The Twentieth Air Force and the B-29s it used to bomb Japan shortened the warby months, perhaps years, and saved, it has been estimated, a million or more U.S. casualties by ending the war before a planned invasion of the heavilydefended Japanese homeland was undertaken — an invasion that was scheduled tobegin November 1, 1945, less than three months after Japan capitulated. Thatinvasion would undoubtedly have taken place had the B-29 program been delayed orhad it and the bombing of Japan not been pushed as fast as humanly possible inspite of the cost in lives and the very difficult odds, choices and problemsthat were encountered.
Throughout the chronicle that follows, I believe that you will be struck bythe number of close calls that the B-29 program itself had — events that couldhave very easily terminated or mortally compromised the program or thecapabilities of the B-29 airplane had it not been for the outstanding courage,foresight and abilities of a relatively few people — both military andcivilian. Two of the very tough choices they made were to put the B-29 intoproduction even before the engineering was completed and to commit them tocombat before developmental testing could get them more fully debugged.
This story about the early development of the XB-29 and particularly about avery important man to that early development — the famous experimental testpilot and Boeing Director of Flight and Research, Edmund T. Allen, who lost hislife trying to make the B-29 combat ready as quickly as possible. Without EddieAllen the B-29 program might never have succeeded. B-29 (42-24579) THE EDDIEALLEN, flown by the 40th Bomb Group was named posthumously in honor of Eddie.
The unsurpassed excellence and ability that Eddie Allen applied to helpingmake the B-29 into the awesome giant that it became can best be appreciated by alook at Eddie’s aeronautical career.
Before we entered World War I, Eddie Allen worked for three years after hisfather died to support his family. He then finished one year at the Universityof Illinois. In 1917 we entered World War I when he was 21. Eddie enlisted inthe Army, learned to fly, became a flight instructor and taught advancedaerobatics. He was sent to the British flight test center in England to learnBritish aircraft flight testing techniques. Before the armistice in November,1918 he returned to the Army’s flight test center at McCook Field to apply hisflight experience and overseas observations. After the armistice he became thefirst test pilot for the National Advisory Committee for Aeronautics —forerunner of today’s NASA. In 1919 he returned to the University of Illinoisfor a year, studied aeronautical engineering for two years at M.I.T. and toppedthat off by entering glider competitions in England and France in a glider hebuilt while at MIT. From 1923 to 1925 he did freelance test piloting and becamea civilian test pilot at McCook Field. From July 1925 to mid-1927 Eddie flewrebuilt WW I DeHavilands as an airmail pilot for the Post Office Department overthe treacherous Rocky Mountain routes between Cheyenne and Salt Lake City —sometimes under extremely adverse conditions. Starting on September 1, 1927 whenthe Post Office Department got out of the flying business, Eddie joined BoeingAir Transport flying Boeing 40-As as an airmail pilot on their new Chicago toSan Francisco run. Over the next five years Eddie began to do more and more testflying — particularly for Boeing Airplane Co., an affiliate of the Boeing AirTransport which later became United Air Lines.
By 1932 Eddie Allen was a recognized, established, highly respected,independent test pilot and consulting aeronautical engineer. In the years thatfollowed his accomplishments became legendary — from landing a Northrup Betawith jammed aileron controls out of a barrel roll to developing the first evereffective cruise control techniques based on some 200 hours of flight testing onthe DC-2 to being a widely published author — mostly on test flying but a hewjust plain old good flying stories — but all with a serious message. He workedfor most if not all of the major aircraft manufacturers at one time or anotherand for Eastern Airlines and Pan American Airways.
For at least some insurance companies, Eddie Allen’s blessing on a newaircraft was a prerequisite to them insuring it. If Eddie was to make the firstflight and do the initial testing on a new design, the insurance premiums wouldbe substantially lower — and the manufacturer could have great confidence thathis creation would come back in one piece. Over the years Eddie made initialflights on over 30 different new models of aircraft. These included the BoeingModel 83 in 1928, the forerunner of the famous F4Bs and P-12s; the Douglas DC-2in 1934; the Sikorsky S43 in 1936; the Boeing XB-15 in 1937; the Boeing B-17B,C, D, E, and F from 1939 to 1942′ the Boeing XPBB-1 and XB-29 in 1942; and theLockheed Model 049 Constellation on January 9, 1943, just 39 days before hisdeath in the number 2 XB-29 crash. That Eddie Allen should be taken by the AirCorps from his vital job at Boeing to make the first flight of the LockheedConstellation is a further testimonial to the high esteem with which he wasregarded.
