A Brief History of Aircraft Carburetors and Fuel Systems
Part 6: Ideas, People and Companies

by Terry Welshans
Bardstown, Kentucky
for the Aircraft Engine Historical Society
August 2013; Revised 29 May 2022

Table of Contents

Glossary
Bibliography

Carburetor Patents: 1920 and Beyond

Discovery of new methods of controlling the fuel–air ratio for all engine loads and speeds continued for more than thirty years, until Turbojet Engines became the predominant aircraft power plant. The advanced fuel controls, with additional sensors and with new portions added, evolved into fuel controls for the newly invented Jet Engines. Many patents issued over the next three decades promoted a large number of improvements to existing carburetors, as well as the development of a number of new designs. Table 1 lists only a small portion of those important patents.

Carburetor Inventors

Many people have received patents for their designs and improvements in the design of aircraft carburetors. Prior to World War I, carburetors were in their infancy, and there were hundreds, if not thousands of patents in this field. Most of these carburetors never left the paper they were printed on, most for very good reasons. Some of these inventors were successful, and their names are familiar to us today. The pioneer inventors have made contributions to carburetors with familiar company names such as Zenith, Marvel, Stromberg, Solex, Robert Bosch, Holley, and the less well known carburetors made by Niles Bement Pond, CECO and Eclipse.

Three of these inventors need special recognition, their work being the holder of the original patents for American aircraft carburetors developed just before World War II. These carburetors, along with the developments made in fuel injection systems brought the art of carburetion to its peak, allowing aircraft engines designed in the 1930-1950 era to meet and better the "one cubic inch-one pound of weight-one horsepower" goal.

Milton J. Kittler, Milton E. Chandler and Franklin C. Mock

Milton J. Kittler and Milton E. Chandler patented a basic pressure type carburetor in 1937. Franklin C. Mock worked at Bendix Stromberg and Chandler. After leaving Bendix, he began working on designs for Chandler Groves and later CECO, refining the pressure carburetor to such a degree that the basic components became fuel injection controls, and eventually into fuel controls for gas turbine engines. The story of the development of the floatless carburetor is contained in a short biography of Milton E. Chandler’s patents and the people and companies with which he was involved. We must begin with Milton J. Kittler, who submitted a patent for a floatless carburetor that was the basis from which Milton E Chandler made improvements. The improvements became patent 2,088,464, issued to Milton E. Chandler and Milton J. Kittler on July 27, 1937. Mr. Kittler awarded his portion of the patent rights to Milton E. Chandler.[131] Milton E. Chandler began working for Stromberg Motor Devices in 1929 as an engineer, and quickly rose through the ranks. By 1934, he was vice president of the Bendix-Stromberg Carburetor Company. Stromberg was at that time a division of the Bendix Corporation, and that placed Chandler in charge of all Stromberg carburetor engineering. He was instrumental in developing all new carburetor technology of that time. Chandler was a prolific inventor with a number of patents, and assigned many, but not all, of his patents to Stromberg.[132] By 1934, Chandler had become dissatisfied with the policies of Bendix management, particularly with the small amount of money they budgeted to research and development. At time, Stromberg policy required the engine manufacturers to pay for most, if not all, of the development cost of any work done for the engine builder’s benefit. Chandler knew that Wright, one of the two major engine manufacturers, was dissatisfied with this arrangement to such a degree that Wright Aeronautical, on its own, began developing a new type of carburetor that it intended to manufacture for use on Wright engines. Despite this, Chandler was still unable to obtain any funds from Bendix for the rapid development of the new carburetors he was developing. Overall, his carburetor designs and patents did not strike much interest in his employer. Chandler wanted to start a new company to compete with Stromberg. He had found that the military services and Wright Aeronautical would be delighted to see competition in the carburetor industry.

Holley Carburetor Company was a manufacturer of automobile and light-airplane carburetors, and had tried to produce a large aircraft carburetor as a private venture, but was only partial successful due to its lack of experience in the field. Chandler eventually received the necessary backing from Holley, and in January 1935, Chandler left Stromberg to become president of the new Chandler-Groves Company. Franklin C. Mock, who had been with Stromberg some years before 1925, replaced Milton E. Chandler when he left Bendix-Stromberg. By 1935, Mock was in charge of the development of fuel injectors at Eclipse Aviation Corporation, another of Bendix’s subsidiaries. Mock had talked about building a floatless carburetor with a discharge nozzle at the supercharger inlet, and had a small contract from the Navy, but had made little progress with it.[133] After returning to Stromberg, Mock's first job was to get satisfactory performance from the Stromberg float-type carburetor. He gave up this approach when a series of Navy tests run on a Wright Cyclone engine in 1935 showed that the Chandler-Groves floatless carburetor was decisively better than the best that Mock could get from the Stromberg float-type carburetor.[134] Mock convinced Bendix management that it was necessary for Stromberg to rapidly develop a floatless carburetor, or the aircraft carburetor market would be lost to the competition. Wright and the Navy definitely preferred the Chandler-Groves to the Stromberg float-type carburetor, but the Army preferred Stromberg to Chandler-Groves. Additionally, the Army had declared in 1934 that all single-engine fighters would soon be equipped with fuel injection.

