Liberty Engine History

8/8/2019 Liberty Engine History

1/124

SMITHSONIAN ANNALS OF FLIGHT

Volume 1 Number 3NATIONAL AIR AND SPACE MUSEUM SMITHSONIAN INSTITUTION


2/124


3/124


4/124

Grover C. Loening and Smithsonian specimen Loening OA-1A Amphib ian "SanFrancisco" of 1926, powered by inverted L-12 Libe rty Engine. Photo take n in 1967.


5/124

S M I T H S O N I A N A N N A L S O F F L I G H TVOLUME 1 NUMBER 3

THE LIBERTY ENGINE1918-1942PHILIP S. DICKEY III

Lieutenant Colonelj United States Air Force (Retired*)

SMITHSONIAN INSTITUTION PRESS NATIONAL AIR AND SPACE MUSEUMWASHINGTON, D.C. 1968


6/124

Second Printing, 1970Third Print ing, 1973Fourth Print ing, 1977Fifth Printing, 1978

For sale by the Superintendent of Documents, U.S. Government Printing OfficeWashington, D.C. 20402 - PriceStock Number 047-005-00003-5


7/124

ContentsPage

F O R E W O R D vnP R E F A C E ixA N T E B E L L U M 1M I S S I O N S A N D C O M M I S S IO N S 5B I R T H O F A C O N C E P T 7T o D E S I G N A N E N G I N E 1 0T H E E N G I N E S 1 9G R O U N D T E S T S 3 3

F L I G H T T E S T S 4 2T H E C R I T I C S 4 5O R G A N I Z A T I O N F O R P R O D U C T I O N 5 5T H E P R O D U C E R S 6 1T H E L I B E R T Y G O E S T O W A R 6 8P O S T B E L L U M 7 3A P P E N D I X 83

1. O r g a n i z a t i o n f or P r o d u c t i o n 8 42 . P r o d u c t i o n S u m m a r y 9 13 . S t a t i s t i c a l C h a r t s 1 0 0

B I B L I O G R A P H Y 1 0 9


8/124


9/124

F o r e w o r dIn this third number of the Smithsonian Annals of Flight, Philip S. Dickeytel ls the history of America 's remarkable contr ibution to ear ly aviat ion,the Lib erty engine, the or iginal mode l of which is now in the Sm ithsonianInst i tu t ion ' s Nat ional Air and Space Museum.In delineating the history of the engine from its brilliant conceptionthrough i ts role in World War I to i ts f inal production and use in WorldWar II and thereaf ter , the author draws on a wealth of information contained in let ters, memoranda, reports , and personal accounts, most ofwhich have hither to been unpublished. His accounts are reinforced bystatistical tables, charts, detailed photographs, specifications and detailsof the engine 's performance. S. P A U L J O H N S T O N , Director

National Air and Space Museum2 Ja nu ar y 1968

Vl l


10/124


11/124

PrefaceT he engines ab out which this pap er is writ ten are called "L ib ert y."They came in series of 4, 6, 8, and 12 cylinders and were developed forinterchangeabil i ty of par ts and ease of production. Without doubt, theLiberty engine was America 's greatest contr ibution to the All ied cause

in Wor ld War I and to Amer ican av ia t ion dur ing the postwar per iod .Designed in six days during late May and ear ly June of 1917, the enginewas still in active use by the Army Air Service in 1936.But the name was not always "Liberty." I t was or iginally the U.S.A.Standardized Aircraf t engine, but , when the name "Liberty" was suggestedby Adm iral D . W . Tay lor in the ear ly par t of the per iod of prod uctio n, i timmedia te ly caught on , 1 and, though there was already a Liberty truckon the market , 2 no other name would do for the new engine.The Liberty became a legend in i ts own t ime and was known even bysome mem bers of a new gene ration of f ighting m en in W orld W a r II .The author, in fact, in 1945 tested a 45-foot Sparkman and Stevens air-corps rescue boat powered by two 500 hp Vimaler t conversions of thefamous Liberty engine.The name "Liberty ," as associated with the engine, was given thedist inction of being registered as the trademark of the U.S.A. Standardizedengines by the United States Government on 17 J u n e 1919, under cer t if icateof regis tration nu m be r 125,853. Th is is the first instance of the U ni te dStates Government register ing a trademark under i ts own laws or the lawsof any other country. 31 History of the Bureau of Aircraft P roduc t ion (M S, T he U ni ted States Air Forc e

M use um , W right-P at terso n A ir Force Base, Oh io) , vol . 7 , p . 1904. Ci ted hereaf ter asHistory of BAP.

2 M cC ook Field d ecima l files (MS , Air Force R ecords Repo si tory , S t . Louis , M o. ) ,452.8 , le t ter , Em mo ns f rom Heaslet , 20 No vem ber 1917. Ci ted hereaf ter as McC ook f iles .(McCook Field no longer exis ts . Located, i n Nor th Day ton , Oh io , du r ing Wor ld War I , i twas the hear t of the Un i ted States ' a i rcraf t - and eng ine-develo pme nt ef for t.3 History of BAP, vol. 7, p. 1904.

ix


12/124

The following examination of the Liberty engine is not intended to bea technical description of the inner workings of an internal combustionengine. It ran the same way a car engine runs. There will be some referenceto pistons, horsepower, battery ignition versus magneto ignition, and likesubjects, but only to put into proper perspective some of the controversiestha t revolved a roun d the engine during i ts testing and prod uction period.There will also be some discussion of the problems of inverting, air-cooling, an d gearin g the engin e. The se are bro ugh t out to show the versatility of design. The postwar story shows the durability of design andtends to prove that the engine was far ahead of its time.If i t appears that Jesse G. Vincent predominates in the references, theappearance is not deceptive, for Vincent was the engine 's codesigner ,act ively par t icip ating in its test ing, modif icat ion, an d prod uctio n unti l heleft the A ir Service after the A rm istic e. A nonflyer, he entered service,became an expert aviator , and, as commander of McCook Field, f lew manyhours behind the Liberty engine. As.an expert engineer and an inventivegenius, he was able to exam ine the operation of the engine on the dyn am om eter, the test stand, and in the air. If any one man can be granted primarycredit for the success of the Liberty, he is Vincent. He was a prolific andlucid writer, and the McCook Field files during his tenure are filled withhis correspondence.E. J. Hall, the engine's codesigner, was moved from the Liberty projectto troubleshoot the D H - 4 production ear ly in 1918. I t is regrettab le tha tso little evidence is left of his role in the Liberty project.


13/124

Ante BellumThis is the history of an engine, but not just a description of the nutsand bolts that held i t together or of the juxtaposit ion of com pon ent par t sthat caused i t to run and produce power. The engine, the pr incipal character in the story, can be properly considered and known only in thecontext of the times and the people and the circumstances that led to itsbir th . The Liberty engine was a child of war , and i t s tands today as amonument to the capacity of a nation to change and, in changing, to movea giant step beyond its peers in the development of an infant technology.The Liberty was an apology for our nation's lack of foresight and for itsfai lure to develop the aircraf t born at Kit ty Hawk.In the ear ly days of March 1917 the United States stood uneasily on thethreshold of a conflict not of its choosing, soon to learn a harsh lesson ineducation for world leadership, begun in the closing years of the lastcen tury . Am erica , the beautiful, was isolated by the high seas from the

lunacy in Europe; America, complacent in the self-hugging warmth of i tsparochialism, was, in time, to become a disorganized mass of patriotismtrying desperately to focus its massive energy on production for war.Fortunately, the nation had developed and prospered pr ior to WorldWar I, partly because of an innate ability to organize and a willingness totry something new if given the prop er im petus. W ith the sinking of Vigilancia,State of Illinois, an d City of Memphis repor ted on 18 M arc h 1917, and thesubsequent declarat ion of war on 6 April , the magnitude of the task ofwar production struck the nation 's leaders with stunning force, for thecountry was to ta l ly unprepared .

ThetUnited States ' lack of preparation for World War I was across theboard; the segment of interest for this story, however, is aviation. It shouldbe remembered that the achievements of industry in the production ofaircraft and of aircraft engines were accomplished against a backdrop ofuniversal competi t ion for the manpower, equipment, supplies, and resourcesrequired to mobil ize the nation for par t icipation in the European War.Before considering the record made by the United States in the designand production of the Liberty engine, the page must be turned back to

1


14/124

show the development of the aviat ion-engine industry in the United Statesand i ts condit ion at the t ime war was declared.Un ti l 14 April 1917, the total production of aircraft in the United Stateshad been 666 machines, none of a more complicated type than trainingplanes built for the British Royal Flying Corps. 1 Soon af ter the UnitedStates entered World War I , the National Advisory Committee for Aeronautics requested capabil i ty studies f rom all aircraf t companies. Theanswers showed a disheartening lack of the immense production potentialwhich would be required to place America's name at the forefront of theaviat ion powers of the world. The government had done nothing to encourage industry or individuals to enter the aircraft or aircraft-engineproduction f ield . The government incentives that character ized the development of our early technology were not applied to the furtherance of theaviat ion industry in the nation that invented the f irst heavier- than-airmachine to lift a man from the ground in sustained, controlled, poweredfl ight . The aircraf t developmental work the Wright Brothers carr ied outalmost in secrecy on the Huffman Plain near Dayton, Ohio, was acceptedand continued with great energy and enthusiasm in Europe. I t was treatedas a curiosity, a passing fancy in the arm ed services of the cou ntry of its bir th.