Between December 31, 1938 and January 20, 1939 Eddie Allen, still as afreelance test pilot, test flew the Boeing 307 33-passenger Stratoliner. Twomonths later on March 18, 1939 on its 19th test flight the 307 crashed killingall on board. Boeing Chief Test Pilot Julius Barr was in the pilot’s seat. Anengineer for a prospective airline customer was in the copilot’s seat. Thatengineer had been pressing Boeing hard to find out what would happen if theairplane were stalled with the #1 and #2 engines throttled and the #3 and #4engines at takeoff power. Boeing refused to demonstrate such a dangerous,unrealistic condition but did agree to approach the condition cautiously. Onecan only speculate as to just what went on in the cockpit and what really causedthe stall, spin, partial recovery, airplane breakup and crash that occurred.Eddie returned to Seattle to testify at the April 3, 1939 CAA Air Safety Boardhearing on the crash. He was there as an expert witness, a highly respected testpilot and the man who had made the first 15 of the test flights prior to theaccident.
While in Seattle for the CAA hearings, Eddie had a conversation with BobMinshall, Boeing Vice President and General Manager. Eddie told Minshall thatcalling in a test pilot to fly a new design after the airplane was built was nolonger a proper approach. Eddie felt that the real need in the aviationcommunity was for exhaustive aeronautical research — both on the ground inlaboratories and wind tunnels and in flight with sophisticated instrumentationand equipment and specialized flight crews. Ground and flight research needed tobe carefully coordinated to compliment each other. The results should becombined with the expertise of the specialized flight crews and engineering testpilots and be applied during the design of any new airplane. Eddie felt thatBoeing was in a unique position to do it. Boeing has the big airplanes needed tocarry all the instrumentation, equipment and specialized flight crews. It hadthe need — its real future was in big airplanes where it already had anenviable background. Minshall liked the concept and so did Claire Egtvedt,Boeing’s President.
The grim reality of the recent Stratoliner accident added emphasis to Eddie’sideas. On April 26, 1939 Edmund T. Allen became Boeing’s first and only Directorof Aerodynamics and Flight Research — a position which he held for almost fourcritical years. It was a fortunate, far reaching event for Boeing and ourcountry. The timing was fortuitous. His beneficial impact on the B-17, B-29 andeven today’s jet fleets would ultimately touch the lives of literally millionsof people — most of whom never knew his name or realized his contributions.This is no exaggeration when one considers the huge B-17 fleets that bombedGermany and the massive B-29 raids on Japan and the lives that were saved andtouched by their parts in bringing WWII to an earlier end. Modern worldwideaerodynamic and flight research which is such a vital part of today’smultifaceted aerospace industries, including modern commercial jet transports,have their roots in and evolve from the ideas that Eddie Allen brought to Boeingin April 1939 and implemented shortly thereafter.
Pan Am B-314
In April 1939 Boeing was many things. It was already a superb designer of thebig airplanes Eddie had referred to such as the B-15 and B-17 bombers, the 314Pan Am Clippers and the 307 Stratoliner. It had only a so-so productionreputation, particularly with the Army whose B-17Bs were behind schedule. The314 deliveries were well underway and the second 307 was nearing flight teststage. Boeing was losing money, was in deep financial trouble and was strugglingto survive. It was a company with people who were courageous, full of vision,imagination, integrity, determination and a dedication to design airplanes thatwere superior and right. Boeing designs were innovative but at the same time,conservative. Boeing would not pursue a poor or even a mediocre design eventhought it might appear to be the politically desirable course.
In late 1938 Boeing has started thinking about a superbomber — an airplanefor which, at that time, there was no established military requirement and nomoney — also, an airplane which no one knew how to build. The Army’s OliverEchols and Bob Olds talked about an airplane with a 5,000 mile range capable ofhitting and enemy aircraft carrier when it was still at least two days offshore.The B-17 could strike a carrier that was only one day out — too close forcomfort. The key to a successful superbomber would be to get the drag way down.Many preliminary design studies were run on numerous configurations includingsuch ideas as new flat liquid cooled engines buried in the wing. Some weretempting but none would really make a good airplane — so the studies werecontinued in an attempt to find an idea the would give the needed breakthrough.
Support for a superbomber spread rapidly. On February 5, 1940 Boeing was oneof several aircraft manufacturers to receive from the Army an invitation to bidon a high altitude, high speed bombardment airplane with a requirement for a5,333 mile range with a 2,000 pound bomb load. A month later Boeing proposed theModel 341 with a gross weight of 85,000 pounds to meet the requirement. Four2,000 horsepower Pratt and Whitney engines would be used. The wing loading wouldbe a whopping 64 pounds per square foot — double what had previously beenconsidered acceptable by the experts. Eddie Allen had convinced the doubtersthat with a very big, properly designed wing flap they could get away with it.The fact that Eddie’s aerodynamics group would have to develop the flap, thatEddie would fly the airplane and that he was confident of success won the day.In addition, extreme measures would be required to further reduce drag as muchas possible. Among the many other things flush rivets and butt joints would berequired and that would add to the manufacturing problems. But confidence washigh that the 341 would be a good airplane.
Several agonizing weeks passed with no word on the superbomber competition.The Air Corps announced that none of the proposals were acceptable. Therequirements had changed as a result of lessons being learned in Europe. Thesuperbomber must have more armament, powered gun turrets, armor plate, selfsealing fuel tanks, higher cabin pressures, a 16,000 pound bomb load capabilityfor shorter flights — and no decrease in performance! A revised proposal wasrequired in 30 days. Back to the drawing board!
