USAAF Compound Engine Project

On 17 Jul 1940, U.S. Army Air Corps Materiel Command Engineering Division Power Plant Laboratory (PPL) civilian employees Opie Chenoweth and A.L. Burger met with Allis-Chalmers representatives C.C. Jordan, R.C. Allen and Dr. J.T. Rattalista at the Allis-Chalmers plant in Milwaukee, Wisconsin to discuss development of a gas turbine for use in a compound engine scheme involving the Allison V-1710 engine. At that time, Allis-Chalmers built a Houdry oil refining process compressor consisting of a 5-stage gas turbine used to drive a 23-stage axial compressor that furnished compressed air for the refining process. With 1,000°F turbine inlet temperature, the gas turbine produced about 1,600 hp at 3,600 rpm. Designed to run for hundreds of thousands of hours, its construction was quite heavy at about 10 lb/hp; this ruled it out as a prime mover. However, Allis-Chalmers was working on turbine materials that would permit operation at 1,500°F, and thought it might be possible to build an acceptable exhaust gas turbine for engine compounding. PPL thought a compounded engine would be able to operate a high altitudes with better specific fuel consumption than the piston engine alone.

After discussing the USAAC needs and Allis-Chalmers capabilities, both parties agreed that the scheme showed promise when based on engine cruising operation at 1,200 hp, 3,000 rpm and a turbine nozzle box temperature of 1,500°F. A 0.110 lb/bhp/min air consumption was expected with about 66% of the exhaust passing through the turbine. Allis-Chalmers agreed to further study and to submit a proposal within 30 days. There is no record that the proposal was ever delivered. [24 Jul 1940 Memorandum Report EXP-M-57-522-48, P100275]

On 2 Mar 1942 the U.S. Army Air Corps became the U.S. Army Air Forces (USAAF) On 18 Nov 1943, PPL Chief Col James Milligan Gillespie wrote Engineering Division Chief Brig Gen F.O. Carroll noting that the Compound Engine Project operating under expenditure order (EO) No. 503-479 had reached the stage where contractors would be contacted and suggested that the project be classified Confidential a classified project number be assigned. Gen Carroll replied on 20 Nov 1943 that the project had been so classified and that confidential project number MX-467 has been assigned.

On 20 Jan 1944, Continental Aviation and Engineering (CAE) representatives R. Insley, J.W. Kinnucan and Carl F. Bachle met with Col J.M. Gillespie, Col G.F. Smith, Maj P.F. May, J.G. Blackwood, N.A. Wolfe and Ford L. Prescott of the PPL at Wright Field concerning compound aircraft engines. The group discussed completion of the three-cylinder test engine, a possible two-stroke engine project to be compounded with a turbine, and the proposed release of specifications and requests for bids on a turbojet propulsion device.

CAE was asked if it would be interested in participating in a compound engine development program, including finishing the partially-complete three-cylinder test engine the PPL had been building. CAE was aware of the compound engine research then ongoing and, based on its experience with O-1430 and IV-1430 development for the past 12 years, had no desire to compete with programs already under way. CAE had built, tested and delivered a 10-cylinder two-stroke radial for the Navy; they found piston problems were worse with two-stroke engines than with four-stroke, and their experience had left them unimpressed with the two-stroke engine field.

Discussion then turned to gas turbine projects for which the Air Technical Services Command (hereinafter ATSC, which was previously the Engineering Division) was preparing specifications and bid requests. This development was interesting to CAE as the O/IV-1430 projects had left an engineering and testing organization that needed work and could contribute to the war effort. CAE was also keen on developing a compounded IV-1430 if the Materiel Command (hereinafter MatCmd) was considering that engine for any future aircraft development. The ATSC agreed to make three-cylinder and gas turbine requirements available to CAE as they became available.