At the beginning of 1917 the United States ranked 14th among thenation s of the w orld in aviat ion. Th e f irst Congressional app rop riat io nspecifically for aviation was $125,000 on 3 March 1911. From 1908 to 1916appropriations for aviation had totaled only $930,000.2 In the same perioda total of only 59 aircraft was delivered to the United States Army. During1916, 83 more were delivered, and, in the first three months of 1917, anaddi t ional 82 .

The Aviation Section of the Army Signal Corps was established by anact of Congress on 18 J u l y 1914. Our aviat ion effor t in the Mexican punit iveexpedit ion am oun ted to 16 aircraf t which, owing to mainte nan ce problem s,seldom operated. The total number of aircraf t delivered to the Armyprior to World War I was 224; and all of these were training aircraft notsuitable for fighting, bombing, or observation service.The personnel complement of the Army's aviat ion section as the UnitedStates entered World War I was 52 officers, 1100 enlisted men, and 200civilians. Although 139 men had received flight training, only 26 could be1 History of BAP, vol. 1, p p . i i - i i i .2 EDGAR S. GORRELL, The M easure of America's W orld Aeron autical Effort (Norwich

Universi ty , Nor thf ie ld , Ve rm ont , Ja m es Jack son C abo t Professorship Lectu res , no . 6 ) ,pp . 2 - 3 .


15/124

considered qualified pilots, and then only in training craft. The UnitedStates was devoid of airports, and there was not enough civil aviation toproduce pilots as an emergency back-up capabil i ty . Before our entry in thewar we had had but two aviation officers in Europe, and only one of them,Colonel George O. Squier , had been permitted to view the aeronauticaleffort at the front.Prior to April 1917 only four United States companies had successfullyproduced a i rcraf t engines. These we re : the Hal l -Scot t Mo tor Car Co m pan yof Berkeley, Calif., the Curtiss Aeroplane and Motor Corporation ofBuffalo, N.Y. , the Wright-Martin Aircraf t Corporation of New Brunswick, N.J. , and the General Vehicle Company of Long Island City, N.Y.Because the demand made upon these companies was sl ight their capacity

was extremely l imited. Th eir produ ct was buil t for dura bil i ty rath er tha nlightness, and power output was adequate only for training purposes.This was to lead to two requirements: an enormous increase in the manufacture of existing types of engines to meet the needs of a wartime flighttraining program, and, at the same t ime, the development here or theadaptat ion from abroad of a wholly new type of engine for batt le work. 3This was the extent of the United States ' preparation to shoulder i ts shareof the battle in the air on 6 April 1917.From the very f irst , the program for plane and aircraf t-engine productionhad one thing in i ts favor: as news of the "derr ing-do" type of war that

was being fought in the skies over Fra nc e was bro ug ht to the State s, thepublic's imagination was aroused, and the desire for an American air fleet,second to none, became almost universal . On 1 J u n e 1917, the jo in t Armyand Navy Technical Board secured approval of the Secretar ies of Warand of the Navy for a program calling for a total of 22,625 aircraft. 4 T h e$639,241,258 app rop riate d for this pro gra m included 45,250 aircraft enginesand was passed by Congress on 24 Ju ly 1917 with almost no discussion andabsolutely no hesi tat ion. This appropriat ion, the largest to that date, provided $125,290,000 for aircraft, $239,850,000 for engines, and the balancefor the myriad requirements associated with such a vast program. 5Unfortunately, al l the money in the world could not , at this point , overcome the problems involved in producing an air arm for our forces. Onthe hom e front the adversary was time. Almost insuperable prob lem s

3 ARTHUR SWEETSER, The American Air Service (New York: Danie l App le ton and Compa ny , 1919), p. 169.

4 History of BAP, vol. 7, pp. 1827-1828.5 G O R R E L L , o p . ci t , pp . 7 -8 .


16/124

had already appeared. Our engineers were behind the "state of the ar t"of mil i tary-plane engines. In the years between the invention of the airplanean d its eme rgence as a pr ime weapon of war , the Un ited States had neglected its mil i tary application. Th e Am erican aircraf t industry requ iredtremendous expansion to produce the numbers of mil i tary aircraf tneeded by the Allies. With only one qualified observer at the war front, theUnited States knew li t t le of modern f ighter and bomber technical requireme nts and ha d absolutely no experience tha t would help build aircraf t andengines superior to those of the enemy.Most American industry was already saturated by the demands of thearmed services. Skil led manpower, already involved in war production or inthe Ar me d Forces, was scarce. Vital ma ter ials were at a pre m ium , an dpriority systems were not at first adequate.Thus the aircraf t and aircraf t-engine programs presented complicatedproduction problems. The f irst at tack had to be made in the engineeringfield. The difficulties involved in aircraft and engine engineering weremuch more complex than had been encountered in the f ields of electr ic ,marine, steam or gasoline locomotion. The emphasis on power and rel iabili ty with extreme l ightness of construction created problems that increasedgeometrically with the increase of power and the decrease of weight. Tomake matters worse, the European technicians who had experience andknowledge in these matters were 3000 miles away and endless red tape hadto be cut and deadly submarines evaded before their skills could be tapped.Steps were immediately taken to br ing them to this country and to sendAmerican mechanics and engineers to Europe, but al l this took t ime andthus played into the enemy's hands.The nucleus of American production experts available to the aviat ionproject centered in the joint Arm y and Na vy Tech nical Bo ard. These m enwere successful manufacturers but had little if any experience with aircraftproduction. There was almost no expert capabil i ty in the Army.Putting all the obstacles together, it was soon apparent that priorities,propaganda, and superhuman effor t would be required if the United Stateswas to take a place of importance in the battle for control of the air. In anation whose every sinew was strained to the war effort, the only answerwas government control . So the government bought the manufacturer hisland, built his shops, hired his labor, procured his raw materials, parts, andequipment, inspected his work, and audited his books. This kind of control ,in turn, spawned a bureaucracy that , when the war was over , consisted of2064 officers, 31,307 enlisted men, and 8969 civilians organized in 8 divisionswith 376 sections. Th is was the Bu reau of Aircraft Pro du ctio n.


17/124

M issions an d Co m m issionsOne of the major problems in gett ing our production of mil i tary aircraf tstarted was our ignorance of which types of aircraft were needed and whichengines would be required to power them. The shortest road to this knowledge was to tap the experience our al l ies had gained in prosecuting the warin Europe. There were two ways to do this: bring missions of allied experts

to this country and send commissions of our experts to the battlefields andfactories of France, England, and Italy.Just prior to the closed-door session that produced the initial design of theLiberty engine, E. A. Deeds, of the Aircraf t Production Board, cal led uponthe French mission to provide as much information as possible regardingaircraf t-engine requirements in batt le . Although J. G. Vincent and E. J .Hall , the designers of the Liberty engine, were more than competent intheir f ield , they lacked information concerning the horsepower requirementsof engines that would be needed the following year at the front. The Frenchmission was the only source for such inform ation in W ash ing ton in M a y1917. The mission had arr ived in Washington on 24 April 1917, headed byMarshal Joffre on the mili tary side and former Prime Minister Rene Vivianion the diplomatic. It consisted of more than 400 officers and technicians.The French mission technicians provided full answers to all questions ofthe American designers. This aviat ion group was headed by Major JosephTulsane. It was initially composed of 25 persons and had expanded to 72by the war 's end. Major Thebault de Channalon, the group 's engine special ist , was permanently assigned to McCook Field. On the diplomaticside, the French mission worked out a program with the United States to

ship 4500 aircraft to the front by the spring of 1918, to train 5000 pilots,50,000 mechanics, and to attain a monthly production of 2000 aircraft and4000 aircraft engines.The British mission arrived in Washington on 31 M a y 1917 undei theleadership of Lord Northclif fe , a noted publisher . Wing Commander W.Briggs, chief engineer for the Royal Naval Air Service, was attached to theAircraft Board as technical advisor to the Engineering Division, Bureau ofAircraf t Production.


18/124

The largest American commission to go to Europe during the war wasthe Boiling Comm ission. Ra yn al Caw thorne Boiling grad uated from H ar va rdLaw School in 1902 and, at the time of his entrance into the service, hadbeen the general counsel for United States Steel Corporation. Boiling hadbeen in the New York National Guard since 1907 and had been first lieutenant and then captain in charge of the First Aero Company since 1915.(This was the first such unit to be formed.) Although commissioned in therelatively mino r rank of major on 9 J u n e 1917, Boiling was eminentlyfitted to be the chief of the commission. He was able to observe aviationactivities in Europe, translate this information into action to be taken bythe United States, and enter into negotiations with the Allies for the mutualbenefit of all. He was probably the only individual in the country at thattime who combined a knowledge of the infant art of flying and its equipmentwith the legal background necessary to the successful negotiation of interna t ional agreements .

Boiling carried both military and civilian credentials, to be used as conditions warranted. He was to secure, if possible:1. Reciprocal patent agreements between the United States and theAllied powers in order to preclude exorbitant demands of the Europeanpaten t owner on the Amer ican manufacturer ;2. A plan whereby the United States would concentrate on the manufacture of training planes and the Allies would continue their efforts in

the manufacture of combat service planes, since the United States hadabsolutely no experience in this field;3. A plan whereby the United States would rush quanti ty productionof aircraft engines for the Allies, whose "choke point" had always been inthis area.Members of the commission sailed from New York on the SS Adriatic on17 J u n e 1917 and arr ived in Liverpool, Eng land, on 26 Ju ne . Th ey stayedin Eng land unti l 2 Jul y, in France unti l 15 Ju ly, in I taly un ti l 27 Ju ly, andthen returned to Paris . About mid-August 1917 the commission broke up;each member submitted a report on his specialty. Boiling's final report wasthe guide for the beginning of American aircraft production, but thedecision to proceed w ith an Am erican-d esigne d series of sta nd ard aircraftengines had been implemented prior to his report. In fact, the decisionhad been made, the 8-cylinder engine was tested, and the 12-cylinderengine was in final test stages before the Boiling commission disbanded.