The Boeing Model 341 became the Model 345. The gross weight went from 85,000pounds to 112,000 pounds (and later to a maximum overload design gross weight of120,000 pounds). The wing span increased from 124 feet to 141 feet. More powerwas required and the new 2,200 horsepower Wright R-3350 engines would have to beused instead of the 2,000 horsepower Pratt and Whitney. The 2,200 horsepowerWright was an undeveloped engine and there were serious reservations aboutwhether it would be a good engine. Boeing was very uncomfortable about the Model345 — about being pushed too far into unexplored areas. To make things evenworse, there was now serious talk about ordering large production quantitiesbefore an experimental prototype could be built. The risks were becoming veryhigh. Boeing came very close to proposing a smaller airplane with which theywould be more comfortable but which would not be what the Air Corps said wasrequired. On the other hand, the war was spreading rapidly in Europe andthreatened to spread much further. The expanded superbomber requirements of theAir Corps might very well prove necessary even thought the technological riskswere very high. After careful soul searching with the war in mind, Boeinguncharacteristically decided to submit the Model 345 configuration with a stringdetermination to do everything possible to make it successful. The Model 345proposal was submitted on May 11, 1940. Within weeks the Air Corps told Boeingthey were issuing a contract for engineering, wind tunnel and a mock-up of theModel 345 which would be the B-29. Furthermore, production contracts for perhaps200 B-29s would be let long before an experimental prototype could be flown.Clearly the Air Corps has joined Boeing in a desperate gamble on the success ofthe Model 345 design. After Paris fell in June 14, 1940 Congress was asked formoney for 990 B-29s. The ante had just been raised!
On September 6, 1940 a formal contract for two XB-29s was released.Engineering studies which had started with only a few people in late 1938 hadnow grown into a full scale production design effort which would require1,433,026 engineering manhours before the first XB-29 would fly. Eddie Allen andGeorge Shairer were kept busy with literally hundreds of wind tunnel and flightresearch investigations to everlastingly reduce drag and to geed the designproject the necessary aerodynamic and flight test data to permit the design tomove forward as rapidly as possible. A PT-19 experimental wing flight testprogram and three specially configured B-17s conducted flight tests of manydifferent configurations of developmental items for the B-29 such as propellers,cowling, turbo superchargers, empennage, rudder, elevators, ailerons and flaps.These tests helped to find the best configurations and to optimize such thingsas control forces and control balance and to reduce the technical risks.
Additional growth during the B-29 design phase increased the design maximumoverload gross weight to 120,000 pounds and the corresponding wing loading to 69pounds per square foot.
There were to be two more serious "wing loading crises" long beforethe XB-29 ever got off the ground. The first was when a new Air Corps"Plane X" with only a 53 pound per square foot wing loading was a"dog" to fly and in addition would not get above 28,000 feet. The AirCorps intently questioned Boeing about the wing loading on the B-29. The secondand even more serious crisis was when a respected aircraft manufacturer’sengineers reviewed Boeing data and told the Air Corps that Boeing was very wrongin its predicted B-29 performance. They said that the B-29 would be 40 mphslower, would have a 5,000 foot lower ceiling and would have 1,000 miles lessrange that Boeing had predicted. In the face of such criticism, it took realcourage and confidence on the part of Boeing and Air Corps principals involvednot to increase the B-29 wing area to substantially reduce the wing loading fromthe planned 69 pounds per square foot — a step which Boeing firmly maintainedwould be catastrophic for performance and, by then, to production schedules.Again it was Eddie Allen’s and George Shairer’s work that was being challengedand who needed to defend their positions of the really had confidence in thepredictions. The price for being wrong either way would have been catastrophicto the B-29 successes of the Twentieth Air Force. They had the courage of theirconvictions and commanded sufficient respect to convince their inquisitors thatthey were right and to continue the rapidly expanding B-29 program withoutchange. Again a catastrophe was averted. It is interesting to note that incombat the B-29s were frequently successfully flown at 140,000 pounds per squarefoot!
While the many B-29 problems were being addressed, Eddie Allen had hadanother extremely important task to accomplish. That was to build the kind offlight research operation that he had outlined to Bob Minshall in his office inearly April 1939. At that time no one realized how crucial it would be to theall out war effort that was to come.
In the following years Eddie built a sophisticated Boeing flight researchcapability that was second to none. His basic purpose was to safely,economically and quickly obtain and disseminate accurate, quantitative flighttest data. This would help find, develop and prove the best possibleconfigurations from the perhaps hundreds of candidates. The data would be usedto determine the safety of the article being tested, the degree to which it metits guarantees and requirements, its adequacy for the purpose intended, areasneeding improvement, ways of improving the existing article or making the nextdesign as good as possible, and finally, the best way of operating the equipmentin service. To accomplish these goals he hired the best people he could get withas close to the qualifications he wanted and trained and developed them into theskillfully expert team that was required to accomplish his vision. Most of theflight crew members and a high percent of the flight test department groundpersonnel were engineers. Each had weeks of formal, structured classroomtraining tailored to specific assignments. There was "hands on"training in the altitude chamber and in the appropriate airplanes with test andsafety equipment. There was periodic recurring training as necessary to maintainthe highest possible skill level to minimize personnel risks and to obtain highquality data.