It was concluded that large engines with low specific fuel consumption were necessary for long range aircraft, and that these aircraft should use four engines. MatCmd recommended that CAE complete the three-cylinder test engine. [7 Feb 1944 Memorandum Report ENG-57-503-1151, P281395]

On 3 Mar 1944, Col Gillespie sent E.A. Hulbert of Continental Aviation and Engineering Corporation four connecting rod drawings for a three-cylinder compound engine that PPL and Continental appeared to be jointly developing:

X43017163 – Rod, Connecting, L.P. Cylinder
X43017164 – Rod, Connecting, H.P. Cylinder
X43023383 – Rod, Assembly, L.P. Connecting
X43023515 – Rod, Assembly, H.P. Connecting

At the 3 Mar 1944 meeting CAE representatives E.A. Hulbert and Carl L. Schiller also met with Capt W.L. Bassett and Ford L. Prescott regarding fabrication and test requirements for the three-cylinder compound engine. Assembly and detail drawings, along with lists of completed and partially completed parts were reviewed. Most small parts were complete, but the crankcase, cylinder blocks and crankshaft were not. No drawings had been made of the flywheel, dynamometer coupling, timing gear assembly and magneto drive. A separate, independently-powered oil pump was to be used, thereby simplifying the basic engine. In addition to preliminary tests, performance tests at various power outputs and operating conditions were to be made, culminating in a 50-hr endurance test that would insure a 5,000 hp multi-cylinder engine based on this design would have desirable endurance characteristics. Hulbert thought the CAE dynamometer stand No. 10, which had been used for O-1430 development, would be useful for this testing. Hulbert requested copies of assembly and detail drawings to help expedite a proposal for the three-cylinder work. [9 Mar 1944 Memorandum Report ENG-57-503-1189, P281272]

At a 17 May 1944 meeting with the Aircraft Laboratory, Allison representatives Atkinson and Dixon requested a set of typical curves depicting brake horsepower required per engine, cruising speed, and altitude versus percent range for two differently loaded long range airplanes. Allison intended to use this data to prepare specific fuel consumption data for the airplane types indicated, with propeller rpm varying from 700 at the flight's start and 500 rpm at its end. On 1 Jun 1944, the Aircraft laboratory sent Allison the following:

Curve Sheet No. 1 for a heavily overloaded airplane flying at long-range ferry conditions, not subject to combat, at the point of 30% range.
Curve Sheet No. 2 for a normally loaded airplane with good all-around performance, with little compromise in range and subject to combat conditions at 30% range. Of particular interest was what the compound engine could do to the range of this airplane.

On 8 Jun 1945, Assistant Chief of Air Staff Materiel and Services Brig Gen J.F. Phillips wrote the Engineering Division Chief noting that while the AAF had for several years been interested in compounded engines with a view toward lowering cruise specific fuel consumption. However, war efforts had taken priority over research and development. Gen Phillips requested a brief resume of AAF engine compounding projects under consideration, along with their priority, developmental status, and Air Technical Services Command's plans for each.

On 23 Jun 1945, Col P.H. Robey, Chief of the Engineering Division Propulsion & Accessories Subdivision, responded to Gen Phillips' request with a list of companies and a discussion of their projects.

USAAF Contract W-33-038-ac-2737 with GE, dated 15 Jun 1944 for $248,000.00 and with a 3A priority, was intended to explore Pratt & Whitney R-4360 turbocompounding and included deliverables:

Item 1 – Eight theoretical reports on various methods of using gas turbine power, including gearing to the engine, driving a supercharger compressor, driving a cooling fan, and the combinations of series and parallel flow for the gas turbine and turbosupercharger system.
Item 2 – A gas turbine that could be inserted in series between the engine and turbosupercharger for altitude supercharging.
Item 3 – A report on a Government-furnished (GFE) R-4360 operated on a test stand for calibration and to check back pressure effects on the engine configured as per Item 2.
Item 4 – reports on turbine operation when remote from the engine and providing supercharging for high-altitude operation. This involved 180 hours testing in a GE high-altitude test stand capable of producing pressure altitudes sea level to 40,000 ft. Data to be collected included engine and turbine power output and maximum bmep, rpm, turbine speed, along with qualitative observation of compressor pulsation and closed-waste-gate engine operation.