19/124

Bir th of a ConceptMass production and standardization of par ts were not new in the UnitedStates in 1917. Eli Whitney, Henry Ford, and others had pioneered the useof standard parts to mass produce arms, automobiles, and other commodities. These techniques, however, had not fully penetrated the aviation field;in the per iod pr ior to World War I there was no requirement for standardization and mass production pr inciples in the manufacture of aircraf tengines because so few were produced. When the necessity arose, the industry of the country had to build machine tools and other mass productionaccouterments before it could translate to moving metal and power thegenius of the design engineers.American aircraft engines at this period were of four basic types:1. Radial : Had a fan-type arrangement of cylinders extending abovethe horizontal to preclude excessive gravity lubrication. When the forced,metered feeding of oil was developed, it was possible to extend the fan into

a circle around the crankshaft and to add individual cylinders or banks ofcylinders to increase power. In this engine the cylinders remained stationary,and the crankshaft revolved. This was a very difficult engine to water-cool,so most were of the air-cooled variety. The bulk of the engine caused greathead resistance; however, a compact, light-weight engine resulted from thiscylinder configuration.2. Rotary: Similar to the radial in that the cylinders formed a circlearound the crankshaft. The crankshaft, however, was fixed and the cylinderswhirled around it. The engine was light and easily air-cooled, but theresistance of the air to the rapid rotation of the cylinders, while it provided

am ple cooling, abso rbed u p to 10 perc ent or mo re of the engine 's pow er. T osolve the lubrication problem an oil-gasoline mixture was used, with consequent excessive oil consumption. Cylinder size was limited (normally five-inch bore) due to cooling problems. As in the radial type, power wasincreased by adding cylinders with a commensurate increase in complexityand maintenance diff iculty .3. Vertical inl ine: The cylinders stand in a row directly above thecrankshaft, as in the ordinary four- or six-cylinder automobile engine. It

2 5 7 - 4 4 1 O - 78 - 2


20/124

was inevitable that this engine would be prom inent in the deve lopm ent ofaircraft engines. The 1912 Graham Clark list of 112 aero engines included42 vertical types. Most aviators did not like the four-cylinder verticalbecause of its vibration, and, as the air war developed, it proved inadequatefor combat flying. The six-cylinder engine was brought to a high degree ofperfection in Germany by Mercedes and Benz. Of the 34 German aircraf tlisted in the summer of 1914, 31 were propelled by vertical aircraft engines(17 were six-cylinder an d 14 were four-cylinder). T his type of engin e,however , was not a strong contender in the United States.

4. V-type inline: The cylinders extend upward at an angle from thecrankshaft in two banks. Although the radials and rotaries were lighter thanthe inlines because of their compact crankshafts and crankcases and theirair-cooling features, they could not develop sufficient power for militaryapplica tions du e to limited cooling capac ity a nd speed of rot atio n. T h eV-type inline was, therefore, the most satisfactory military engine from apower-to-weight ratio. This was the basic design of the Liberty engine. 1W ith the declarat ion of war cam e the need for imm ediate and far-reachingdecisions if the United States was to pull its share of the load in the conflict.The country was faced with the double requirement of producing aircraf tengines in unheard of quantities and in horsepower ratings never beforeachieved. Two basic courses of action were open: to reproduce the mosteffective of the European engines, or to design and build a completelyAmer ican engine .

Examination showed that the All ies were developing or manufactur ing60 different engines, while the Germans, concentrat ing on no more than5 varieties, were outproducing the Allies. Allied production potential wasnot realized because an excessive effort was spent on design. The Americansolution was to place production under government control , al lowing thepooling of all available foreign and domestic experience for use of participating engine designers.The production of European engines in American facilities was difficultbecause the foreign engines were handmade, their par ts were not inter

changeable, and the metr ic system of measurement was used on al l technicaldrawings .2 The Wright-Mar t in Company had spent near ly two years and$3,000,000 on the production of the Hispano-Suiza engine, and by 1917had at tained only l imited production. The same results had been experi-History of BAP, vol. 7, pp. 1823-1826.SWEETSER, op . cit., p. 175.


21/124


22/124

To Des ign an EngineEdward A. Deeds, an engineer and industr ial ist , put the facts together ,arrived at a decision, and proved it right to the government's decisionmakers. He was appointed to the Aircraf t Production Board on 17 M a y1917, after service on the Munit ions Standards Board, and brought greattalent and experience to the problem of organization for aircraft production.O n 2 August 1917 he was appo inted Acting Chief of the Eq uip m ent Divisionof the Signal Corps, and became Chief of the division after his commissioningas a colonel in the regular Army on 24 August . In January 1918 he wasmade Industrial Executive in the Office of the Chief Signal Officer. On24 May 1918 Deeds was rel ieved of duty in the Equipment Division andordered to cooperate in the Hughes aviat ion investigation, in which hebecame the principal figure and was condemned for unjustified delay inthe aircraf t-production program and for waste and mismanagement of thep rog ram. xIt is not the intention of this paper to discuss the several armament

investigations that were launched during the last year of World War I .A dispassionate examination of those investigations connected with aircraftproduction shows that achievements outweighed shortcomings, and propercredit is owing to the men who organized and guided pro duc tion thro ug hits first months of hectic growth.Edward Deeds was the catalyst in the decision-making process. Twoother men provided the abil i ty , experience, and knowledge needed tomake the vital engineering decisions. They were Elbert John Hall and JesseG. Vincent .Hall was born in San Jose, California, 8 April 1882. After educatinghimself by courses from a correspondence school and by night courses inSan Francisco, he was employed by the I . L. Benton Machine Works inSan Franc isco as a steam enginee r at the age of 16. W ith in four years hewas made half owner in the company, and during this per iod he gainedexperience in the design of marine, hoisting, and gasoline engines. In 1903

MARCOSSON, op. cit., pp. 217-233.10


23/124

the company star ted to build auto engines, and by 1905 Hall was workingwith the Heine-Velox Company making comple te au tos . Both companieswere wiped out in the 1906 e a r t h q u a k e .In 1910, after a period of building the "Comet," an auto of his owndesign, he joined with Bert C. Scott in the Ha ll-Scott M oto r Ca r C om pa nyto manufacture industr ial locomotives, gas-dr iven rai lway coaches, inter-urban car bodies for electric railroads, and auto and aviation engines. By1913 aircraft equipped with Hall-Scott four- and six-cylinder aviationengines we re being flown by the best know n aviato rs in A m erica . In 1915the Hall-Scott six-cylinder A-5 was being sold as a military engine to thegovernments o f Russia , China , Japan , Hol land , Norway, Aust ra l ia ,England , and the Uni ted Sta tes .Hall was commissioned a major in the Signal Corps in October 1917 and

was promoted to l ieutenant colonel in April 1918. In addition to his co-design of the Liberty engine, Hall adapted the Le Rhone engine to American metho ds of produ ction, w as a troubleshooter in start ing U.S . produ ctionof the DeHaviland-4 aircraft, and was sent to France at the request ofGeneral Pershing. There, in October 1918, he became chief of the AirService Technical Section. He received the Dist inguished Service Medalfor his wartime service.2Vin cent was born on 10 Fe bru ary 1880 at Charleston, Arkansas. Hisformal education was l imited to at tendance at a country school near Pana,I l l inois, and the Cote Bril l iant Grammar School, St . Louis, Missouri . Hisengineering education was acquired through correspondence courses.From 1897 to 1898 he was with Smith, Vincent and Company, commission merchants in St . Louis. From there he worked as a machinist andtoolm aker in St. Louis until 1902, and in 1903 he becam e sup erin tend entof inventions for the Burroughs Adding Machine Company. He remainedwith the Burroughs Company unt i l 1910. His patents up to this time fill avolume more than an inch thick.F r o m 1910 to 1912 he was chief engineer of the Hudson Motor CarCom pany, and f rom 29 Ju ly 1912 to 10 August 1917 he was vice presidentfor engineering for the Packard Motor Car Company. His patents of thisperiod fill a book more than two inches thick.

