Share This Article

But for the intervention of WWII, Curtiss-Wright’s big CW-20 might have displaced the Douglas DC-3 on many major air carrier routes.

When First Lieutenant Edward D. Michalek, a former B-24 co-pilot, joined the Air Force Reserve in 1949, he never dreamed that within the space of a year he’d leave his civilian occupation to become a Curtiss-Wright C-46 first pilot in an overseas combat zone. In August 1950, his unit, the 84th Squadron, 437th Troop Carrier Wing, was among the very first Air Force Reserves ordered to active duty in support of the United Nations “police action” in Korea. That November, after completing their training and preparation in the United States, Michalek and more than 2,000 other members of the 437th TCW arrived with their C-46s at Brady Air Field on the northern coast of Kyushu, Japan.

“Our outfit immediately began making scheduled runs both within Japan itself and to and from Korea,” Michalek recalled, “almost like an airline. And we carried anything that would fit into a C-46: people, military supplies, medical supplies, spare parts, munitions, gasoline—you name it.” According to him, the weather along their routes, especially around Japan during the winter months, was very unpredictable. “One of the trips I frequently drew,” Michalek said, “was a round robin from Brady to Tokyo and back. We’d fly up the Inland Sea between Honshu and Shikoku and stop at various American bases along the route, dropping off and picking up passengers and cargo. On those trips, you could almost count on flying under instrument conditions some of the way.”

He cited two factors that helped them maintain a fairly regular schedule: “First, the C-46 was a pretty good instrument flying platform, and the other thing was GCA [ground controlled approach].” Developed during World War II, GCA consisted of two precision radars, one for azimuth (course) and another for elevation (glide slope), in which a ground controller, watching the aircraft’s position with respect to the desired glide path and course, would literally “talk” the pilot down to the runway. “When our destination airfield was socked-in with a low overcast,” Michalek recalled, “GCA saved our bacon more than once. On a number of occasions, I remember dropping down through the soup at minimum ceiling, say 200 feet—less than that sometimes—and bingo! The GCA controller would have us positioned right over the runway—you had to look down to see it, not ahead. I’d immediately chop the power and drop the C-46 on the runway like a load of bricks. After one of those landings, I’d always try to track down the GCA controller who’d been on duty— typically a senior NCO—and shake the man’s hand.

“In addition to the regular routes,” Michalek said, “our unit was called upon many times to support military developments in Korea. I particularly remember the time after the Chinese Reds counterattacked, when we had to fly missions into the North to evacuate our troops and wounded. That was when our C-46s really showed their worth: They could operate out of most of the forward airfields normally used by C-47s, but carried twice their load.” Michalek continued on active duty in the Korean combat theater until June 1952, after which time the 437th TCW was deactivated and its personnel returned to the States.

The Curtiss-Wright CW-20, best known as the C-46 Commando for its military version during World War II and later, was originally conceived to compete against the Douglas DC-3 in the rapidly expanding American air transportation market of the late 1930s. Beginning with the DC-2 in 1934, followed by the DST/DC-3 in 1936, Douglas-built transports displaced virtually every other type of plane on major domestic airline routes in the United States within only a few years. The key to their overwhelming success was efficiency. Their combination of safety, reliability, comfort, speed, range and load-carrying ability enabled air carriers, for the first time, to make a profit carrying passengers alone. Airlines were now freed from dependence upon U.S. Mail contracts and the prescribed routes that went with them. If any aircraft manufacturer hoped to compete in this new market, Douglas’ DST/DC-3 was clearly the plane to beat.

Technology is seldom at a standstill, however, and Curtiss-Wright Corporation, the largest U.S. airframe and aircraft engine manufacturer at that time, was eager to secure its share of the emerging air travel market. The company wasn’t exactly a newcomer to the airliner business: In 1928, as a civil outgrowth of its Army B-2 bomber design, Curtiss (which became Curtiss-Wright in 1929) had marketed a twin-engine, 18-passenger biplane transport known as the Condor CO. The model offered no significant advantage over existing Ford and Fokker trimotor monoplanes, however, and three of the six built saw only limited service with Eastern Air Transport, a Curtiss-owned subsidiary. Curtiss-Wright experienced far better success with its completely redesigned and re-engined T-32 Condor II biplane, building 30 civil versions between 1933-1934 primarily for the transcontinental “sleeper” market. Sleepers, intended to compete with similarly equipped trains, were popular with overnight travelers, and the Condor IIs, though 50 mph slower than contemporaneous Boeing 247s and Douglas DC-2s, nevertheless offered a roomy, soundproofed fuselage that could accommodate 12 sleeper berths. But Curtiss-Wright’s success was to be comparatively short-lived: In 1936 the arrival of an enlarged development of the DC-2, the 16-berth Douglas Sleeper Transport (DST—also sold as the DC-3 21-seat day version), led to all Condor IIs being withdrawn from service the same year.