[It should be noted here in passing that there were three completely separateflight test groups at Boeing with entirely difference people reporting throughdifferent organizational lines. One was Production Flight Test which wasresponsible for flying every new production airplane to make sure there were nomanufacturing or quality control problems and to make any necessary adjustmentsbefore turning the airplane over to the customer for acceptance. Another was thecustomer flight acceptance group. In the case of B-17s and B-29s they were AirCorps officers who flew and accepted the airplanes for the Army. The third wasthe Research Flight Test Department which was Eddie Allen’s creation and isreferred throughout this discussion. It conducted engineering, experimental andresearch flight testing. In the purest sense those were really three differentkinds of flight testing what were all conducted in Eddie Allen’s department. Forthe most part all three were done by the same people using the same methodsalthough it was recognized that certain tests in any of the categories mightpotentially require specialized or exceptional skills and this warrant selectivepicking of specific flight crew members. First flights on new models ofairplanes fell in this category. Because of the similarities in methods andcrews and the fact that some flights might involve engineering, experimental andresearch testing, the three terms are often used interchangeable with somethingless than precise regard for the differences.]
Eddie Allen’s Flight research Department was under Al Reed, Chief of FlightTest and Chief Test Pilot. It was organized into functional groups such as:pilots and copilots; the other specialized flight crew members, for the mostpart flight test engineers; the instrumentation group who was responsible forobtaining or designing and making, calibrating, installing, servicing andmaintaining the vast amounts of standard and specialized instrumentation andphotographic equipment required to measure and record the many variables thatneeded to be measured; the analysis group who transcribed, corrected withcalibration data, plotted or tabulated the corrected data, analyzed the resultsand prepared the final reports for distribution; the liaison group who workedwith the mechanics and shops to make sure that the airplane configuration andinstrumentation were in accordance with requirements established by the ProjectFlight Test Engineer in charge of each test airplane’ a flight equipment groupto service, store and maintain items such as parachutes, oxygen masks, bailoutand walk-around oxygen bottles, etc.: and an administrative support group.
Prior to each flight the airplane was prepared to conform with the veryspecific written test and configuration requirements. A very detailed, specificPlan of Test setting forth each test condition was prepared for each testflight, given to each of some 10 flight crew members and then gone over indetail in the pre-flight conference so everyone knew exactly what to expect andwhat was expected of him in flight.
During normal flight the basic flight crew performed their duties in theconventional manner. During the flight test phases the Project Flight TestEngineer would be slightly aft of and between the pilot and copilot to providethe best possible communication and awareness between those three people and, inthe case of the B-29, the flight engineer as well. A normal flight test crew onB-17s and B-29s consisted of about 10 people. The additional people manned thespecial instrumentation and equipment involved in the test. All had interphonecontact. A typical instrumentation load might include two photorecorders with 40or 50 instruments and a camera in each; two or three manometer boards to record40 or 50 pressures; one or two potentiometers to record 50 to 100 temperatures;a Brown recorder that could be selectively set for continuous recording of anyone of many different potentially critical temperatures; and perhaps anoscillograph to record strain gage or vibration data on structural demonstrationor flutter flight tests. Large bundles of wires or tubing connected eachinstrument with the appropriate transmitters on propellers, engines, nacelles,wings, control surfaces, etc. Manual and photographic recording of the data wasroutine. Frequency of recording depended upon the requirements of the testcondition that was set up. For instance, automatic recording once a second forperhaps three minutes during a stabilized performance condition was common —and produced a lot of performance and cooling data to be analyzed. The ProjectFlight Test Engineer coordinated the activities of the entire crew, kept amaster log of events and set the appropriate recording frequency of all camerasfrom his master control. At every recording station there was a coordinationlight and a coordination counter that clicked over once a second that providedprecise coordination of all manual and photographic data from before takeoff toafter landing. Typically one pilot would concentrate entirely on flying theairplane to precisely stabilize and maintain the planned flight condition. Theother pilot would set up the engine power, set cowl or wing flap positions,maybe operate special equipment such as an engine water injection system andmonitor everything going on inside and outside the airplane to be able toanticipate and react immediately to cope with any emergency. In the XB-29 theflight engineer helped with particularly the powerplant related tasks.