The number of employees working on Stromberg aviation carburetors quickly increased from four to about ten engineers. Stromberg expenses for aircraft carburetors development in 1935 were twice that of 1934 and in 1936 and 1937, they were still greater by an additional 25%.[135] Before 1936 ended, Stromberg had acquired an engine test stand: the significance of this was not small, as it was a signal that Stromberg had given up relying on tests run by the engine companies and the essential information obtained only by running carburetors on actual engines.

Mock's new design eliminated the principle of supplying fuel to the jet at a constant, but low pressure and using the venturi suction to create the fuel flow. In his new design, fuel forced through the jet by pressure in the fuel supply metered the fuel flow.[136] The carburetor included a venturi, whose suction creates a force on one side of a diaphragm with pressure from the inlet duct on the other, while the drop in pressure across the fuel jet created an opposing force on another diaphragm. The two linked diaphragms also connected to a valve that regulated the fuel supply to keep the forces on the two diaphragms equal, thereby maintaining the fuel flow in proportion to the airflow. Bendix discovered that the principle of using the double-diaphragm system received a patent several years before for use as a chemical process controller. Bendix bought that patent.[137]

Chandler intended his new company to compete with Stromberg partly based on improved performance, but principally based on lower price. He knew that the Stromberg carburetor consists of a set of very complicated castings, machined in a long series of operations. If an error occurred at any step in the process,it was necessary to scrap the part, wasting all the previous work. The scrap rate was as much as 25% on the latest Stromberg models, and frequently brought production to a complete stop. A carburetor designed for simpler production promised an important advantage in price; that met Chandler's goal. A secondary benefit of his design was that Chandler-Groves simply did not have foundry facilities capable of producing castings for a Stromberg-type carburetor.

The Chandler-Groves carburetor differed from the Stromberg float-type aircraft carburetor in two respects. One of Chandler’s carburetor patents used a variable venturi as the throttle, a completely different design compared to the disk-type butterfly valve that was located in the air stream after the air had passed through a fixed venturi.[138] The Chandler venturi had a variable cross-section so that it could serve as the throttle before the fuel spray entered the airflow. With no fuel spray hitting the throttle; it was not as likely to freeze up in icing conditions. It also was much simpler to make this new carburetor because the rectangular throttle pieces made it possible to build the carburetor by simply bolting a number of castings together. As a bonus, each casting was very simple and easy to cast. A series of openings located at the narrowest opening position of the venturi/throttle plates measured the vacuum created by the air flowing through the venturi. A number of small impact tubes before the venturi measured the pressure of air flowing into the carburetor. Using the venturi and impact pressures, an internal diaphragm controlled fuel flow.

The second characteristic feature of the design was how the fuel pressure delivered to the jet remained at constant pressure. The valve controlling the entrance of fuel to the chamber was a pressure regulator to keep the pressure within the chamber constant. The pressure of the fuel on the diaphragm moves the valve that controlled the fuel flow, and therefore, the pressure in the fuel bowl. This change eliminated problems associated with floats in an aircraft carburetor. Chandler's primary motive, however, was to incorporate a novelty feature (a "selling point") to compete with Stromberg. The float was satisfactory in most types of flight. The need for fighter and dive-bomber engines, where the float carburetor did not work well, was only a small part of the total engine market.

Initially, the new carburetor received only a minimum of resources. Prototypes were ready for test early in 1935, but Chandler-Groves had neither facilities for running the carburetor on an actual engine nor an "air box" that was large enough to test a carburetor of this size. Wright Field conducted the first air-box tests, and Wright Aeronautical Corporation plant made the first runs on an actual engine.[139] Although Wright and the military services were eager from the beginning to have competition in the carburetor industry, no one was very enthusiastic over the Chandler-Groves carburetor design until late in 1935. The Navy’s inflight testing of the new carburetor demonstrated that it was very much less subject to icing than the Stromberg float-type carburetor. The Cyclone engine lost a great deal of power when using air heated to approximately 150°F, the temperature required to eliminate ice, once ice formation had started. Both Wright and the Navy at once became very much interested, and gave wholehearted support, to further development of the Chandler-Groves carburetor.[140]