George A. Page, who had been the chief design engineer of the Condor II, initiated a series of studies in the early 1930s aimed at producing a next-generation Curtiss-Wright transport that would embody recent innovations such as a monoplane layout, all-metal structure, landing flaps and retractable landing gear. Page’s team had come up with a final design concept by the end of 1936, and the project received the company designation CW-20 (Curtiss-Wright Model 20) in April 1937. Work on a full-scale mock-up began shortly thereafter, with construction of the prototype commencing at the company’s St. Louis plant in 1938.

Although the project was technically a private venture, the design requirements for the CW-20 were established in close consultation with major airline operators, from whom Curtiss-Wright officials fully expected to obtain orders for a minimum of 25 aircraft and a potential production run of up to 50. With those criteria in mind, the final design emerged as a twin-engine, medium-range (i.e., 600-800 miles) airliner that could be configured in either 30-seat day or 16-berth sleeper versions. It also featured a novel fuselage configuration that offered pressurization in the crew and passenger compartments plus a spacious cargo bay beneath the floor of the main cabin.

One of the critical decisions made early on by Page and his staff was to use a two-engine instead of a four-engine layout, in anticipation that more powerful power plants (Wright R-2600 Duplex Cyclones) would become available by the time the prototype flew. A detailed study by Curtiss-Wright engineer A.E. Lombard concluded that a less complex, two-engine scheme would not only be more cost-effective than four engines but would also be safer. From an economic perspective of performance (speed, range and useful load) versus operating costs (acquisition price, maintenance, specific fuel consumption, aircrew requirements, etc.), the twin-engine arrangement was categorically more efficient. And for domestic airliners operating over land, Lombard’s report claimed there was no inherent safety advantage to be derived from four engines. Specifically, he argued, a two-man crew on a less complex twin-engine plane could more easily handle routine in-flight emergencies, such as engine shutdowns, than the three-man crew (two pilots and a flight engineer) required on a four-engine ship.

Two factors predominated in the CW-20’s fuselage design: passenger comfort and cargo capacity. Cabin pressurization that could be maintained at a constant 6,500 feet would offer an unprecedented level of passenger comfort by permitting high-altitude operations (for example, a 20,000-foot cruising level), above most inclement weather. However, the ideal cross section for pressurization is a perfect circle, which on a plane the size of the CW-20 would create excessive frontal area and thus result in aerodynamic drag. The solution, attributed to Page, was a unique “double-bubble” configuration—two intersecting circles of different diameters that converged at the floor of the upper main cabin. This arrangement also permitted the main wing spar to pass below the passenger cabin. The lower level of the bubble resulted in a 455-cubic-foot cargo compartment near the aircraft’s center of gravity. Further aerodynamic refinements included a cockpit/windscreen enclosure that completely blended in with the nose contour, plus a longitudinal fairing installed over the crease where the fuselage sections joined.

Another innovative characteristic of the CW-20 was its engine nacelle configuration. Termed a “tunnel cowl,” it was designed so that air was ducted in and expelled through the bottom of the cowl, which eliminated turbulent airflow and induced drag across the upper wing surface. Moreover, the length of the bottom nacelle section permitted the main landing gear to be fully enclosed by wheel well doors. Other efforts to minimize drag were a fully retracting tailwheel and butt-joining and flush riveting sections of metal skin.

CW-20 engineers also placed great emphasis on operational safety features such as optimal control placement and design, monitoring systems and gyro-stabilization devices. Key controls, levers and switches were simplified and reduced by a factor of one-third in comparison to the cockpit arrangements found on Douglas and Boeing transports of the same period. “Tell-tale” (annunciator) lights were installed on the instrument panel to forewarn pilots of irregularities in 47 different operating systems (fuel, hydraulics, electrical, engines, etc.). The control system featured a “gyropilot” that provided one-axis directional control and would boost the rudder in the event of an engine failure on takeoff. Pilots enjoyed improved visibility thanks to large side windows and a long-stroke tailwheel leg that reduced the ground angle.

