to 20,000 feet. He also recommended a 5-minute limit at 50 degrees CAT. Eighth Fighter Command was more conservative; they eventually established a War Emergency Power rating of 45" up to 25,000 feet. Wright Field, with more responsibility, was even more conservative and recommended a Military Power of only 41” (1150 bhp). War Emergency Power was not recommended. This was essentially the same power available to the XP-38 in 1939!

In the end, the various U.S. Air Forces set their limits somewhere between Kelsey and Wright Field. Actually, it was the fighter pilots and their crew chiefs that often had the last word on how the powerplant controls were rigged, and it was not uncommon practice to remove the throttle stops on operational P-38s. This provided full throttle (60-70" of manifold pressure) at lower altitudes, but it greatly increased the chance of blowing the engine. Many pilots thought that since they were the ones risking their lives on the cutting edge, it was only fair that they should decide how to use the power. The author has a friend who flew the P-38H, J and L models with the 55th Fighter Group, Eighth Air Force. He told me that he had the throttle stops taken off all three aircraft and, when necessary, used full throttle in combat. He had no problems with the H model overheating.

The P-38H entered service in mid 1943, and it was definitely a transition aircraft. With the wing leading edge intercoolers it looked almost identical to the F and G models, but it had V-1710-89/91 engines with much higher power ratings. It also had improvements to the various cooling systems. The Flight Handbook shows Takeoff and Military Power as 1425 bhp, and War Emergency as 1600 bhp. The Flight Handbook also shows maximum carburetor air temperature as 45 degrees C., creating an immediate conflict because of intercooler limitations. Wright Field rated the H model at only 1240 bhp due to "inadequate cooling." Nevertheless, many of the more aggressive pilots disregarded all of this and used the full capability of the new powerplants when it appeared necessary.

The P-38J arrived soon after the H model and became prominent by early 1944. Featuring the new core-type intercoolers and re-designed oil and coolant radiator ducts, it was the first P-38 capable of using full rated power without fear of exceeding 45 degrees CAT. WEP was now 1600 bhp up to maximum turbo rpm, which usually occurred at around 25,000 feet. High CAT was no longer a problem. Unfortunately, this fix came at a price. The new intercoolers produced considerable drag, and at the same power the P-38J was 10-12 mph slower than previous models. Nonetheless, the J had a lot more useable power, especially at altitude, and had better overall performance, it also carried 100 US gallons more internal fuel than the F and G models, using the space once occupied by the original intercoolers.

The following curves compare the speed and climb performance of the P-38G, using the conservative Wright-Field data as a baseline (red), and that generated by Kelsey and Hough (orange) as the maximum limit (notice that the cleaner G model is faster than the J when at the same power). The power settings actually used in combat fell somewhere in between. Notice how the high CAT-induced power reductions above 15,000 feet affected P-38G performance. In fact, any high power setting with the carburetor air temperature above 45-50 degrees C was, supposedly, risky. Also shown are curves for a P-38J at WEP using both 130 and 150-grade fuels. 150-grade fuel was used by the US Eighth Air Force from mid 1944 to the end of the war. The P-38J curves are modified Wright Field data verified by power required calculations.

Merlin Powered Lightning

This of course is pure speculation and never happened, but what effect would it have had on P-38 performance? The Lightning’s powerplant installation turned out to be its major nemesis. The wing leading edge intercoolers kept power almost static until the arrival of the J model. Unfortunately the P-38J aircraft used by the Eighth Air Force brought with them an unexpected batch of engine failures. As this occurred only within the Eighth AF it has always been something of a mystery and beyond the scope of this paper. What follows concerns studies investigating the possibility of installing two-speed Merlin engines in the Lightning in an attempt at simplifying the design and improving performance.

Considering the above, this would be a good time to review the advantages and disadvantages of the Lightning’s V-1710/turbosupercharger powerplant. The primary disadvantages include the complexity and nature of the system. Add to that the considerable space needed to house all the ducting. The Lightning had approximately 12 feet of intake ducting per engine, which included a number of separate sections and joints. All of this required tight seals and proper alignment to ensure adequate intake pressure, which affected critical altitude and turbo rpm. In fact, this was an ongoing problem with the Lightning. Some airplanes tested by Wright Field showed a critical altitude much lower than advertised. After properly sealing and aligning the ducts, the airplanes performed normally. The most important problem was, however, the lack of exhaust thrust. At that time, standard turbosuperchargers could take very little back pressure, and collecting the turbine exhaust in a thrust-producing manifold was out of the question. As it turned out, exhaust thrust was very important. Approximately one-third of the heat energy produced by a piston engine is available for useful work, the rest is wasted! A well-designed jet (ejector) exhaust system can capture a considerable amount of this energy, and can easily increase rated horsepower by 10 per-cent or more. At 25,000 feet a 400-mph airplane could expect a gain of around 20-25 mph… and it cost almost nothing.