Lacking the necessary facilities in his own plant, Chandler himself participated in the very extensive running of the carburetor on actual engines at Wright's expense at the Wright plant, and on the refrigerated air box at the Naval Aircraft Factory at the Navy's expense. During testing of the new carburetor on the Navy air box, which could simulate performance at altitude at the proper temperature as well as the proper pressure, there a completely unexpected feature was discovered. The carburetor had an aneroid mixture control, but the original model had been unsatisfactory and the carburetor test continued with it disconnected. This running led to the observation that automatic mixture control or altitude compensation appeared to be inherent in the basic carburetor itself, although the reason was unknown.[141] Elimination of the now unnecessary special mixture control resulted in a price saving, and reducing trouble due to its increasing complexity. The great deal of assistance contributed to the development of the Chandler-Groves carburetor in the form of experimental running by Wright and by the Navy meant that even though Chandler had taken no development contracts, in order to avoid loss of rights in the design, his company had to invest some $100,000 in development costs before placing the carburetor in production in the first half of 1937.

The Chandler-Groves carburetor became standard on all Navy Cyclone G-100 models that had entered production about this time. The airlines also became interested in the new carburetor as soon as they learned of its performance.[142] The airlines were attracted by the freedom from icing, and by the lower price, which had been the guiding principle of Chandler’s original design. In March 1937, TWA chose a Chandler-Groves carburetor for its regular service, and it became standard on new airline Cyclones G-100 models, along with the Navy engines. The Army, however, refused to use the Chandler-Groves carburetor. Engineers at Wright Field had discovered that its "magic" altitude compensation was due, at least in part, to the absence of any provision for the escape of gasoline vapor. This resulted in the fuel boiling from the reduced pressure at altitude , thereby leaning the mixture. But that was not the only factor involved – the variable venturi of the carburetor gave the intake air a velocity at or above the speed of sound when operating at high altitude. This also tended to lean the mixture. Neither of these principles was ideal as a method for altitude compensation, as the amount of compensation was partly dependent on the vapor pressure of the fuel. This meant that the amount of compensation would vary with fuel’s volatility. The Army was unwilling to use a carburetor based on an unsound principle, and made no flight tests, preferring to use the old float-type Stromberg carburetor, which by 1936, was equipped with a true automatic mixture control. This loss of sales resulted in the Chandler-Groves Company going out of business.

After the Chandler-Groves Company dissolution in 1938, a new partnership formed between Milton Chandler and Edward S. Evans Jr., the owner of the Evans Appliance Company, forming the Chandler Evans Company (CECO), with financial backing from Niles-Bement-Pond, the largest machinery manufacturer at the time.[143] Chandler immediately began designing a new type of floatless carburetor. Chandler's new design, which he called the hydro-metering carburetor, resembled the new Stromberg carburetor except for the way the Chandler-Groves used the diaphragms to measure air flow and fuel pressure. The fuel could be discharged at any location rather than using the venturi suction directly to draw the fuel out of the fuel bowl as in a float-type carburetor. The basic difference from the Stromberg was that instead of balancing the fuel and air pressure drops directly against each other, the Chandler-Groves carburetor used a hydraulic connection that permitted a very small air suction to control a much larger drop in fuel pressure. This was to be very valuable as it was then possible to have much less constriction in the venturi, thereby reducing losses.

Wright Field, Wright Aeronautical, and Pratt & Whitney all energetically gave support and assistance to CECO carburetor development. Wright Aero was particularly interested in the CECO carburetor for the new Wright R-3350 engine then under development. This new engine was for the new Boeing B-29. Wright tried to persuade the three carburetor suppliers to produce a suitable product. All three agreed, but Holley never produced a carburetor for test, and the original Stromberg entry, besides being enormous, had many imperfections that Stromberg was unable to eliminate due to the volume of other work at hand. In 1942, Wright decided to use the CECO carburetor on the R-3350, even though its development was not yet complete.

The Army and Pratt & Whitney desired a second source of carburetors, because Stromberg did not have a large production capacity. They were completely dissatisfied with Holley, and were interested in the CECO for the R-1830 engine, used on the B-24 in very great numbers. In 1942, the Army decided on using the Holley 1375F carburetor on the B-24's R-1830 engines, subject to service test. In the fall of 1942, CECO delivered hydro-metering carburetors for service test on the R-1830 engines. The usual difficulties soon appeared, particularly with acceleration, and while Chandler and the Army were attempting to eliminate these, Pratt & Whitney continued to have trouble with inconsistent performance on the test stand. Early in 1943, Pratt & Whitney decided that the need for production was so urgent, they would alter the CECO carburetor to use the already-approved Stromberg regulating unit. Pratt & Whitney produced an experimental carburetor that was a success on the test stand. Stromberg gave CECO a license to use their control system, and the CECO "direct-metering" carburetor went into production. This is the carburetor used on the B-24 R-1830 engines.