Vincent was commissioned a major in the Signal Corps on 15 August1917, and was honorably discharged as a l ieutenant colonel on 30 November1918. H e return ed to Packard in his former posit ion on 1 Ja n ua ry 1919,

2 Elbert John Hall (New York: Ja m es T. Wh i te & Co. , 1924), pp . 1-7 .11


24/124


25/124

for final tes t ing when Hal l was summoned to Washington in May 1917 onoff icial business with the Navy Department. 9 About this same t ime, acombined French and Brit ish mission was surveying the great industr ieslocated in Detroit . On 26 May they visi ted the Packard plant . Vincent spentconsiderable t ime with them, asked innumerable questions, and gained agood idea of the latest developments in European aircraf t . The visi tors wereimpressed with the quali ty of Vincent 's experimental 12-cylinder engine,but informed him that the weight per horsepower was too great for combatuse.10

The next morning, while reading the latest issue of Automobile, Vincen tsaw an article by a Mr. Bradley which discussed the multiplicity of enginesbeing produced by the Allies. Bradley pointed out that this led to high cost,low production, and a serious shortage of parts in the field. This article,added to the previous day 's discussions with the French and Brit ish, determined Vincent on a course of act ion that had been forming in his mindfor some time.He realized that i f immediate and strong action was not taken, the UnitedStates would find itself in the same position as its allies. Even though it wasSunday morning (27 May 1917) Vincen t went to see the Packard Com pany 'spresident , Alvan McCauley, and explained his fears. Quick to realize thedanger , McCauley agreed tha t Vincent should go to Washington , D.C. , a tonce to discuss the establishment of a standardized line of aircraft engines

with Howard Coff in , chairman of the Aircraf t Production Board.On the morn ing of 28 May 1917 Vincent conferred in Washington withCoff in , Deeds, and Sidney Waldon at the Lafayette Hotel . Vincent discussed his l ine of reasoning with the Aircraf t Production Board members;the Bradley ar t icle and the information he had received from the Europeanmission indicated the need for a lighter and more powerful standardizedline of engines. Vincent said he was convinced that some of the Packardengine 's great weightoccasioned by safety factorscould be reducedwithout sacrificing reliability.1 19 Pertinent Facts About the Liberty Motor (San F ranc i sco : Ha l l -Sco t t Moto r Car Company ,

n. d.) Cited hereafter as Pertinent Facts. History of BAP, loc. cit . , indicates this engine was300 h p a nd too heavy for mili ta ry use.10 M cC ook f iles , le t ter , Em mo ns f rom V incent , 4 Dec em ber 1918. This le t ter pre datesthe Vince nt repor t to the Society of Auto mo t ive En gineers in Feb ruary 1919, f rom whichhe is most often quoted.11 I b i d . ; MARCOSSON, op. cit . , p. 238, however, gives Deeds complete credit for thestanda rdized-e ngine idea. He does not ment ion the 28 M ay mee t ing and indicates th atHal l and Vincent were called, in on the 29th and that thei r par t ic ipat ion in the project

13


26/124

Deeds had been thinking along the l ines that Vincent explained in hisdiscussions with the board members. He believed that an engine had to bedeveloped for max im um pow er and efficiency w ith, m inim um w eigh t; i tmust be capable of maximum power and speed dur ing a la rge percen tageof i ts operational t ime; and i t must be economical in the consumption offuel and oil.12 His conversations with Vincent solidified his thinking, and heasked Vincent to work with E. J . Hall on the design of a standardizedengine. Hall was stopped at Cleveland on his way back to California andagreed to re turn to Washington .Both Vincent and Hall had come to Washington to sel l their own products,but Deeds, from his more objective position, could see that the force of theUnited States Government must be the dr iving power of the project , s incethe number of engines required would tax the capabil i ty of the entireindust ry .Hall returned to Washington on the morning of 29 May and ear ly in theaf ternoon of the same day met with Vincent , Deeds, and Waldon in Deeds'suite at the new Willard Hotel . Hall and Vincent knew each other only byreputa t ion . 1 3 They got down to business immediately and by mid afternoonwere laying down two views of a proposed eight-cylinder aircraft engine. 1 4Deeds impressed them with the need for speed and, to assure as rapidproduction as possible, cautioned them to use no untried devices or designsin the new engine. 1 5Vincent called the chief of the Washington branch of the Society ofAutomotive Engineers, a Mr. Zimmershied, to obtain draf t ing mater ialsand draftsmen but was able to get only some tools. Vincent and Hallbought the necessary draf t ing tables and paper .1 6 Vincent cal led Dr.Strat ton of the Bureau of Standards at midnight to get the latest data onBrit ish and French engines. Strat ton had the information for them by themorning of 1 J u n e . 1 7O n the ev ening of 29 M ay , De eds called a session in his suite at wh ichwas completely unexpected. History of BAP, loc . c i t . , pp . 1891-1892, indicates that Deedshad made the decis ion and discussed i t wi th Waldon before Vincent came to Washington.I t is probable that both versions of the story are true, as the facts were evident and bothDeeds and Vincent had access to them.

12 MARCOSSON, op. ci t . , pp. 236-237.13 McCook files, loc. cit. (footnote 10).14 Ib id .]r' MA RCO SSON , op. cit . , p. 238.Ui M cC ook files, loc. cit . MARCOSSON, loc. ci t . , stated th at De eds and. W ald on w ent out an d

bought the required mater ia ls .17 MAR COSSO N, op. cit . , p. 239.

14


27/124

members of the French mission were present . The French, closely questionedby Vincent and Hall , were able to produce a great deal of up-to-dateinformat ion .1 8 The next morn ing Zimmersh ied provided Vincent and Hal lwith a volunteer , J . M. Schoonmaker , who, though he had done no draf t ingfor many years, took over the task of drawing so that Vincent could dictatethe report he and Hall had prepared the night before.19 By the afternoonof 31 May 1917, the report was finished, and the drawings were fairlywell blocked out . At about 3:00 pm, Waldon called, asking that the reportand drawings be brought to the office of George Squier (later Chief SignalOfficer) for presenta t ion to a join t me eting of the Arm y-N avy Aircraf tProduct ion Board . Waldon read the repor t , and Vincent and Hal l exp la inedthe drawings. The Board gave approval to go ahead with complete drawings.

O n 1 Ju ne , two capable layout men came in f rom D etro i t and workedon layouts of the construction features being completed by Vincent , Hall ,and Schoonmaker . The f ive men worked s t ra igh t th rough unt i l Mondayafternoon, 4 Ju ne . 2 0 At this point layouts had been completed of thelon gitu din al, transve rse, rear-ele vation , and camshaft-assembly views ofthe eight-cylinder engine. At midnight on the 4th, Hall and Vincentapp ear ed before the join t comm ittee of the Arm y-N avy Aircraf t Pro ductio nBoard, showed the finished drawings, and further detailed their plans forthe engine. The Board approved the building of f ive 8-cyUnder and five12-cylinder engines and asked that the first 8-cylinder engine be producedas soon as possible. Deeds promised it in seven weeks. Vincent sent thelayout men to Detroit to work on detai led drawings of "long t ime parts"such as cylinders and crankcases. 2 1

Hall then proceeded to Indianapolis and Vincent to New York, wherehe met M cCa uley on the m orn ing of 5 Ju ne 1917. The next day theyre turned to Washington and met wi th Deeds and Waldon . McCauleyagreed that Packard would pioneer the standardized engine and f inance i tunti l reimbursement could be made. He also agreed to loan Vincent to thegovernment for three months and to give top pr ior i ty to the engine projectin the Pac kard p lan t . This t ru ly was the day on which the U.S . S tandard ize d18 Ib id . MARCOSSON does not ment ion th is meet ing.19 Ib id . MARCOSSON, op. c i t ., p . 238, s ta ted th at Deeds an d Wald on met Sch oonm aker

whi le they were buying draf t ing mater ia ls and shanghaied him into assis t ing wi th thedraf t ing. Vinc ent ' s version is more plausible .

20 Ib id .21 Ib id . Al though Vincent d id not say the layout men took the drawings wi th them, i tseems logical that they would. MARCOSSON, op. cit . , p. 239, stated that the drawings weresent to the Packard plant in Detroi t as soon as each was completed and approved byDeeds. Vincent 's version is more logical .

15


28/124

Aircraft engine idea became a fact. 22 On th is same day Vincent requested$250,000 from the Aircraf t Production Board to continue the design andlayout work that was necessary before construction could begin. 2 3Vincent and Hal l le f t Washington for Detro i t Thursday noon , 7 June1917, to supervise the jo b of prod ucing the experim ental engines at thePackard plant. It was decided that the series would include 4-, 6-, 8-, and12-cylinder engines, each having a five-inch bore and a seven-inch strokewith a maximum of interchangeable par ts . The 8-cylinder engine, ratedat 225 hp, was designed to anticipate the requirements at the front for thespring of 1918, and the 12-cylinder, rated at 330 hp, was to be the enginefor 1919 an d 1920. It is im po rta nt to note th at every foreign aircraft exp ertin Washington at that t ime agreed that the 225-hp 8-cylinder engine wasadequate for 1918. Despi te th is unanimity , in May 1917within 90 daysi t was clear , and al l were equally unanimous, that the 12-cylinder enginemust be rushed to fill the need in the spring of 1918. Such was the rap id i tyof developments in the air war . 2 4

Up on ar r iva l in Detro i t a t 8 :00 am, 8 Ju ne , Hal l and Vince nt w entdirectly to the Packard plant to get things "cranked up." O . E . Hun t , t hechief engineer, had already procured some billets of steel for cylinders. Hallleft for Cleveland to get crankshaft forgings made up; he authorized theParke Drop Forge Company to "dig out" his dies for the Hall-Scott engineso th at th e jo b could be done faster: i t was completed in three days. Vincentcalled for volunteers for weekend work in the drafting department, andevery draf tsman volunteered. 2 5 This workforce amounted to 150 m e n .About 86 percent of the detai led draf t ing was completed during this per iod.These were paper drawings and only of the 8-cylinder engine. Almost theentire Pa ck ard dra fting force worked on the jo b the following week an dfinished the remaining paper drawings for the 8-cylinder engine. The jobwas then transferred to Washington, and Vincent took about 25 draf tsmenfrom Detroit and Buffalo furnished by Dodge Brothers, Packard, Cadillac,and Pierce Arrow. When al l paper drawings for the 8-cylinder and the12-cylinder engines had been completed, they were sent out to various

22Ib id . In th is le t ter Vincent showed, the meet ing wi th Waldon and Deeds to havetaken p l ace on the morn ing o f 5 Ju ne . Th e Fe bruary 1919 repo r t to the SAE is date d the

6th . This was a typographical er ror in the typescr ip t of the le t ter , as he could not have metMcCauley in New York on the morning of the 5 th and have conferred, i n Wash ing ton a tthe same t ime. The rest of the chronology of Vincent ' s movements dur ing th is per iodsupports the date of the 6th.