Immediately after every flight there was a highly structured but prettyinformal post-flight conference that was recorded verbatim in its entirety by acourt type stenotypist. The conference was attended by the entire flight crewand any key ground personnel who had a direct interest in the flight. Thesecould include: design project and staff engineers who had requested specifictest conditions and who might have to design the corrections or requestadditional tests based on problems encountered and the data obtained; technicaland management representatives from outside suppliers whose components werebeing tested such as engines, propellers, carburetors, accessories, brakes,armament; flight test instrumentation engineers who wanted to know how theirinstrumentation worked and what they needed to do before the next flight; thedata analysis supervisor whose people would have to take the vast amounts ofmanual and recorded data and sort out what should be processed’ the shop foremanand quality control supervisor who wanted to know of any airplane problems andany special actions needed from them for the next flight; customerrepresentatives, usually at least an Air Corps quality control supervisor; andfor particularly important flights perhaps high level company and customermanagement. A post-flight conference might have as few as a dozen or as many as40 or more people. It might last for only five minutes or as long as a couple ofhours. The Project flight Test Engineer or perhaps the Project Test Pilot waschairman. The short items were usually disposed of first so most of the peoplecould leave and get back to their work. The test conditions that were run wereeach reviewed using the Plan of Test as the agenda. Any unusual evens werenoted. Any clarifying questions were asked and answered while circumstances werestill clearly remembered. Plans for the next flight were tentatively made beforethe conference adjourned.
Before their day was over the stenotypists would have transcribed theirverbatim recording of the post-flight conference so that it could be distributedthe next morning to all those with a need to know. The Project Flight TestEngineer would make every effort to complete and distribute his "Report ofTest" also on the following day. It was a written summary of the testflight conditions run along with his log sheet. The system was not allowed toget bogged down. Flight test data promptly got to those who needed it.
Eddie Allen said "Flight Testing is a Sound Business" and wrote apaper proving it. It is also an expensive business usually involving heavilyinstrumented airplanes that would he hard, expensive, and time consuming toreplace. Sometimes they are one of a kind. With the highly organized, structuredapproach that Eddie developed, risks, costs, and time were minimized whileresults and accuracy were maximized.
Eddie’s drive to make airplanes as safe as possible extended to the specialneeds of military aircraft. He and Boeing worked particularly hard to design andbuild combat damage tolerant aircraft. It allowed many B-17 and B-29 crews toget to safety in spite of extreme combat damage.
When Pearl Harbor hit on December 7, 1941, Eddie Allen had his organizationset up as described. It was operating smoothly and he was in the process ofexpanding it. Including Eddie, there were only four pilots at that time doingengineering flight testing at Boeing. In early January 1942 I started workingfor Eddie along with six other new copilots and twice that many new flightengineers. We were moved quickly through school and the formal training programand acquired B-17 experience with the production acceptance crews. Most of ourengineering flight test efforts in 1942 were spent trying to find out how tomake crews safer and allow them to operate more efficiently while unpressurizedat altitudes to and a little above 35,000 feet. Between April and the end of1942 I flew a number of times with Eddie as his copilot on the B-17 and XPBB-1twin engine flying boat.
When I first met Eddie in January 1942 I was surprised. Although I had nopreconceived ideas, I did not expect the world renowned test pilot to be of soslight a build and so unassuming. He weighed about 145 pounds and was about5’8" tall. In those moments he fit better my image of a naturally friendly,soft spoken, mild mannered midwestern farmer. It was hard then to visualize himskillfully controlling the sometimes huge, sometimes balky airplanes he hadtested.
As I got to know Eddie better of the subsequent 14 months, I came to havevery great respect, admiration and affections for him. I have never heard anyonesay an unkind word about Eddie Allen. On the contrary, there have been many verycomplimentary words used to describe Eddie. The include: calm, competent,skillful, precise, earnest, ingenious, courageous, intensely curious, dedicated,sincere, pleasant, congenial, gentlemanly, retiring, friendly, unassuming,generous — and the list goes on — and I’ll bet there is at least one story oract of Eddie’s to fit each word. No wonder his people were so dedicated to him.In spite of his great personal ability, he let me, his copilot, do most of theflying when I was with him. He was a kind but precise teacher. I learned a lotfrom him in flight and on the ground. He made me feel that he had great faith inme. I believe he was the same with most of us. It made one determined to doeverything possible to justify that faith and confidence. Although he ran atight ship with highly structured procedures, I don’t ever remember feelingresentment or rebellion against the discipline — perhaps because it seemed soright, so logical, so proper. He was a great team leader and a tremendousinspiration to us.
Eddie was a conservative test pilot — not prone to take chances. Heunderstood his limitations and those of the equipment he was testing. He did notlike the then common Hollywood depiction of a test pilot as a brash, wild,flamboyant daredevil. He felt keenly responsible for protecting the hugeinvestment that an experimental airplane and its crew represented. He said thathe was afraid to take risks! He felt that fear is healthy, whereas panic isdebilitating. He stayed cool under pressure.
This, then was the man and the organization he had built which was to begintesting of the first XB-29 in September 1942.