The original larger CECO carburetor worked better on the R-3350 than the smaller size designated for the R-1830. The Army preferred to change the CECO on the R-3350 to the direct-metering system, but production capacity prevented this. The revised CECO unit was working well by the time the R-3350 entered production. All of the B-29's R-3350s received the CECO direct-metering carburetor, except for a few converted to fuel injection just before the end of the war. At war's end, the Navy was considering it for use on the Pratt & Whitney R-4360. The Army adopted fuel injection on all its large engines after the war, and production of the small number of carburetors sold for Navy and airline use did not promise to be profitable, resulting in CECO leaving the carburetor business.

Chandler Evans Company History

For many years, the intertwined Chandler Evans and Niles-Bement-Pond Company (NBP) bought and sold subsidiaries, merging some and divesting others. For that reason, we must first look at the Niles-Bement-Pond Company family tree. In 1898, Niles Tool Works purchased control of the Pond Machine Tool Works. During the next year, a great consolidation took place when several major builders of large machine tools, including the Niles Tool Works, Bement, Miles & Co., the Pond Machine Tool Company, and the Philadelphia Engineering Works merged, becoming the Niles-Bement-Pond Company.

In 1901, NBP purchased the Pratt & Whitney Machine Tool Company (P&WMT) which led to reorganization at P&WMT. At that time, internal bickering within P&WT was tearing the company apart, and it was unable to resist a take-over bid. Mr. F.W. Gordon soon traveled to Hartford, where he acted as General Manager of P&WMT as a representative of NBP. He thoroughly reorganized the shop. Under Mr. Gordon the company’s officers paid more attention to manufacturing on a quantity basis. NBP acquired additional companies including John Bertram in Canada, Ridgeway Machine Company and the Milwaukee Machine Tool Company, along with numerous smaller companies. Because of expanding markets, improved export sales, a strong home trade and their take-over activities, NBP became, at least for a time, the largest machine-tool company in the world.

Mr. Dudley Seymour, Mr. C.C. Tyler and Mr. B.M.W. Hanson were associates of Mr. Gordon at NBP. Mr. Hanson afterward became a Vice-President. Later, NBP consolidated its businesses in New York once many P&WMT officers departed the company. Mr. Hanson became the General Manager, a position he held until July 1917. Pratt & Whitney Machine Tool Company became the P&WMT Division of NBP. In 1914, the Pope Manufacturing Company buildings, which were located next-door to P&WMT Division in Hartford, became available, and NBP purchased the entire plant. This supplied a great deal more floor space than needed. Having idle factory space and capital available for investment wherever good return seemed available, P&WMT saw the postwar aviation industry, both military and civil (commercial, private), as one with some of the greatest growth and development prospects available anywhere for the next few decades.

In April 1925, Frederick Rentschler, an Ohio native and a former executive at Wright Aeronautical, was determined to start an aviation-related business of his own. His social network included Edward Deeds, another prominent Ohioan of the early aviation industry, and Frederick's brother Gordon Rentschler, both of whom were on the board of NBP. Frederick Rentschler approached these men as he sought capital and assets for his new venture. Edward Deeds and Gordon Rentschler persuaded NBP Board that their P&WMT subsidiary should provide the funding and location to build a new aircraft engine then under development by Frederick Rentschler, George J. Mead, and their colleagues, all of whom were formerly of Wright Aeronautical. Conceived and designed by Mead, the new engine would be a large, air-cooled, radial design. P&WMT loaned Rentschler $250,000, the use of the Pratt & Whitney name, and space in its building. This was the beginning of the Pratt & Whitney Aircraft Company. Pratt & Whitney completed its first engine, called the Wasp, on Christmas Eve 1925. The Wasp developed 425 hp on its third test run. It easily passed the Navy qualification test in March 1926, and by October, the Navy had ordered 200 engines under the military designation R-1340. The Wasp exhibited speed, climb, performance and reliability that revolutionized American aviation. The R-1340 engine powered the aircraft of Wiley Post and Amelia Earhart, along with aircraft used on many other record flights.