23 SWEETSER, op . cit., p. 176.24 History of BAP, vol. 7, pp. 1892-1893.25 McCook files, loc. cit. (footnote 10).

16


29/124

automobile concerns to have tracings made from the paper drawings.During the week required to complete the tracings, at least 300 draf tsmenwere working on th e jo b at var ious plants. All draf t ing work was com pletedby 15 Ju ne .Meanwhi le , pa t te rns and a wood model were under const ruc t ion . Thewood model was completed and shipped to the Bureau of Standards on16 J u n e . P arts ha d also been ord ere d for the first 8-cylinder e ngin e asfollows: bronze-back, babbit- l ined bearings and aluminum castings fromGenera l Aluminum and Brass Company, Detro i t ; connect ing rods , connecting-rod upper-end bushings, connecting-rod bolts , and rocker-armassemblies f rom Cadil lac Company; camshafts f rom L. O. Gordon Manufac tur ing Company, Muskegon; c rankshaf ts f rom Parke Drop Forge Comp a n y , Cleveland; all bevel gears f rom Hall-Scott Company, San Francisco;bal l bear ings f rom Hess-Bright Manufac tur ing Com pany, Ph i la delph ia ;piston r ings from Bord High Compression Ring Company, Rockford;pistons from Aluminum Casting Company, Cleveland; valves from RichTool Com pany, De tro i t ; springs f rom the Gibson Com pany, M uskeg on;and al l patterns, many dies, and the production of al l other par ts providedby the Packard Company. Assembly and test ing was accomplished in thePacka rd p lan t .

The first sample 8-cylinder engine was delivered to the Bureau of Standards on 3 Ju ly 1917, just over a month from its conception. During thisperiod, while the engine was being rushed to completion, Deeds had thedesign submitted to such well-known engineers as H. M. Crane, chiefengineer o f the Wright-Mar t in Company; David Fergusson , o f the PierceArrow Company; a Mr . Fekete , o f the Hudson Company, and D. McCal lWhite, of Cadil lac, who had designed the Napier . Deeds also had a committee of machine-tool makers approve the design to assure no trouble intooling up, and a group of manufacturers (e .g . , H. M. Leland, C. HaroldWills, of Ford, F. F. Beall, of Packard, and Walter Chrysler, then of Buick)to give their blessing from the production standpoint. All were well satisfiedwith the design.2 6

In this way the Liberty ser ies of engines was launched. Although theelapsed time from the initial design conference to a .completed engine wasphenomenally short , there were adequate safeguards against an infer iorproduct . The engine was designed in six days, but this was the culminationof years of thought and experimentation on the par t of Hall and Vincent ,and the product proved the quality of their effort.MAR COSSO N, op. c it . , p . 240.

17


30/124


31/124

The EnginesIn discussing the Liberty engine as a child of American technology, itmust be remembered that it was designed as a series of engines4, 6, 8, and12 cylinderswith s tandard ized , in terchangeable , mass-produced par ts .This was the most important basic premise on which i ts designers worked.The fact that only the 12-cylinder engine was used extensively should notdetract from the soundness of the series principle. The exigencies of the warsituation relegated the smaller engines to the role of curiosities. All weresound engines, however , with the exception of the L-8, which was takenfrom production because it vibrated excessively.Total production of the different sizes varied considerably. Of the 4-cylinder engines, only 2 were built1; there were 52 6-cylinder engines 2; 15of the 8-cylinder,3 and, overshadowing its smaller sisters, a total of20,478 12-cylinder engines was produced.4 The 12-cylinder power plantwas used experimentally as a 24-cylinder X type, as a double-crankshaft

type, with spur-gear or epicyclic reduction, and inverted and air-cooled.So few of the smaller engines were built that only brief reference will bemade to them. Because the basic premise of the Liberty design was inter-changeability of parts, all engines used the same cylinders, pistons, and1 M cC ook f iles , le t ter , Bureau of S team E nginee r ing (N avy) f rom V incent , 14 M arc h

1918. This reference stated f latly that , as of that date at least , only two had been buil t .In a letter to G. H. Bordel of the Packard C omp any , 3 Nov embe r 1917, Vincen t d i r ec t edthat an L-8 be rebui l t in to two L-4s. These L-4s were to be bui l t by the Hudson MotorCar Company, whereas the f i r s t two were bui l t f rom "scratch" by Nordyke and. M a r m o n .Th ere is no evidence, howe ver , tha t the la t ter were actual ly bui l t . The re i s no other indication of any others being buil t .

2 Ib id . , Chief, Engineer ing Divis ion, f rom Chief, Air Service , 8 M arc h 1924. Th isinform ation is not precise. Th e reference stated tha t 52 engines were stored at Fairf ieldAir Depot , Ohio , but d id not indicate th is as the to ta l produced.

3 History of BAP, vol. 7, p. 1889.4 Ibid. This f igure was also used in a let ter to the Aviation and Aeronautical Engineeringmagazine f rom the Chief, Informat ion Group, Air Service , 18 September 1919. GORRELL,o p . cit . , gave a total of 13,574, but did not include engines produced after 11 N o v e m b e r1918.

19


32/124


33/124

Figure 2. L-6 , 1918. (Smithson ian photo A1746) Air Force MuseumAir Service, about his retention desires. The answer, 17 March , was tha t theengines were not required in the McCook Field program and should not beretained by the Air Service. On 20 August 1924, the Chief, Air Service,telegraphed the Chief, Engineering Division, that he understood McCooknow wanted to retain the engines. The Chief, Engineering Division, wiredback on 21 August tha t he wanted to re ta in the Wright and Thomas-MorseL-6s, but that the Hall-Scott L-6s were not desired.5 Following thisexchange, the files are silent.The Liberty 8-cylinder (L-8) was manufactured by General Motors at acost of $3000 each. It was a 45-angle V-type, weighed 575 pounds, developed 270 hp at 1850 rpm, weighed 2.12 pounds per horsepower, averagedfuel consumption of .547 pounds per horsepower hour and oil consumptionof .050 pounds per horsepower hour . Fit ted with improved intake headersand carburetors, the L-8 generated 330 hp at 1950 rpm.

Like the L-4 and L-6, the L-8 was a vict im of competi t ion with a tr iedand proved engine, in this case the Hispano-Suiza-300. The L-8 was also5 M cC ook fi les, let ter , Ha llett from Clark , 18 D ec em be r 1918. T he Hall-Scotts were

not t rue L iber ty engines. Th ey were cal led "L ib er t y" by the com pan y but d id not usepar ts that were in terchangeable wi th the Liber ty ser ies .

21


34/124

Figure 3. L-8 , 1917. F irst L iber ty Engine. NASM spec imen(Smithsonian photo A54391)the victim of a mo re serious pr ob lem : vib ration . Th is was so serious th atpro du ctio n was stopped at the Buick pla nt whe n a total of 15 ha d bee nbuil t . In January 1918 six L-8s had been produced and were located asfollows: nu m ber 1 engine at McCo ok, num ber 2 in De troit (Pac kard p lan t) ,one at the LWF plant , College Point , N.Y. , one at the Bureau of Standards,one a t Delco , Dayton , Ohio , and one a t the Hudson Motor Car Companyin Detroit .The L-8 was the first of the Liberty series to be built. Recognizing thehistorical significance of this fact, Dr. Stratton, of the Bureau of Standards,requested that one of the original Liberty engines be shipped to him for ahistorical collection. Vincent was quick to agree and ordered the firstLiberty to be carefully "pickled" in castor oil, crated, and shipped to theBureau of Standards.6 This was the engine that had been buil t in 21 days,

6 Ibid., Marmon from Vincent, 24 Ju ne 1918.22


35/124

Figure 4.L12, 1918. Also know n as Mo del A Buick , from NASM's FokkerT -2 Airpla ne. First coast to coast f l ig ht, 1923. (Sm ithso nian photo A4870 D)

and it was first received at the Bureau of Standards on 3 July 1917. T h i sengine, which had been sent to McCook Field from the Bureau of Standardsfor further testing, arrived at the Bureau for the second time about themiddle of Ju ly 1918 for per m an ent display.The Liberty twelve (L-12) is the engine referred to as "the LibertyEngine." Its vital statistics were:T y p e :Cyl inders :Horsepower:R P M :Bore and stroke:

V 451240018005 " x 7 "

Weight per h p :Fuel per hp hour :Oil per hp hour:Average cost:

2.11 lbs.509 lbs.037 lbs$4000T h e w eight of the eng ine varied acco rding to its accessories an d equ ipment. The basic enginedry and wi thou t radiatorweighed as little as 786pounds, but fully equipped for flight it would weigh in excess of 900 pounds.