There had been many tough decisions made and significant risks taken in theshort three years since the Model 341 concept had sparked real hope for asuperbomber. It was only two very compressed years since the contract for twoXB-29s had been signed. Now, after investing more than 1,400,000 engineeringmanhours in the XB-29, flight testing was bout to begin: flight testing whichwould prove whether Eddie Allen and George Shairer had been right in their manydecisions including defending a 69 pound per square foot wing loading; whetherthe Air Corps had been right in building two new plants, in starting B-29production by Boeing, Bell, and Martin in four plants in Wichita, Renton ,Marietta, and Omaha, and in already ordering 764 B-29s before the first XB-29ever flew; and whether the thousands of other decisions that had been made wereright.
Everyone had been under tremendous pressures and time had now allowed as muchpre-flight development testing as most would have liked. For instance, theengines which Eddie was about to fly with had been cleared for only 35 hours!The need for flight test answers was enormous.
The pressure was really on Eddie Allen and still he "kept his cool"in spite of it all. Eddie estimated that with an all out effort it should takefive months and 200 flying hours to do a reasonable job of shaking down theXB-29, determining its capabilities and getting the minimum performance andoperating the data the Air Corps needed to start training and place theforthcoming production airplanes in service.
Taxi tests and a couple of very short hops were made by Eddie on therelatively short 5,200 foot runway at Boeing Field in Seattle in the first partof September 1942. Although there were some system problems, Eddie felt thatmeaningful testing in flight could be conducted while solutions to theindentified problems were being worked on.
On September 21, 1942 the first XB-29 flew for the first time, and Al Reed asEddie’s copilot. Eddie climbed to 6,000 feet and checked lateral, directionaland longitudinal stability and control. He checked controlability and generalperformance with the #1 engine throttled. Power off stalls were checked. Controlresponse, forces and effectiveness were noted. Everything that should be checkedon a first flight was satisfactorily accomplished in the 1:15 flight. It was apretty uneventful flight and first indications were certainly favorable. ButEddie and the others knew there was a great deal of work ahead.
There were to be very few additional uneventful flights. The troubles startedadding up. By December 28 Eddie had been able to make only 23 flights in 27hours of flying. There had been 16 engine changes, 22 carburetor changes and 19exhaust system revisions in those three months. In addition there were propellergoverning and feathering difficulties, runaway engines that oversped to 3,600RPM and a host of lesser problems. The longest flight was 2:19 The averageflight was only 1:10 long. It was almost impossible to get much meaningfulquantitative data when flights were that short — particularly when much of thetime was spent fighting the problems and getting back to the field. One of thefew bright spots was that the aerodynamics of the airplane seemed to be justwhat Eddie and George had worked so hard to achieve. Later testing would confirmthat early assessment. Performance and handling qualities were excellent. Nosignificant aerodynamic changes were ever made except for research work on therudder which resulted in being able to simplify and improve the airplane byeliminating the rudder boost. Eddie and George were vindicated.
The flight on December 28 was intended to check the service ceiling and setperformance data. The #1 engine failed at 6,800 feet and the flight wasterminated after 26 minutes. Ground inspection of the #2 engine showed metalchips in the sump — it, too, was about to fail. That was the last time EddieAllen or Al Reed would fly the #1 XB-29. Subsequent events kept the airplanegrounded for more than seven months — until August.
There is an interesting personal sidelight to that December 28th flight. Sixdays before on December 22 I flew with Eddie as his copilot on the 62,000 poundXPBB-1 twin engine flying boat. It turned out to be the last time Eddie flew theXPBB-1 and also the last time I ever flew with Eddie. The purpose of the flightwas to complete a few tests prior to flying final demonstration for the Navy ina few days. Eddie let me fly the airplane including the required power offlanding which called for cutting the ignition on both engines at 1,000 feet. Thehigh drag boat came down like a brick! It was the first time I had ever donethat! Eddie just sat there and watched. Fortunately it was a good landing inspite of the extremely steep glide path. I had no inkling that that would be mycheckout flight (and I doubt that Eddie did) until the morning of December 28th,when Eddie came to my desk and very casually asked me to fly final demonstrationfor the Navy that day on the XPbb-1 because he and Al Reed needed to fly theXB-29. I managed, in my amazement, to stammer something like "I’d be gladto." The demonstration went well and I have always been extremely gratefulto Eddie for giving me that opportunity and for placing that much trust andconfidence in me. Incidentally, he let me have the fun of making the deliveryflight to the San Point Naval Air Station two weeks later.