The most important advantage of the turbo was a very high critical altitude (if all the ducts were tight), and the small amount of engine horsepower required to run the turbosupercharger. Much of the energy used to run the turbo came from the tremendous heat (roughly 1700 degrees F.) of the exhaust air. This greatly decreased fuel consumption and extended the range.

In 1940 Packard Motors of Detroit began building the two-speed Merlin V-1650-1 (Merlin 28) under license from Rolls Royce. This engine had 1170 horsepower in high blower with a critical altitude of 21,000 feet. Lockheed ran a study comparing a Merlin XX powered Lightning with a standard V-1710 powered variant. The reported speed difference was over 25 mph, favoring the Merlin powered airplane. Climb performance was similar to the Allison powered machine.

Another Merlin vs. Allison comparison in 1942 involved the V-1710-89/91 Allisons (engines used in standard P-38J) and the Packard V-1650-3 two-speed, two-stage Merlin used in the P-51B/C. Utilizing Military Power speed was almost identical.

Yet another study in 1944 compared V-1710s producing 1725 bhp and "advanced" Merlins using "special" fuel and producing 2000 bhp (no altitude specified). The Merlin powered version could supposedly attain 468 mph at 30,000 feet, which was considerably better than the Allison powered version.

These studies were all conducted by Lockheed and exhibit a certain amount of optimism in regard to maximum speed for both types, but the consensus clearly shows better performance with the Merlin powered Lightning.

Taking a different tack, the author has personally calculated the performance of a V-1650-7 (P-51D) powered Lightning with surprising results. I have spent over 25 years working with performance calculation, and in my opinion the power required/ power available method works very well for speed and climb data. Of course, the main problem is an awareness of all the factors affecting the variables. The results of this study show a P-38 with almost identical speed when compared to the P-51D (around 440 mph at 24,000 feet), and slightly better climb. The key to the comparison is the value of the drag coefficient used, and the weight estimate. If the basic, zero-lift drag coefficient of the early P-38s (E-H models) is used, the V-1650-7 powered Lightning is slightly faster than the Mustang. Using P-38J drag data, the Mustang is slightly faster. In either case the P-38 climbs better.

To help substantiate the calculations, consider this. For speed comparison, the thrust horsepower divided by parasite area for either airplane is very similar (again, it depends on which drag coefficient is selected). This provides a quick estimate of thrust vs. drag, and works for any propeller driven airplane. For climb performance, the power loading for both airplanes is also similar. The actual performance of a V-1650-7 powered Lightning would depend on Lockheed’s ability to keep weight down and drag as low as practicable.

So, with the above in mind the logical question would be, why didn’t they switch to Merlins?

As it turns out there were several reasons, but one of the most important would have been the drop in production caused by the transition. At that time the US needed every first-line fighter they could lay their hands on, and a drop in P-38 production was out of the question.


The twin-engine P-38 Lightning was the result of a 1937 US Army Air Corps requirement for a 360-400 mph "interceptor". A twin-engine fighter was not in favor, but it was calculated that two engines would be necessary to meet the requirement. With its introduction in early 1939 it became the first 400-mph fighter in history, but it would take another year and a half to get it into production.

The P-38s 400-mph speed and 40,000-foot ceiling placed it in a pioneering role when it came to high altitude dives and the effect of compressibility. As it turned out, all World War II fighters had some trouble with compressibility, usually encountered in dives from high altitude. Unfortunately the Lightning’s unusual shape induced a high-speed airflow over the wing root resulting in a low critical Mach number, and caused it to enter compressibility induced control problems at a relatively low speed. This turned out to be a major problem for any dive started above 25,000 feet. In Europe, German pilots used this to their advantage against P-38s engaged in the high-altitude bomber escort role. The problem was eventually improved by installing dive-recovery flaps under the wings, but it was never totally eliminated. Below 25,000 feet the Lightning could generally out-dive the Bf-109 or the FW-190.