232 5 7 - 4 4 1 O - 78 - 3


36/124

As will be seen in "The Crit ics," pp. 45-54, there was much controversyabout some of the L-12 's design features. Probably the most accuratecriticism was against the scupper oiling system in the original L-12. Whenit became necessary to increase the horsepower of the engine to the 450 to500 range in order to stay abreast of German developments, a forced-feedsystem had to be used. Engineers such as H. M. Crane and D. McCallWhite consulted with O. E. Hunt and E. J . Hall to effect this and to makechanges necessitated by the "beef-up" of the engine to bearings, crankshaft,and o ther par ts . 7 It should not be implied, however, that these changes,with the exception of the oiling system, were design changes in the samesense of redesigning a basically poor engine. The many tests to which theoriginal L-12 was subjected all gave the same result; the 350-hp L-12 w asa fine, economical, dependable engine.The Liberty engine was the product of the thought and experience ofmany men. Although Vincent and Hall had put i t on paper , they drewheavily from their own proved designs as well as from the designs of othermanufacturers to assure the best product possible.Vincent pointed out that:Every feature going into the Liberty motor had been thoroughly proved outin Europe and also by experimental work in this country. I had personallyspent two years at the P ackard Factory developing the improved type of valveaction which was used in the Liberty motor, as well as light steel cylinders,the water jacketed intake headers, the two part box-section crankcase, and soon through the list of features, which are now well-known as being importantfeatures of the Liberty motor.8In another letter he lists the proved design features he had contributed:

On account of the experience which I had gained at the expense of thePackard Company, I was able to contribute the following major features ofdesign :(A) Crankcase construction split on the center line with the bearings carriedbetween the two halves and through bolts running from top to bottom;(B) Steel cylinders of the Mercedes type of construction, but designedfor rapid production;(C) Camshaft and valve rocker arm construction;(D) Intake Header and Carburetor arrangement including means forheating intake header;7 GROVER CLEVELAND LOENING, Our Wings Grow Faster (Garden Ci ty , N .Y. : Doub leday ,

Do ran and Co. , Inc . , 1935), p . 81 . He re Lo ening says tha t the engine was redesigned byCrane, Whi te , and others . This does not agree wi th the s tory as to ld by Vincentand others .

8 M cCo ok fi les , le t ter, Em mo ns f rom V incent , 4 De cem ber 1918.24


37/124

Figure 5.L-12 with spur reduction gearing, 1918. Air Force Museum(Smithsonian photo A693-C)(E) 45 degree included angle of cylinders;(F) Water pump des ign, loca t ion and dr ive , inc luding self-takeup on thestuffing box;(G) Conn ect ing rods and bea r ings ;(H ) Oi ling system as finally ad op ted in clu din g full pressure feed, no groovesin the bearings and tripple oil pump to accomplish dry crankcase. 9

T h e H a l l - S c o t t C o m p a n y , i n s u b m i t t i n g a n a c c o u n t i n g o f t h e i r c o n n e c t i o n w i t h t h e L i b e r t y t o t h e B u r e a u of A i r c r a f t P r o d u c t i o n , s h o w e dt h a t m a n y o f E . J . H a l l ' s d e s i g n f e a t u r e s w e r e a ls o u s e d i n t h e L i b e r t y :

I t is a noteworth y fact tha t the exper ience obta ined from the ma nufa c tureand product ion of the A-5, A-7, A-5a , and A-7a engines brought out manyfea tures and pr inc iples embodied by our Mr. Hal l in the Liber ty engine .Several of these were as follows:a. The special heavy duty type aluminum pistons. This design was neverused before to our knowledge in any aviation engine.Ib id . , Pot ter f rom Vin cen t , 9 M ay 1918.

25


38/124

s******Air Force MuseumFigure 6.Allison V.G . 1410 (Air cooled Inve rted Liber ty L-12). Geared supercharger and Al l ison epicycl ic propeller reduct ion gear ing. (Sm ithson ian photoA1042D)

b . The method of camshaft dr ive , which resembled in a grea t many waysthe Mercedes drive, excepting that the camshaft was entirely enclosed, whichat the t ime the Mercedes was not.c . The special method of drawing water from the exhaust valve s ide ofthe cylinder, which tends to equalize the heat dis tribution of the cylinder.This is clearly a Hall-Scott feature and was one of the things that lead [s ic]to the success of the Liberty cylinder.d. The propeller flange drive was designed and manufactured solely byHall-Scott and was s imilar upon all types of our engine. This flange as a unitwas adopted upon the Liberty engine and was an entire success .e. The Hall-Scott crankshaft was used throughout all of our engines andthe bear ing diameter e tc . , proport ioned as to the horsepower carr ied on thevarious types. It was a noteworthy fact that this same shaft with correct proportions to the horsepower carried by the Liberty engine was installed in that

26


39/124

NASM spec imenFigure 7.L-12 with Ge neral E lectric supercharger instal led in 1922. (S m ith sonian photo A1099)engine. In fact all the Hall-Scott dies were used in producing the firs t Libertyengine . . . T h e direct drive feature of the Liberty engine was one of thegrea test reasons of the success of the Lib ert y engine . . . Th is design wa sinsis ted upon by our Mr. Hall at the t ime of designing the Liberty engine andapproved by Mr. Vincent only af te r Mr. Hal l had agreed to take the ent i reresponsibil i ty in the event the directly driven job should fail .10A f u r t h e r i n d i c a t i o n o f H a l l ' s c o n t r i b u t i o n t o t h e L i b e r t y i s c o n t a i n e d i n a

b o o k l e t p u b l i s h e d b y t h e H a l l - S c o t t C o m p a n y a f te r t h e w a r . I t p o i n t e d o u tt h a t , p r i o r t o 1 M a y 1917 , t h e H a l l - S c o t t C o m p a n y h a d a 12-cylinderd e s i g n c a l l e d t h e A - 8 w h i c h w a s d e s i g n e d f o r 4 5 0 h p . T h i s e n g i n e w a sr e a d y f o r t e s t w h e n H a l l w e n t t o W a s h i n g t o n o n o f f i c i a l b u s i n e s s w i t h t h eN a v y a n d w a s p u l l e d i n t o t h e L i b e r t y p r o j e c t b y D e e d s . T h e A - 8 a n d t h e

10 History of BAP, loc. cit., pp. 1848-1849.27


40/124


41/124

Figure 8.24 Cyl inder L iber ty " X " , 1918. Air Force Museum(Smithsonian photo A1747)

engine from the ground instead of f rom a maintenance stand. The relocationof the cylinder under the engine provided maximum visibility for the pilot,and, when the inverted engine was air-cooled, the air scoops did not interferewith visibility. Inversion increased the weight of the engine but also gave aslight increase in power.Th e next inversion test was mad e between 5 and 7 Fe bru ary 1919. Th etest showed that inversion was feasible if the problems of oil scavenging andwater flow could be solved. Four years went by before another seriousatte m pt to solve the inversion proble m w as mad e, bu t this time the job wasdone correctly .After the modifications were made, tests proved to be very successful,except that the oil had a tendency to overheat . The inverted enginedeveloped 422 hp at full throttle, with a fuel consumption of .499 poundsper horsepower hour . At 90 percent of normal speed under a propellerload, oi l consumption was 8.7 pounds per engine hour . This engine, L-12#E0501121, weighed 915 pounds, including generator but less star ter , airintake pipe, and exhaust pipe. I t was shipped to Grover Loening in 1924

29


42/124

for installation in one of his designs, although it had initially been testedin a D H -4 B . T he first flight, of 14-minute dura t ion , was ma de o n5 September 1923 by Lieutenant James Dooli t t le . Flight Off icer Carrollmade a cross-country flight to Toledo on 9 September and in October flewin the Pulitzer races to St. Louis and return. The first flight of an invertedLiber ty engine in an am phib ia n was on 7 Ju ne 1923, but in this flightthe aircraft was wrecked on landing when it struck an obstacle in thewa ter . Th e next tests in the am phib ian occurred in Ja nu ar y 1925 wh en,on the 8th, the amphibian was flown from Mitchell Field to Boiling Field,on 2 Fe br ua ry 1925 from Boiling to Lan gley, an d on 13 Fe br ua ry fromLangley to McCook.Grover Loening was quite interested in the inverted Liberty engine.H e mad e "n um erou s visits" to Mc Cook , and on 20 April 1923 m ade astatement to the press that he had tried to interest the Air Service ininverted Liberty engines "f ive years ago." He had also pointed out inlectures r ight af ter World War I that inversion would provide betterpropeller clearance and visibility. When inversion finally came, Loeningdesigned his amphibian for i t . Twenty inverted Liberty engines were buil tfo r the Loening Amphib ian (COA-1) in the Air Service inventory. Thiswas the total production of inverted Liberty engines through f iscal year1925; the programmed 30 engines for that f iscal year were reduced to 20.The program had been for 5 engines up to February 1924 and 20 throughthe balance of the year. The total in the program for 1926 was 15, 10 forthe Air Service and 5 for the Coast Guard. The All ison Company ofIndianapolis , Indiana, was the contractor for conversion; the pr ice perconverted engine was $1472. After 1926 the McCook files do not indicateadditional conversions of Liberty engines to the inverted configuration.In Ja nu ar y 1918 work star ted on gearing the propeller of the L ibertyengine to provide more power. The engines produced for the war in Europewere direct dr ive; this method of propulsion was cheaper to produce andlighter and easier to maintain than a geared method. The pr ime Liberty-equipped aircraf t for use in France by the United States was the D H - 4 ;the direct dr ive L-12 was more than adequate for this aircraf t . I ts use inlarger , multi-engined aircraft and flying boats, however, would requiremore power so that a geared engine was necessary. (Gearing provideshigher engine rpm, which increases power, and lower propeller rpm,which increases efficiency.)

Vincent, after study, chose an epicyclic gear-reduction design as themost promising because it made use of only known and successful con-

30


43/124


44/124

Air Force MuseumFigure 9.Air cooled L-12 wi th geared superch arger, circa 1925. (Sm ithso nianphoto A1042 C)

concerns an engine that had been in stock for a number of years and thatwas made avai lab le to the Army Aeronaut ica l Museum on 1 J a n u a r y 1936.The development of the air-cooled L-12 began in August 1923 whenspecifications were sent to contractors. Air cooling improved flying qualitiesthrough reduction of weight , streamlining and pilot vision, reduced noise,cleaner and smoother operatio n, reduced cost, and improve d ma intain abil i ty .The air cooling of a Liberty engine reduced its basic weight by 141 pounds,and it pulled 436 hp at full throttle. The power was as good or better thanthe water-cooled engine, so the weight adv anta ge w as signif icant. Prod uctio nin fiscal year 1924 was five to seven engines, and in fiscal year 1925 ano the rf ive for use in a grou nd-a ttack plan e. This num ber w as subsequently reducedto two. The cost of these engines, geared and inverted, was more than$8500 apiece, about twice the cost estimated per engine in 100-engine lots.There is no indication that this type of engine, regardless of its advantages,was ever produced in more than two- to five-engine lots.