P&WMT was going through a period of self-revision at the time in order to prepare itself for the post–Great War era by discontinuing old product lines and incubating new ones. The Great War had been profitable to P&WMT, but the peace brought a predictable glut to the machine tool market, due to canceled government contracts. The used and recently rebuilt machine tool market competed against newly manufactured machines. P&WMT's future growth would depend on innovation. In 1929, Frederick Rentschler ended his association with P&WMT and formed United Aircraft and Transport Corporation, the predecessor to today's United Technologies. The agreement allowed Rentschler to continue use of the Pratt & Whitney name with his new corporation.[144]

In 1938, a partnership between Milton Chandler and James Evans, the owner of Evans Appliance Company, formed the Chandler Evans Company (CECO). In 1943, NBP acquired CECO, and in 1945, Chandler-Evans absorbed NBP. In 1954, Pennsylvania Coal and Coke Corporation diversified from mining into manufacturing, changing its name to Penn-Texas Corporation. Penn-Texas then bought control of financially troubled Colt's Firearm Manufacturing of Hartford, Connecticut. In 1955, Chandler Evans merged into Penn-Texas Corporation, spinning out the P&WMT division. The Company’s name reverted to the original Pratt & Whitney Machine Tool Co. name. In 1958, Penn-Texas merged with Fairbanks-Morse, and the name of the combined company changed to Fairbanks-Whitney. In 1959, gun maker Colt Industries, continuing to operate the gun factory until 1999, when it acquired CECO. In the following years, CECO supplied the MFP-100 fuel control for the Pratt & Whitney J58 engines used on the Lockheed A-12, YF-12A and SR-71. In 1964, Fairbanks-Whitney changed its corporate name to Colt Industries, and then sold its firearms-making business, changing its name to Coltec Industries. In 1999, Goodrich acquired Coltec Industries and CECO was renamed the Goodrich Pump and Engine Control Division of the Goodrich Corporation.

On July 26, 2012, United Technologies Corporation purchased Goodrich Corporation, integrating it with Hamilton Standard and Sundstrand, forming a new division of parent United Technologies Aerospace Systems. That division is part of a larger group within United Technologies Corporation, called UTC Propulsion and Aerospace Systems, which also includes East Hartford-based Pratt & Whitney Aircraft Company. Frederick Rentschler’s United Aircraft and Transport Corporation is the predecessor of today's United Technologies. Pratt & Whitney Aircraft Company has returned to its original ownership. The world is a small place indeed. On March 18, 2013, Triumph Group, Inc. announced that it completed the acquisition of the pump and engine control systems business of Goodrich Corporation (Goodrich Pump & Engine Control Systems) from parent United Technologies Corporation. The business will operate as Triumph Engine Control Systems, LLC and be included in the Triumph Aerospace Systems Group.[145][146][147]

Rob Kolp, Vice President, R&D, The Lee Company, recalls hearing many of Leighton Lee's stories of the early days and his transitioning from carbureted and fuel injected engines to jets. Mr. Lee spoke of Mr. Chandler and how he influenced the path of The Lee Company. According to Mr Lee, many of the wartime patents were not issued until later because of secrecy. For example US 2,402,332 was filed in May 1944 and US 2,423,394 was actually filed earlier in December 1943. During the war, Mr. Lee traveled between Hartford and Wright-Patterson AFB to check on carburetor installations and institute aircraft upgrades. As the war progressed he was moved over to work on the secret jet program. When the war wound down he felt he could use his expertise and the contacts that he had made during the war and start his own company. US 2,660,474 was an early jet development program through The Lee Company. In the early days these patents were just assigned to him.

Endnotes

[131] US Patent 2,088,464.
[132] Development of Aircraft Engines, pp 517-521.
[133] Development of Aircraft Engines, p 521.
[134] Development of Aircraft Engines, p 521.
[135] Development of Aircraft Engines, p 522.
[136] Development of Aircraft Engines, p 522.
[137] Development of Aircraft Engines, p 522.
[138] US Patent 2,088,464.
[139] Development of Aircraft Engines, p 522.
[140] Development of Aircraft Engines, p 522.
[141] Development of Aircraft Engines, p 522.
[142] Development of Aircraft Engines, p 522.
[143] Development of Aircraft Engines, p 525.
[144] Excess Profits, pp 23–53, ch 2.
[145] The Hartford Courant, Mar 18, 2013.
[146] www.goodrich.com/history.
[147] http://www.prattandwhitney.com/history.

Send mail to with questions or comments about this web site.
This website depends on cookies to make it function. If you continue to browse, scroll, click or otherwise interact, you are providing implicit acknowledgement of and agreement to this.
Copyright © 2002-2024 Aircraft Engine Historical Society, Inc.