On December 30th the #2 XB-29 (AAF41-003) was ready for this initial flight.It too had engines that were cleared for only 35 hours in positions 1, 3 and 4.It was to be a thorough functional check of the airplane and its extensiveinstrumentation The weather was marginal. The functional check proceedednormally until the #4 propeller would not feather and governing was erratic.Eddie elected to discontinue the flight and immediately headed hack to BoeingField at which time he was advised that the weather was deteriorating rapidly.About 6 minutes out, the #4 engine caught on fire, the propeller oversped to3,500 RPM, the propeller would not feather and smoke, sparks and flame werecoming from the exhausts. Shutting off the fuel and the use of fireextinguishers were ineffective. The fire continued to get worse. About 2 minutesout the fire was burning fiercely in the accessory compartment. Flames werepouring from the nacelle access door and from the intercooler exit area. Heavysmoke and long fingers of flame were trailing off the wing. In the meantimeheavy smoke was pouring from the bomb bay into the cabin making it increasinglydifficult to see or breathe. Eddie landed downwind, choking, partially blinded,on the 5,200 foot long, 200 foot wide runway. The intense fire was put out byfire equipment on the ground. Eddie later received the Air Medal for his skilland bravery during that harrowing 32 minute flight. Ground inspection showedmore trouble. A fire had just started in engine #1 and engine #3 was close tofailure, too. Those three 35-hour engines each had less than three hours totalground and flight time. Because of engine shortages, two of the three engineshad to be replaced with engines cannibalized from the XB-29 which was laid upfor some modifications. In addition the fire in #4 had been so severe that the#4 nacelle had to be replaced with the #4 nacelle also cannibalized from the #1XB-29. At least the #2 XB-29 now had 4 so-called "unlimited" engines.
Unfortunately, engine/nacelle fires similar to the #4 fire continued tooccasionally haunt production B-29s and caused at least 19 serious B-29accidents between February 1943 and September 1944. While Boeing and Wrighttried hard to find and correct the cause or causes, there was a natural tendencyfor each to blame the other. It was 15 months before there was positive proofthat the R-3350 was susceptible to induction system fires which could veryrapidly get out of hand and become uncontrollable magnesium fires which thendestroyed evidence of the fire’s origin. That proof came on March 24, 1944 whenI had an induction system fire on the #4 engine during a routine test flight onthe #1 XB-29. I was fortunate enough to get the engine feathered and the fireout before it broke out of the blower section or the intake pipes and became anexternal fire. The partially burned magnesium impeller and interior of theblower case were irrefutable evidence. In the face of that evidence Wrightdeveloped the fuel injection system to eliminate the potential for inductionsystem fires.
It was almost a month before the #2 XB-29 flew again on January 29, 1943. Inthe next three weeks emphasis was on engine, propeller, governing, and airplaneperformance testing. Catastrophic engine failures eased up but that was aboutall. During descent for landing on February 2 there was a strong odor ofgasoline emanating from the bomb bay into the cabin. A thorough inspectionuncovered nothing conclusive. On a flight on February 17th there was a bad fuelleak over the wing from the #4 fuel filler cap. The leaking cap was fixed.
By February 17, 1943 the #2 XB-29 had made eight flights totalling 7:27 hours— an average of only 56 minutes per flight. In the five months since the firstXB-29 flight on September 21, there had been only 31 flights totalling 34:27 —a long way from what Eddie had estimated in September could be done. And withand overall average flight time of only 1:07 the amount of meaningful test datawas pretty sparse from that meager 34:27. As hard as everyone was working tosolve the problems the answers were coming painfully slowly. As Eddie and isProject Flight Test Engineer left the airplane that afternoon and walked acrossthe ramp to the post-flight conference, Eddie expressed to him the gravereservations he had about continuing flight testing until at least the moreserious of the XB-29 problems could be fixed. Unfortunately, the fastest, andmaybe the only way to fix some of them was to try out the various fixes inflight — the "try, try, try again" approach that had been sosuccessfully used by Eddie and George Schairer over the years. But now Eddiefaced a real dilemma. The B-29 was potentially a fine airplane. It was urgentlyneeded in the Pacific. It was committed to production — 1,600 B-29s were now onorder at four separate plants. Flight test was way behind its expected scheduleand the data was badly needed to: prove the airplane; quickly find and correctthe problems; minimize production disruptions’ develop training and operatingprocedures and manuals. But it was currently a dangerous airplane. Majorimprovements were badly needed. Temporary grounding would be normal, prudentthing to do. But they were not normal times. The sooner the B-29 could be usedin combat, the sooner the was would end and the sooner the casualties andcarnage would stop. Eddie concluded that he must continue flight testing asrapidly as possible. His entire crew had also to know the risks — to a man theystayed with him.
The primary objectives of the February 18, 1943 flight were to measure climband level flight performance and get engine cooling data with 4 and 2 enginesoperating. Maximum altitude would be limited to 25,000 feet because of theexcessive trouble that had been encountered with low engine nose oil pressuresabove that altitude. The effectiveness of fixes for some of the past problemswould also be evaluated. Takeoff would be at the normal design gross weight of105,000 pounds with full fuel tanks — 5,410 gallons of gasoline.