32


45/124

Ground Tes tsD uri ng the f irst pa r t of Ju ly 1917 the work of consolidating and numberingthe drawings and of checking and correcting tracings, bi l ls of mater ial ,l imits , and the like was com pleted. O n S aturda y, 21 Ju ly, O . E. H un t, atthe Packard plant , informed Vincent that the f irst standard 8-cylinderengine would be ready to run on 23 Ju ly . 1 Vincen t could no t leave Wa shington just then a nd so told H u nt to "work the engine in " imm ediately. 2Hall arr ived in Detroit on 21 July, and on the following day he and Huntran the engine under i ts own power. Vincent arr ived at the Packard .planton 25 Ju ly an d star ted test ing.

I was very anxious to test the motor for smoothness and immediately pulledsome tests from low speed up to 2000 RPM under full throttle conditions andalso under light throttle conditions. I quickly found that the motor was verysmooth, and that up to 2000 RPM at least, there was not even a suggestion ofa period of vibration. This was, indeed, gratifying as it thoroughly provedout my contention that even a large bore eight cylinder motor could be madesmooth by setting the cylinders at an included angle of 45 degrees.

3

Hall and Hunt had run the engine about eight hours the day before andhad found no problems. They had even cut the intake water to br ing thetemperature to boil ing, but no steam pockets had developed.O n 26 Ju ly a nd the morn ing of the 27th the engine was m oun ted on atruck and fitted w ith a propeller. I n the afternoon of 27 Ju ly 1917, H . M .Leland , W . C. Leland , Char les King , Glen M ar t in , H . B . Joy , and Alvan1 Vincen t M SS , "Hi s to ry of t he Deve lopm en t of t he USA S tand ard ized Ai r cr a ft Eng ines"

(proposed repor t ) . This f i r s t s tandard L-8 should not be confused wi th the L-8 whichwas del ivered to W ashin gton on 3 Ju ly 1917. T he W ashing ton engine was not bui l t to runbut to show that an engine had been developed.

2 Ib id . The "working in" was accompl ished by power ing the engine f rom an externalsource.

3 Ib id . The 45 included angle was cr i t ic ized by a number of people as being exper i mental . The pros and. cons of th is controversy wi l l be d iscussed below. No explanat ion wasfound as to why there was no vibrat ion dur ing the test of the L-8 , and yet product ion wasla t e r canceled, because of excessive vibration. I t is probable that the more solid base of at es t p l a t fo rm may have t ended to dampen v ib r a t ion .

33


46/124

McCauley came out to see the run. The engine star ted on the f irst pull ofthe propeller but had to be stopped since the oil pressure line had not beenconnected. The second star t required more t ime because the operators werenot familiar with the new engine and overprimed it. After the second start,i t ran for an hour and a half with no trouble. The f irst hour was at nine-tenths power, and the last half hour at full power. After the run, the oiltemperature in the crankcase was 127F and less than half a gallon of oilhad been used. Glen Martin , one of the few aircraf t producers in Americaat the time, was very pleased with the design and the test. Vincent, alsopleased with the test , said: "I was very much pleased with the ignit ionbecause I had found i t to be l ighter than magnetos, probably more rel iable,and best of al l, . . . i t m ade for easy star t in g." 4With the successful conclusion of the first tests, Vincent called Deeds forpermission to ship the truck and engine to Washington. Deeds agreed, andon 28 Ju ly prep aratio ns were mad e. On th e 29th the unit was shipped,arr iving in Washington on the 30th. Vincent and Hall also arr ived inWashington on the 30th and supervised the unloading and preparation forrun at the Bureau of Standards on the 31st . The engine was ready by8:00 pm, and Hall , Vincent , and Frank Trego, of Trego Motors, ran someshort tests.W edne sday af ternoon, 1 Augu st 1917, Waldon and Deeds ar ranged for atest before the join t technical comm ittee of the Arm y an d Na vy. T he teststar ted at 4:30 pm and continued unti l 6:00 pm. The engine ran perfectlywithout at tention or adjustments and used only one quart of oi l . Vincentto ld the commit tee he recommended immedia te manufacture of the engine ,al though he assured them that exhaustive tests would also be conducted.There was no dissent to his recommendation. 5 The commit tee was a lsoshown the disassembled parts of the or iginal engine that had arr ived inW ashington on 3 Ju ly 1917.

More tests were run on 2 and 3 August for members of the Frenchcommission and Mr. Riker of the Locomobile Company. On 4 August af inal exhibit ion run was made for members of the Committee on Mili taryAffairs of the House and Senate. Again the engine ran perfectly withouta t ten t ion . O n 6 August 1917, the test engine was dismantled, and the par tswere minu tely exam ined and found to be in excellent shape . Hig her pressure

4 Ibid . T his ignit ion, ano the r controversia l point of design, will be discussed below .5 Ib id . This appears to be the f i r s t t ime that the technique of "concurrency" of tes t

and manufac tu r e had been suggested . This technique was not used, again for many years ,but is now an accepted pro duct io n me thod.

34


47/124


48/124

After prel iminary runs, number 29 chokes had to be made and instal led;accordingly, the engine was not ready for the 50-hour test until 3:00 p m ,22 August . A variance was found in the carburetors that could not beadjusted out, but, even though gas economy would be poor, it was decidedto make the test rather than wait for new intake headers.

At 5:00 pm, 22 August , the engine was star ted, and, af ter a warm-upand adjustment, the official 50-hour test run was begun at 5:20 pm. Anonstop run of five hours (until 10:20 pm) was made. There had been somebackfire through the carburetor on the first 5-hour run so 2 hours and 35minutes were expended clear ing the fuel system, the pr imary problemhaving been air-bound water traps. The second f ive-hour run began at12:55 am, 23 August , and was completed at 5:55 am. No problems wereencountered during this run; the engine performed beautifully throughout.After a 10-minute shutdown, the third five-hour run was started at 6:06am and completed at 11:05 am. The oil was changed and a 10-hour runwas begun a t 11:50 am, 23 August . At 9:50 pm, the engine was againstopped and carefully exam ined. No thing was found wrong . Th e nextrun commenced a t 11:00 pm and was completed 10 hours later at 9:00am, 24 August 1917. Again the engine performed superbly, and examination showed no problems. The final run was started at 10:30 am, 24 August,and was completed 15 hours later at 1:30 am, 25 Aug ust . D uri ng the50-hour test run, no adjustments had been made; not even a sparkplugwas changed . The average rpm was 1584; average horsepower, 3 1 5 ; an daverage gas consumption, .58 pounds per horsepower.

Upon completion of the 50-hour run, the engine was transferred to thedynamometer , and a power curve was run wi thout change or ad jus tmentto the engine. The curve showed 337 hp at 1700 rp m an d 346 h p at 1800rpm. The engine was then disassembled and examined, and al l par ts weremeasured under the supervision of the government inspector , Lynn Reynolds. His report showed a m ax im um variat ion of 5 perc ent in rpm with thebrake horsepower for the first run averaging 304 minimum, and for thesecond run averaging 319.5 maximum, an increase of 15.5 hp. This was 6.5percent in excess of the engine 's rated horsepower. The average thermalefficiency of the engine was 23.1 percent; the average oil consumption.0238 poun ds per horsepower hou r; the average wa ter-outlet tem per atu rewas 183, and the average temperature difference was 20.

The accessories used, as shown by the report, were: Delco ignit ionsystem with battery and two distr ibutors; two duplex Zenith carburetors;and AC Ti tan spark p lugs f rom the Champion Ign i t ion Company. Theengine was fueled by Red Crown gasoline with a Baume gravity test of36


49/124

58 at 60 F, and lubrication was provided by grade-B Mobil oi l combinedwith 25 percent castor oil.During the test there was no undue vibrat ion, and, upon completion ofthe last run, the engine was disassembled for examination of par ts . Theensuing inspection showed the valves to be in good condition, with noevidence of warping, overheating, leaking, or scoring and with an acceptable seat surface. The bearing surface on the piston rings and cylinders waswithin tolerance, and only slight wear was evident. These conditions indicated that the lubrication and cooling systems were adequate and thatthere was no leakage of lubricating oil.

Reynolds' report suggested that alteration of the intake manifold wouldincrease engine efficiency as would the installation of an auxiliary breatherin the ver tical-shaft com part me nt. H e also recom me nded tha t a new typeof babbit bushing be designed for the lower connecting-rod bearings topreclude cracking at high speed.After this examination, the engine was reassembled and a five-hour runwas made for Lieutenant Scofield , U.S. Navy, on 27'August 1917. Afterthis run, the engine was disassembled, checked, reassembled, and shippedto Washington, D.C. , on Tuesday, 28 August 1917.The completion of the 50-hour run in an elapsed time of 55 hours was,in itself, a record; most 50-hour acceptance runs of that per iod took aroundfive days to complete. Reynolds' foreword to his completed test reportbears witness to the competency of the basic design:The appended report is a survey of the main phases of the 50-hour endurance test, maximum power curve calibration on the electric dynamometer,and inspection of dismantled parts of the USA twelve Aircraft Engine Nu m ber1, which were made under the supervision of the Equipment Division, SignalCorps at the Packard Factory, Detroit, Michigan, August 22-25, 1917.A consideration of the data collected, we believe, will show that the fundamental construction is such that very satisfactory service with a long life anda high order of efficiency will be given by this power plant, and the designhas passed from the experimental stage into the field of proven engines.7

In November 1917, Lieutenant Emmons directed that the no. 5 production engine from Packard be run to destruction at the Packard plant;Captain Heaslet and Mr. Reynolds were to conduct the test . "This is forthe purpose of avoiding any cr i t icism of Major Hall or Major Vincent ,designers of the engine, conducting this essential test." 8 Vincent was upse tby being excluded from the tests and voiced his opinion strongly to Colonel7 Ib id . Vincent quotes f rom Reynolds ' repor t on the 50-hour test .8 M cCo ok fi les , le t ter , Heaslet and H al l from Em mo ns, 1 No vem ber 1917.