The contract Items were to be delivered within 15 months of contract signing, provided the turbine was approved within 3 months of contract signing. On 19 Jun 1945, GE representatives Fisher and Sawtell visited the Air Technical Services Command (ATSC) and explained that little progress had been made because of other, higher-priority projects. GE was considering a system that used extra power from the turbine, which, after driving a supercharger, would drive an axial-flow compressor that supplied cooling air to a radiator or to an air-cooled engine, both at sufficiently high pressure to then pass through a nozzle and produce jet thrust. This scheme increased power plant power and lowered fuel consumption without mechanically connecting the turbine to the engine. (Cooling of engines operating at extreme altitudes was a problem due to the low air density. Those advancing the art were considering schemes to remove waste heat from both air- and liquid-cooled engines.) GE planned to send a letter to the ATSC explaining this concept.

Contract W-33-038-ac-7191 with Allison, dated 5 Jan 1945, for $148,524, sought to mount an exhaust turbine aft of a V-1710-E auxiliary stage supercharger and connect the turbine output to the crankshaft via a shaft extending through the auxiliary supercharger impeller shaft. Allison envisioned this engine, with a 2.48:1 propeller reduction gear, for use in a Bell P-63 airplane. This project had a 3A priority. The following military ratings were predicted:

1,450 bhp at 3,000 rpm and sea level
1,600 bhp at 3,000 rpm and 25,000 ft
1,550 bhp at 3,200 rpm and 29,000 ft

The contract included the following deliverables:

Item 1 – Report covering at least 300 hours of dynamometer calibrations and preliminary development tests on a contractor-furnished V-1710-127 (E-27) engine. The report was due 270 days after receipt of the GFE GE CT1 exhaust turbine.
Item 2 – Services and all new engine parts to convert three GEF V-1710-109 to model V-1710-127, incorporating all changes found necessary during the execution of Item 1. Delivery was scheduled 90 days after receipt of the GFE engines and exhaust turbines or 210 days after contract approval, whichever was later. Additional engines were to be delivered every 20 days.
Item 3 – Report covering 50-hr Preliminary Flight Rating Test on a GFE V-1710-127 engine, delivery 60 days after receipt of the GFE engine.
Item 4 – Report covering 7.5-hr War Emergency Rating Test based on a rating based on Item 1 results and to be established by mutual agreement between the Government and Allison. This testing was to be conducted using Specification 28R Grade 100/130 aviation gasoline. Delivery was to be 30 days after satisfactory completion of Item 3 testing or 30 days after receipt of GFE engine, whichever was later.
Item 5 – Report covering 7.5-hr War Emergency Rating Test based on a rating based on Item 1 results and to be established by mutual agreement between the Government and Allison. This testing was to be conducted using 115/145 aviation gasoline. Delivery was to be 30 days after satisfactory completion of Item 4 testing or 30 days after receipt of GFE engine, whichever was later.
Item 6 – Drawings and parts lists, delivery 30 days after delivery of the last Item 2 engine.

This project involved the use of three GE CT1 exhaust turbines, which were similar to the GE CH5 turbosupercharger except the turbine rotated in the opposite direction to accommodate the Allison engine power transfer gearing. One CT1 turbine had been delivered to Allison and testing was expected to begin at once. This engine's weight, complete with auxiliary stage supercharger, extension shaft, reduction gearing, exhaust turbine, automatic water injection controls, exhaust manifolds and exhaust shrouds, was predicted a 1,983 lb. Further development using aftercooling was also anticipated. Allison projected the following ratings with 50% aftercooling:

Military Rating:
1,450 bhp at 3,200 rpm and sea level
1,620 bhp at 3,200 rpm and 30,000 ft
War Emergency Rating using Specification 28R Grade 100/130 Fuel
1,900 bhp at 3,200 rpm and sea level
2,030 bhp at 3,200 rpm and 24,000 ft
War Emergency Rating using Grade 150 Fuel
2,370 bhp at 3,200 rpm and sea level
2,430 bhp at 3,200 rpm and 18,000 ft

The ATSC Allison project engineer was scheduled to visit Allison on 19 Jun 1945 to discuss the V-1710 project and also the possibility of compounding the V-3420. A suggested scheme used one turbine similar to the V-1710-127 development geared to the engine and supplied with exhaust gas from two cylinder banks; a second exhaust turbine, connected to the other two banks, would drive a compressor of sufficient capacity to furnish cruising power at altitude. By using cross-over piping and valves it might be possible to use the turbosupercharger for military power at altitude, provided the turbocharger turbine and compressor had sufficient range. Various other compounding combinations were to be discussed.