Eight minutes after that 12:09 PM takeoff to the south, while climbingthrough 5,000 feet with rated power, a fire was reported in the #1 engine.Mixture and fuel to #1 were cut off, propeller was feathered, cowl flaps wereclosed, a CO2 fire extinguisher bottle was discharged and a descent and returnto Boeing Field was initiated. Since the fire appeared to have been put out andeverything seemed under control, Eddie elected to make a normal landing patternand land form the north on runway 13 (128 magnetic) to the SSE into the 5 MPHwind rather than making a downwind laning on the 5,200 foot runway with a heavyairplane. At 12:24 PM the radio operator routinely reported altitude at 1,500feet at a point 4 miles NE of the field. They were on the downwind leg, headedNNW and starting a left turn onto base leg. No one suspected the drastic changethat would take place in the next 2 minutes. At 12:25 they had just completedturning onto base leg, had just crossed the heavily populated west shore of LakeWashington about 5 miles NNE of the field, were at about 1,200 feet altitude andwere heading SW approaching the commercial and industrial south side of downtownSeattle. At that point ground witnesses heard an explosion that sounded like aloud backfire and a piece of metal fell from the airplane. About that time theradio operator, who could see into the forward bomb bay and the wing centersection front spar, was overheard by the Boeing tower on an open microphone tosay "Allen, better get this thing down in a hurry. The wing spar is burningbadly." He told Boeing Radio on a different frequency "Have fireequipment ready. Am coming in with a wing on fire." About a mile down theflight part from the explosion, burned parts of a deicer valve, hose clamps, andinstrumentation tubing were later found. The had come from an area normallyinside the wing leading edge, ahead of the front spar, and just outboard of the#2 nacelle near the #2 fuel tank filler neck which was rubber like the selfsealing fuel cell. The airplane now turned south on an oblique final approach ina desperate effort to reach Boeing Field just 4 miles away. Eddie was about 250feet high and ground witnesses later reported that part of the wing leading edgebetween #1 and #2 engines was missing. In the next mile the flight engineer’sdata sheet was found and three of the forward compartment crew members left theairplane — too low for their parachutes to open. At 12:26 PM, only 3 miles fromBoeing Field, the #2 XB-29 crashed into the Frye Meat Packing Plant killingpilots Eddie Allen, Bob Dansfield, and the other 6 crew members on board. Thecrash and resulting fire killed an additional 20 people on the ground anddestroyed much of the airplane and the plant. There was clear evidence that fireand dense smoke had gone through the bomb bay into the cockpit in the lastmoments before impact. Burns on the bodies and clothing of the 3 crew membersbailed out just before impact were a part of that evidence. Eddie Allen and hiscrew died service their country the best way they knew how. In 1 minute the firehad gone from undetectable to catastrophic.
At 12:26 PM on the 18th day of February, 1943, the saga of Eddie Allen ended.However, not so his legacy which has continued to this day to benefit his fellowmen for whom he always showed such great respect.
The scientific flight testing methods which Eddie Allen developed continuedto serve his country well throughout the war. And they have continued to thisday to evolve and improve and keep pace with technology and to serve man—justas Eddie Allen would have wanted.
The flight test team that Eddie has assembled and trained was decimated,devastated and demoralized. Some of its members would probably never completelyget over his loss — but they did put the pieces back together and continued to"fight the battles" and get the answers as Eddie would expect them to.
On April 23, 1946, three years after Eddie Allen’s death in the February 18thcrash of #2 XB-29, he was posthumously awarded the Air Medal — an honor rarelybestowed upon a civilian — by direction of the President of the United States.
The medal was presented to Florence Allen Howard, Eddie’s widow, by MajorGeneral Benjamin T. Chidlaw, Deputy Commander for Engineering for the AirMaterial Command at Wright Field, during ceremonies at the Boeing Plant #2. AtMrs. Howard’s request, General Chidlaw pinned the medal on Turney Allen, the sixyear old daughter of the late pilot.
The Citation reads:
"To Mr. Edmund T. Allen, Civilian Test Pilot, for meritorious achievement in aerial flight on December 30, 1942. On this occasion while piloting an Army Air Force XB-29 type aircraft under extremely unfavorable flying conditions, an uncontrollable fire developed in the number four engine. In spite of the fact that he would have been justified in abandoning the airplane under such conditions, Mr. Allen elected to remain at the controls and attempt to safely land it. As a result of his skill and daring invaluable test data and a prototype airplane were saved, the loss of which would have immeasurably retarded the entire B-29 Program at a crucial time in its development."
It is signed by President Harry S. Truman.
"In the course of a great war such as we have only recently concluded, there are a great many unsung heroes — men who labor and work in relative obscurity while others garner the laurels of combat accomplishments. Of course, the men who flew the planes in combat and met the enemy on his own ground deserve the plaudits which have been accorded them. But in the air war there were other men whose work and without whose sacrifice it would not have been possible to get into combat the planes that finally won the war.
"Especially this was true in the case of the aircraft test pilots — the men who took the planes in their experimental stages, tested their potentialities, ironed out their defects and brought in the reports that made it possible to fashion these airplanes into formidable weapons of war. Theirs was the contribution of a scientific objectivity combined with the daring and fearlessness of the pioneer, and the contribution was a magnificent one. They have earned the admiration and the respect of the men who flew the planes that grew out of their efforts and accomplishments and, as a matter of fact, they were really a part of the great Air Force team that bombed the enemy to defeat.
"Eddie Allen was outstanding among these men."