The CW-20’s wing design was generally similar to the multicellular structure of the DC-2/DC-3 series and featured variation of airfoil sections combined with washout to offset wingtip stalling. Takeoff and landing characteristics were improved by the addition of slotted trailing-edge wing flaps, which in a fully lowered position were designed to produce a touchdown speed of approximately 70 mph. Another interesting feature of the wings was the tubular main fuel tanks in the outer wing panels; this location not only afforded better fire protection, but with dihedral also gave built-in gravity feed and drainage. The landing gear was exceptionally strong, the main struts having been designed with extra-length oleos that permitted landings at up to a 900-feet-perminute rate of descent, an attribute that would later permit significant increases in takeoff and landing weights. The empennage was originally configured with twin fins and rudders plus dihedral in the horizontal stabilizer. It was believed at the time that this layout would give greater pitch and yaw stability and permit wider variations in the center of gravity, but concerns over tail flutter and structural integrity led to the adoption of a single-fin arrangement shortly after the CW-20 prototype flew.

When it was completed in 1940, the CW-20 was the largest twin-engine aircraft in the world. In terms of sheer size and load, it was larger than the DC-3 by a factor of one-third, and was roughly comparable in overall performance to the four-engine Boeing 307 Stratoliner. The CW-20 first took to the air on March 26, 1940, with experienced test pilot Eddie Allen at the controls. After initial testing, the most significant change made to the prototype was the conversion to a single tail unit during mid- 1940. The prototype’s performance was listed as a 254-mph maximum speed, a 222-mph cruising speed, a 26,900-foot service ceiling, a 38,000-pound gross weight and a 1,500-mile range.

As early as the summer of 1940, the U.S. Army Air Corps had shown a strong interest in acquiring military transport versions of the CW-20, and soon thereafter awarded Curtiss-Wright a contract to produce 46 examples under the designation C-46. The military production contract called for a switch from 1,700-hp R-2600 engines to the newer Pratt & Whitney R- 2800s generating 2,000 hp. The single CW-20 prototype was delivered to the newly created U.S. Army Air Forces in June 1941 as the one and only C-55, but was returned to Curtiss-Wright after 350 hours of testing. In November 1941, Curtiss-Wright sold it to British Overseas Airways Corporation, where it remained in service until it was scrapped for lack of spare parts in late 1943.

America’s rapid movement toward a wartime footing during 1940-1941 effectively put to rest any plans to produce the CW-20 commercially. Deliveries of the first 25 C-46s to the Army Air Forces commenced in July 1942, after which production shifted to the unpressurized C-46A. Other modifications included deletion of the fairing over the fuselage seam, reduction of the number of side windows to five and the addition of double cargo doors, a hydraulic cargo winch and a reinforced floor. A total of 1,454 C-46As were subsequently produced under military contracts, 40 of which were delivered to the U.S. Marine Corps as R5C-1s. They were followed in 1944-1945 by production of 1,410 C-46Ds, which differed in having a single door to facilitate paratroop drops.

The one-off XC-46B was an experimental model that featured a stepped windscreen and nose section, similar to that of a C-47, and R-2800-34W engines with water injection. The 17 C-46Es that were built incorporated the stepped windscreen of the XC-46B as well as a single cargo door, and came with three-bladed Hamilton-Standard propellers rather than the standard Curtiss-Electric four-bladed types used on previous C-46 variants. The C-46E was followed by 234 C-46Fs, which retained the nose contour of the C-46A/D but included the other refinements of the C-46E plus squared-off wingtips. The sole C-46G had a stepped windscreen and squared wingtips and was to have followed the C-46F, but the Army Air Forces canceled production plans when the war ended.

At least two C-46 proposals were never completed: a C-46H with dual tailwheels and a C-46K meant to be powered by Wright R-3350 engines. Three XC-46Ls with R-3350 engines were delivered in 1945. The single C-46G became the XC-113 after it was fitted with one General Electric TG-100 turboprop engine as a flying test-bed. When wartime production of the C-46 finally ended, a total of 3,182 aircraft in all variants had been completed.

The first production C-46s, by this time dubbed Commandos by Curtiss-Wright, were initially delivered to the Military Transport Division (MTD) of Eastern Air Lines during the middle of 1942 for service evaluation, and by September had been placed in MTD service along the South Atlantic ferry route. Eastern’s experienced airline pilots recommended more than 300 modifications that were subsequently incorporated in C-46 production.