While the ATSC had no direct compounding contract with P&WA, ATSC engineers had asked P&WA to cooperate with GE in the R-4360 compounding under GE contract W-33-038-ac-2736. P&WA had also submitted a report showing what might be accomplished if the dual superchargers were removed from a R-2800-E and replaced with exhaust turbines. A single speed ratio was to be used between the turbine and engine. While this scheme lacked the altitude performance necessary for fighter applications, it might have been applicable for long-range cargo or bombardment types.

Although Studebaker's then-current contract made no mention of compounding, the company was discussing the subject with GE. Studebaker had furnished GE with XH-9350 performance estimates and further work on compounding was anticipated.

Lycoming, engaged in getting the XR-7755 up and running, had postponed any discussion of compounding.

While the ATSC had discussed compounding with WAC on several occasions, most of WAC's efforts in this regard were directed at driving cooling fans rather than improving altitude performance. The U.S. Navy Bureau of Aeronautics was thought to be in discussion with WAC about a compounded engine that did not involve GE.

The compounded engines that were subjects of the NACA studies were probably Pratt & Whitney's R-2800 and Allison's V-1710.

Pratt & Whitney Turbo-Compound-Modified R-2800-E Estimated Performance
	Brake Horsepower (bhp)	Specific Fuel Consumption (lb/hp/hr)
Flight Regimen	Altitude (ft)	Engine Only	Compound	Engine Only	Compound
Military – Overdrive	4,500	2,200	2,550	0.795	0.654
Military – Low (Minimum Slip)	14,500	1,900	2,500	0.845	0.634
Military – High (Minimum Slip)	27,000	1,350	1,950	0.938	0.635
Cruise – Overdrive	10,500	1,350	1,150	0.426	0.379
Cruise – Low (Minimum Slip)	18,500	1,250	1,270	0.444	0.369
Cruise – High (Minimum Slip)	22,000	1,150	1,250	0.460	0.368

On 26 Jul 1946, seven P&WA personnel met with fourteen MatCmd personnel to deliver four reports and present some quarter-scale compounded R-4360 models. The Pratt & Whitney reports were:

TR-25 – Wasp Major R-4360-C Single-Stage Engine Compounded with Dual Feedback Turbines
TR-27 – Wasp Major R-4360-C Two-Stage Engine Compounded with Dual Feedback Turbines
TR-28 – Wasp Major R-4360-C Single-Stage Engine Compounded with Auxiliary Compressor, Burners and Feedback Turbine
TR-29 – A Comparison of the Estimated Performance of Several R-4360-C Compound Engines Operating Under Flight Conditions.

Based on the conference discussions it was agreed that P&WA would study R-4360 compounding as follows:

Extend data up to 45,000 ft
Extend range calculations to 10,000 miles
Consider fan cooling as an engine requirement
Include a comparison of single stage engine compounded with dual feedback turbines and with separate exhaust turbine superchargers (not coupled to engine) against a R-4360-C engine with turbine supercharger (not coupled to engine), and also against the single stage engine compounded with an auxiliary compressor, burners, and feedback turbine. Also comparison to other configurations presented in TR-29.
Use 50% and 85% normal rated power as values to be used in cruise power and economy studies.
Use speed schedules as in TR-28, Curve 9, for further studies.

During the conference it was stressed that fuel economy was more important than power increases for future aircraft use of reciprocating power plants of the R-4360's size.