Lockheed could not have anticipated this problem. In fact, Kelly Johnson probably knew as much about compressibility as anyone in the world at that time. Unfortunately, in 1937 no one could predict with any degree of certainty the effect of compressibility on aircraft control.

Speed and climb performance normally depends on powerplant development, and the Lightning’s Allison V-1710s grew from 1150 bhp in 1941 to 1600 in 1943. The wing leading edge intercoolers officially prevented any noticeable gain in performance until the arrival of the J model. Nonetheless, two energetic (and apparently fearless) Colonels named Kelsey and Hough showed everyone the edges of the Lightning’s envelope, and allowed pilots to get all the performance the airplane had to offer. With the engine problems resolved, and the addition of hydraulically boosted ailerons and dive recovery flaps, the aircraft remained competitive to the end of the war.

A Brief Operational Assessment

The P-38 Lightning was the only successful twin-engine dogfighter of the war, and it served from Europe to the Pacific. It performed well in North Africa, Italy and on the continent with the Ninth Tactical Air Force, but the records show that the P-38 did not do well as a long range, high altitude bomber escort with the Eighth Air Force. The major problem would appear to be the unusually large number of engine failures that occurred during the crucial first six months of 1944. There was also the restriction requiring a low dive speed above 20,000 feet, which was the standard altitude for escort fighters. The engine problems were under control by mid 1944 and the other problems were eventually eliminated or improved.

If the Lightning did not do that well with the Eighth Air Force (for whatever reason) it more than made up for it in the Pacific. It was over this vast, rugged area fighting the Japanese that the P-38 came into its own. In this theater Lightnings shot down more enemy aircraft than any other AAF type. America’s two top aces got all their kills in the Lightning, and it was used to intercept and shoot down Admiral Yamamoto’s plane. At that time, the P-38 was the only American fighter in the theater with the range to perform this 750-mile mission.

The P-38 was definitely the AAF fighter of choice in the Pacific, primarily due to its twin-engine design and long range. One of the nagging fears of any pilot operating over the Pacific was being forced down at sea. The chance of friendly units locating you before the enemy (or the sharks) was not always good. Having two engines provided a great boost in morale, and many P-38 pilots returned to base on one engine. In fact, if the war had lasted a few months longer the British DeHavilland Hornet, a Merlin powered twin-engine dogfighter with very high performance, was also slated for the Pacific Theater.

The Pacific war was fought at a lower altitude than that in Europe, due to differences in targets and lack of heavy anti-aircraft batteries. As a result, compressibility was rarely an issue. The level speed and climb performance of the P-38 was also good, and throughout the war the Lightning proved to be generally faster that most Japanese fighters… this includes the KI-84 Hayate and N1K2-J Shiden Kai.

With the exception of the gains associated with the use of 150 grade fuel by the US Eighth Air Force, the Lightning’s performance was not improved beyond the standard P-38J/L; consequently, speed and climb remained otherwise constant from late 1943 to wars end. The Focke-Wulf 190D-9 and late model Bf-109s had better speed and climb performance than the Lightning, but with boosted ailerons and combat flaps, it was more maneuverable. It was also as good or better in a dive, if below 25,000 feet.

The P-38 Lightning turned out to be a real "work horse" for the USAAF. It served around the world as a fighter, fighter-bomber, and photographic/ reconnaissance aircraft and will always be considered one of three great USAAF fighters of World War II.


Bodie, Warren M. The Lockheed P-38 Lightning. Hiawassee, GA: Widewing Publications, 1991.

Hough, Lt. Col. Cass S. Performance of P-38 vs. German and British Fighters During the North African Campaign. Air Force Historical Research Agency. Maxwell AFB, AL. 1943

Kinzie, Burt. P-38 Lightning: In Detail and Scale. Carrollton, TX: Squadron Signal Publications, 1998.

Pace, Steve. Lockheed P-38 Lightning. Osceola, WI. Motorbooks International. 1996

Whitney, Daniel D. Vee’s For Victory. Atglen, PA: Schiffer Military History, 1998.

Erection and Maintenance Manual for the P-38H-1 through the P-38J-25, AN 01-75-2, 1944

Pilots Handbook for P-38H, J and L, AN 01-75-1, 1944

US National Archives, RG-18, Army Air Forces, 1941-1945.