Due to its superior range, operating altitude and lifting capacity, the C-46 saw its most extensive employment in the China-Burma-India (CBI) theater, where it was used to ferry supplies over “the Hump”— the Himalayas—between India and China. During wartime service, the C-46’s maximum overload takeoff weight rose to 50,000 pounds (and often more than that, according to unofficial sources), and in the CBI theater particularly C-46s were routinely required to sustain flight operations under primitive conditions—unpaved airfields and local maintenance—for which they had never been intended.

Whether deserved or not, in operational service the C-46 began acquiring a reputation as a dangerous aircraft. Many of its early accidents were eventually traced to faults in the propeller pitch controls and vapor locks in the fuel system. And in the CBI, numerous C-46 crashes were the direct result of operating an overloaded aircraft above its operational ceiling in some the world’s worst flying weather. Pilot experience (or inexperience) was undoubtedly a factor too, for in critical emergencies the C-46 was not nearly as forgiving as the C-47. Michalek, who was an experienced multi-engine pilot before he went to Korea, recalled that because of its large tail and side area, the C-46 could be very tricky to land in gusty crosswinds.

After WWII ended, many C-46s were retained in regular military service. A smaller number ended up in reserve units, but the majority were either placed in storage or sold as surplus. The largest postwar military user of the C-46 was the U.S. Air Force, operating the type in various active and reserve troop carrier and air transport units. The C-46 was replaced by Fairchild C-119s and C-123s during the 1950s. The last operational U.S. Air Force C-46s were used by the 1st Air Commando Group for jungle supply drops and deliveries during the Vietnam War, and were finally retired in 1968.

As wartime production slowed, Curtiss-Wright began actively making plans to introduce a new civil variant, to be marketed as the 36-passenger CW-20E, an R-3350-powered version with a stepped windscreen that was based largely upon the canceled XC-46K. The company received a provisional order from Eastern Air Lines sometime in mid-1945, but soon after the war ended the huge influx of military surplus C-46s and C-47s into the civilian market—at comparatively nominal costs—effectively destroyed any potential market for the CW-20E. Eastern subsequently canceled its order, and the project was soon abandoned.

Surplus C-46A/D/Fs were put into service in the late 1940s by major air carriers such as Pan American, Delta and Braniff, but only in cargo configurations. Northeast Airlines, American Export & Import Company and Wein Air Alaska are known to have operated C-46s in both 36-seat passenger and cargo versions. From the late 1940s onward many large all-freight carriers like Slick Airways, Flying Tiger Line, Riddle Airlines (later Airlift International) and Zantop International operated sizable fleets of C-46s on their medium-haul routes. Civilian use reached its peak in the early 1960s, when the type reportedly served with more than 90 scheduled, nonscheduled and charter air carriers. Over the next two decades, however, C-46s were rapidly replaced by turboprops and jets; only about 60 remained on the U.S. registry by 1980, mostly with very small freight and charter operators. Today few, if any, are operating commercially in the United States. The Commemorative Air Force maintains at least two C-46s in flying condition: Tinker Bell, with the headquarters squadron at Midland, Texas; and China Doll, based with the Southern California Wing. Both of those aircraft can often be seen at airshows.

Beginning in the 1950s, many surplus and ex-U.S. civilian C-46s made their way into the hands of air operators around the world. They were especially popular in Central and South America, where their functionality at higher altitudes enabled them to fly routes across the mountain ranges of those regions.

By 1960, some 600 C-46s were estimated to be operating in South and Central America, and in contrast to U.S. carriers, many of the foreign operators routinely used their C-46s to haul passengers as well as freight, sometimes filling as many as 62 seats in a high-density layout. By 1980, the number of active C-46s in the region had dwindled to about 60 aircraft. It seems likely that a few tired old examples are still down there below the equator, doing what they do best—hauling people and freight in and out of primitive airstrips and over rugged terrain.

In order to truly appreciate the size and mass of a C-46, you need to see one up close. There are quite few on exhibit around the country, including at the National Museum of the U.S. Air Force in Dayton, Ohio; the Naval Aviation Museum in Pensacola, Fla.; the U.S. Army Airborne Museum at Fort Bragg, N.C.; the Glenn H. Curtiss Museum in Hammondsport, N.Y.; and the Castle AFB Museum near Merced, Calif.


E.R. Johnson writes frequently for Aviation History on aircraft types. For additional reading, he suggests: Curtiss Aircraft 1907-1947, by Peter M. Bowers; and Curtiss-Wright: Greatness and Decline, by Louis R. Eltscher and Edward M. Young.

Originally published in the September 2007 issue of Aviation History. To subscribe, click here.