On 17 Oct 1946, six P&WA and fourteen MatCmd personnel reconvened at Wright Field. P&WA presented six reports:

TR-33 was a performance analysis of a single-stage-single-speed R-4360-C compounded with dual CHM-2 feedback turbines and dual two-stage turbo compressors that provided supercharging to high altitude.
TR-34 was a performance analysis single-stage, single-speed R-4360-C supercharged to high altitude with two CHM-1 Turbosuperchargers with two-stage compressors.
TR-35 compared R-4360-C compounded engine systems that were highly supercharged and moderately supercharged, a conventional R-4360-C, and a form of propeller turbine power plant. (Could this have been the PT-1? More about this later.)
TR-36 estimated the performance of a R-4360-C engine compounded and supercharged by a J33-type turbojet unit.
TR-37 was a second analysis of a single-stage R-4360-C compounded with dual feedback turbines.

Based on P&WA's preliminary work, MatCmd awarded P&WA Contract No. W-33-038-ac-14294 for a single-stage, single-speed R-4360-c with two feedback turbines and two type CHM turbosuperchargers with two-stage compressors, for airplanes with 45,000 ft critical altitudes and with two type CH turbosuperchargers for airplanes with 35,000 ft critical altitudes.

A fan would be required for engine cooling at altitudes above 35,000 ft, and its most suitable location was forward of the propeller reduction gearbox. With the fan mounted between the propeller and engine, the problem of propeller control would have to be solved by the propeller manufacturer.

P&WA had considered using a contra-rotating propeller, but development had been postponed until a definite requirement existed. P&WA was to continue design studies on engine systems intended for 35,000 – 45,000 ft critical altitude operation. [7 Aug 1946 Memorandum Report TSEPP-503-3279, P350498]

These concepts evolved into the R-4360 VDT, which despite significant development effort, never entered production.

Work had been delayed by the war and work on gas turbine engines; compound engine turbine work was in 11th place on GE's priority list. However, no project had been terminated and the ATSC planned to accelerate projects to the extent possible within their priority status limitations.

On 6 Sep 1946, Propulsion & Accessories Subdivision Chief Col R.J. Minty wrote the Engineering Division's Technical Control Branch requesting a higher priority for the Compound Engine Project, EO No. 502-479, be assigned a 2B priority. It seems that Charles Hawkins, the one mechanic assigned to the three-cylinder compound engine, would retire on 1 Dec 1946. Col Minty wanted personnel trained, parts completed and testing started before Hawkins left. The project's priority was subsequently raised to 2A.

As early as 1939, P&WA engineering manager Leonard (Luke) S. Hobbs began investigating the idea of transport aircraft powered by gas turbines. In 1940, Hobbs, in conjunction with the Massachusetts Institute of Technology, initiated research into aircraft power plants based on gas turbines. The P&WA PT1 arose from this research. It consisted of a free-piston, two-stroke diesel cycle gas generator whose output drove a gas turbine that was geared to a propeller. P&WA got the PT1 running in 1943 and demonstrated that it had a specific fuel consumption comparable to the best piston radials. In 1944, Hobbs decided to abandon further PT1 development in favor of the R-4360 VDT and axial-flow gas turbine. Still, the PT1 was a fair example of a compounded piston engine, and the project taught the P&WA engineering team much about the complexities of gas turbine design and construction [Connors, Jack. The Engines of Pratt & Whitney: A Technical History (Reston, Virginia: American Institute of Aeronautics and Astronautics (2010) pp 164-172].

None of the USAAF compound engine projects ever entered production. For the most part, they were fundamentally sound concepts that were overtaken by turbojet and turboprop development. However, in Britain, D. Napier and Sons built what was probably the ultimate turbo-compound engine, the Napier Nomad. In the U.S., Curtiss-Wright, on the Navy's dime, developed the R-3350 Turbo-Cyclone, which used three blow-down turbines that drove the crankshaft via fluid couplings to extract as much as 600 additional horsepower from the engine's exhaust. Curtiss-Wright rapidly pushed the engine into civilian service as the Turbo Compound, where it powered the Douglas DC-7 and later Lockheed Constellation models, which ruled the skies until the Boeing 707 was introduced. As such, the Curtiss-Wright compound aircraft engine was the only one to ever enter production.

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.


P&WA PT1 Gas Generator (Author)