37


50/124

Deeds. He thought the test was for engineering purposes prel iminary topublic tests at Dayton and that no engineering information would beobta ined " . . . because there wi ll be nobody on the job tha t rea lly knowsthe structure in i ts numerous detai ls ."

9Vincent went on to say:These are, of course, orders, and I will be governed accordingly. I want tostate, however, that I consider this is being handled in an unnecessarilydangerous way, and I don't believe that the information will be obtainedthat would have been obtained under the plans which I had laid. You, ofcourse, know that I am the only one that has followed this job in all its detailsfrom the first; I am the only one that knows just what moves have been madein order to step up the M . E. P., and I know better than anyone else whe re thedanger points lie.10

T he above quo tat io n is only one exam ple of the almost jealous interestVincent continued to have in the Liberty throughout his service career .He was completely dedicated to assuring the success of the engine, reactingimmediately and forcefully whenever, in his opinion, this goal was injeopa rdy . Deeds was no t unaw are of Vince nt ' s con tr ibu t ion and knowledge ,and so, in the instance in point, quickly took action to insure that Vincentand Hall would not be excluded. He wrote to Vincent with copies to Hall ,Heaslet , and the Pack ard C om pan y modifying the tests so tha t prel im inaryruns would be made by Vincent and Hall to el ici t whatever engineeringinformation they might desire; a 50-hour test would then be run controlledby Heaslet and Reynolds. The test to destruction would also be controlledby Heaslet and Reynolds, but Hall and Vincent would be present asobservers and would be available to help remedy any ser ious trouble thatmight develop. Deeds' quick conciliatory action in this case typifies theleadership he exerted during his tenure as the head of production to assurethe harmonious cooperation of his subordinates in order to achieve the goalof rapid production of the Liberty engine.

For reasons not now apparent in the McCook Field f i les, Vincent wasnot present at the destruction tests. As it turned out, the tests were run ontwo engines, no. 5 as originally scheduled an d no . 12; the latte r was aproduction model, the former an experimental model. These tests had beenpreceded by a series of 50-hour tests to prove out a "beef-up" of the basicdesign from 315 to 400 plus hp. This "beef-up" was symptomatic of therequirement for more and more horsepower to support the air war inFrance. In effect, the war brought on a compression in time of the development of aircraft and aircraft engines. This compression was particularly

9 Ib id . Emm ons from Vincen t , 3 Novem ber 1917.10 Ib id .38


51/124

felt in the United States because of our almost nonexistent research andproduction pr ior to World War I . The Liberty engine is a good example.I t was thought in May 1917 that a 250-hp eight would be the engine forthe spring of 1918. Two months la ter a 300-350-hp twelve was required,and by September 1917 the horsepower race required 400-450. I t is atribute to the basic design of the 12-cylinder Liberty that this escalationcould be absorbed a nd m ore than 1000 engines produc ed w ithin one yearof the initial design conference.

The basic changes that raised the horsepower of the L-12 from 315 to400-plus were an improved intake-header design and increased valve lift.Fifty-hour tests in September and early October, following these changes,but based also on the experience of our Allies in combat, resulted in amodified oil pump and propeller hub. Because of the increase in power, thestrength of many of the engine partsdesigned for a hundred horsepowerlesswas questioned. A series of 50-hour tests, started in November andcontinued into Fe bru ary 1918, dictated chang es to connecting rods, thecrankshaft, connecting rod bearings, piston-pin retainer, and other details.The oil system was also changed from the scupper to the forced-feed type toprovide more positive lubrication. These tests and the resulting changeswere the responsibil i ty of Major Hall and O. E. Hunt of the PackardCompany. Other than his par t icipation in the design of the engine, this wasE. J. Hall 's major contribution to its success, since the changes he and Huntworked out remained valid, for the most part, for the duration of the war.Shortly after the breakdown tests of the nos. 5 and 12 engines, Hall wasrel ieved from further contact with the engine program and was sent to Dayton to assist in getting the American-built DH-4 in to product ion .1 1

The tests of the two engines were the culmination and proof of the"beefed-up" engine. The first test of no. 5, equipped with low-compressionpistons, was of 29-hour duration, 20 hours at full throttle. Although oilcorisumption was low, the test resulted in excessive ring wear, gave evidenceof valve overheating, and showed that the connecting-rod bearings hadsuffered from insufficient lubrication. The engine was pulled down andreassembled with new rings, the valves were ground, and the main andconnecting-rod bearings were grooved to provide more oil to their surfaces.The next run had lasted four hours when a connecting rod b roke ; examina t ion proved the connecting-rod bearings had been too t ight . The engine wasreassembled, al lowing more clearance, and an experimental high-chromeexhaust valve was installed.

11 History of BAP, vol. 7, pp. 1937-1938, let ter , Ingoldsley from Hunt, 5 March 1919(quoted in full) .

392 5 7 - 4 4 1 O - 78 - 4


52/124

The next run lasted f ive hours, when an exhaust valve burned out .Thirty-eight hours, 25 at full throttle, had been run at this point at anaverage horsepower of 404.9. Back in test the engine ran 11.5 more hou r s(9.5 at full pow er) and brok e the crankshaft (this was the light shaft d esignedfor the 315-hp engine); the chrome exhaust valve held up well . The enginewas reassembled with the redesigned heavy crankshaft and other improvements and was run wide-open for nine hou r s ; this t ime stoppage w as causedby a broken piston pin. The total run was 63.5 hours, 43.5 at full throttle.The average horsepower was 412.48.T h e n o . 12 engine was set up for test with the heavy connecting-rodbearings, the high-chrome exhaust valve from the no. 5 engine, and withupdated parts that had been developed in the series of 50-hour tests runfrom November through January. The f irst run was for 27.21 hours , 21.21

hours wide open. The average horsepower was 407.87. W h e n the enginewas torn down, it was found to be slightly underlubricated; some valvesprings were broken, and all the connecting-rod bearings were slightlycracked. The t iming was 10 late. The engine was rebuil t with specialgrooved connecting-rod bearings, and grooves were scored in the mainbearing s. Th e next ru n w as of 31.35 ho urs ' du rati on , 24.35 hou rs at fullpower. The average horsepower was 399.48. Teardown af ter this runshowed the connecting-rod bearings cracked and four valve springsbroken but lubrication was good. Reassembly was accomplished with stockconnecting-rod bearings and valve springs. After a short run, the engine lost200 horsepower, and i t was found that the connecting rod bearings weretoo tight.

Special Ford connecting-rod bearings were now instal led, and the enginewas run to a total of 108.91 hours, 85.41 wide open. Average horsepowerto this point was 403.74. Shutdown was required only for new spark plugs.The next run was terminated af ter 19 hours, when the t iming gear broke.Examination of the Ford bearings showed them to be at least 50 percentbetter than any others previously tested. After replacement of the timinggear, a 12-hour run was terminated because of a cracked cylinder. Thisfailure was not the fault of design, but rather owing to improper designspecifications for the cylinder. After replacement, a final run of 30 hourswas made with no more trouble. Teardown proved the par ts , including theexperimental chrome exhaust valve, to be within tolerance. Total test was156.66 hours, 128.16 wide open. 1 2

Following these tests a committee of manufacturers was appointed toassist in design matters f rom a production standpoint . O. E. Hunt was a12 McCook f i les , le t ter , Vincent f rom Hunt , 6 February 1918.

40


53/124

me mb er , as well as such men as H enry Cran e , D. Mc Cal l W hi te , and o thersof equal experience and abil i ty . During March, April , and May 1918 thetrem en dou s effort of the preced ing year bore fruit, an d by J u n e 1918 theLiberty engine was acknowledged a success.

41


54/124

Flight TestsThe first flight of a Liberty engine was made on 29 August 1917 a tBuffalo, New York, in an LWF airplane. L-8 no. 3 was the engine tested.Th e p lane was pu l led ou t on the f ie ld a t 3 :15 pm , and " . . . p rom pt lyat 3:30 Aviator Blakely pulled the throttle wide open and left the groundin exactly 3 seconds." J Blakely made a 20-minute flight. Three more flights

were made af ter which Vincentwho was observing with Mr. Flint of theLWF Company and Mr . Crane of the Crane-Simplex Companynotedthat the cooling was perfect with an average temperature of 85F. Lubrication and ignition were also perfect; carburetion, however, was only fair.The carburetion problem was investigated, and i t was found that a rubberfuel line ha d collapsed an d some jets ha d bee n installed th at were a q ua rte rinch too short. When these faults were corrected, the engine ran wellthrough subsequent tests, although still a little rich. On 30 August moreflights were made, and speeds up to 104 mph and 17,000 feet al t i tude werereco rded . Th e first A rm y officer to fly a Lib erty engin e, M ajo r K ilne r, ofthe Air Service, made two short flights and reported that this was the firstaircraft