The High-Speed Frontier. Case Histories of Four NACA Programs, 1920 - 1950

8/8/2019 The High-Speed Frontier. Case Histories of Four NACA Programs, 1920 - 1950

1/196

THE HIGH-SPEED FRONTIERCase Histories of Four NACA Programs, 1920-1950

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION( N A S A - S P - 4 4 5 (01)) T H E d l G H - S P J i E D F R O N T I E R . N81-15969C A S E dlSTOKIES O f F O U f i N A C A P B O G f i A H S , 1920 - T i i f i O1950 (National Aeronautics and Space No1-1597jAdministration) 196 p HF A01; SOD HC ib.5J Uncias

CSCL 01B H1/01 13916


2/196

THE HIGH-SPEED FRONTIER


3/196

NASA SP-445

THE HIGH-SPEED FRONTIERCase Histories of Four NACA Programs,1920-1950

By JOHN V. BECKER

Scientific and Technical Information Branch 1980Nat iona l Aeron au t i cs and S pace Adm in i st ra ti onWashington, DC


4/196

Library of Congress Cataloging in Publication DataBecker, John Vernon, 1913-

The high-speed frontier.( N A S A SP ; 445)Includes bibliographical references and index.Supt. of Docs, no.: NAS 1.21:4451. United States. Nat ional Advisory Committee for Aeronautics. 2 . High-speedaeronauticsResearchUnited States. I. Title. II. Series: United States. NationalAeronautics and Space Adminis t ra t ion . NASA SP ; 445.

TL521.B39 629.132!3 80-607935

l-'or sale l iy the S u p e r i n t e n d e n tof Documents . U.S . G o v e r n m e n t P r i n t i n g OfflqpW a s h i n g t o n . D.f. 20402

PRECEDING PAGE B U N K NO T F J L M C D


5/196

Foreword

It is refreshing as well as un usu al to find such an accoun t as this ofpast technical programs that were so importan t to aeronautical progress.T he autho r deals not only wi th the research in wh ich he was inti m atelyinvolved but also with th e personalities of the participants and the doubts,false starts, and misconceptions that occurred before the final solutionswere achieved.In my view, the flavor imparted to these case histories by the verypersonal impressions of the impact of certain of the key players is anecessary ingredient in getting to the bottom line of how and why thingsworked the way they did in the prime years of the National AdvisoryCommittee for Aeronautics ( N A C A ) .Each of the four programs described grew from small beginnings inthe third and four th decades of this cen tury to become substantialelements of the NACA contr ibution to the achievement of high-subsonicand transonic flight. All of the programs had been essentially completedby the time of the termination of N A C A in 1958 and the transitionto N A S A .

WILLIAM S. AIKEN, JR .Office of Space TechnologyNational Aeronautics and

Space Adminis tra t ion


6/196

Contents

Chapter PageI I N T R O D U C T I O N 1

I I THE H I G H - S P E E D A I R F O I L P R O G R A M 3Background and Origins (1745-1927) 3T he Quest for Understanding (1928-1935) 13Increasing the Critical Speed (1936-1944) 21Supercr it ical and Transonic A erodynamics (1 94 5 -1 95 6)_ _ 36Supercritical Airfoils (1957-1978) 55I I I T R A N S O N I C W I N D T U N N E L D E V E L O P M E N T (1940-1950)- 61T he Choking Problem 62The R epowered 8-Foot High-Speed T unnel; S mall M odelTechniques 68Transonic Airfoil Facilities 78The Annular Transonic Tunnel 80Wing-Flow and Bu m p M eth ods 84The Body-Drop and R ocket -M odel Techniques 85High-Speed Research Airplanes 88The Slotted Transonic Tunnel 98C omment on M anagement M ethods 117

IV THE HIGH-SPEED P R O P E L L E R P R O G R A M 119T he Emergency Propeller Program 121Full-Scale Propellers in the 16-Foot High-Speed Tunnel __ 125Propeller Blad e Pressure D istributions at H igh Speeds 127One-Blade Propeller Tests 129

PRECEDING PAGE BLANK NOT FILMED


7/196


8/196

C H A P T E R I

Introduction

Previous w rit ings about N A C A research achievements, for exampleG. W. Gray's Frontiers of Flight, contain generally excellent descriptionsof the problems of aeronau tics and the solutions developed . To anyon epersonally involved in these programs, however, there are serious omis-sions, particularly th e absence of vital in format ion on how the solutionsactually evolved . M ore often than not the solutions seem to have emergedautomaticallythe inevitable result of wise management, inventiveresearchers, and unparalleled facilities. In the four programs consideredhere the previous treatments passed over so much of the important actionI had seen as a part ic ipant that I was inspired to under take th is effortto complete the record.

To provide fundamenta l ins ights in to NACA's technical accomplish-ments the record should include the doubts and misconceptions thatexisted in the beginning of a project, the unproductive approaches thatwere tried and abandoned , the st im ulati ng peer d iscussions that providednew insights, and the gradual evolut ion of the final solut ion . This kindof in format ion is hard to find. Only bi ts and pieces of it appear in thewritten records in N A C A files. M ost of it is stored in the minds ofthose who par t ic ipated in the NACA programs. A par t ic ipant-author candraw on obvious major assets in establishing this part of the recordhis personal knowledge of the fertile areas to probe, th e roles played byth e others, and the profitable quest ions to ask. T he true facts can belearned through th e process of pooling and edi t ing th e recollections ofall the principal part icipants . In the present study I drew heavily onthe help of many former colleagues who are identified in the acknowl-edgments and elsewhere throughout the text.


9/196

2 THE H IGH -SPEED F R ONT IERThree other historical documents have recently been authored by

former NACA eng ineers : E . P . Har tman ' s A d v e n t u r e s in Research: AHistory of Ames Research Center , 1940-1965 ( N A S A SP-4302); J. A.Shortal 's A New Dimension. Wal lops Island Fl ight Test R a n g e : TheFirst F i fteen Y ears (NA SA RP-1028); and J. L. Sloop's L i q u i d Hydro-gen as a Propulsion Fuel, 1945-1959 ( N A S A SP-4404). These workshad different objectives than the present study and each covers vastlylarger territory, dealing extensively with management and adminis t ra t iveoperations in addi t ion to research activities. Beyond any question eachof these books contains innumerable important contributions to therecord which would otherwise have been lost if these knowledgeablepart icipant-authors had not taken up their pens.

From the large body of NACA's to ta l contr ibut ion to high-speedtechnology the particular programs treated here were selected for tworeasons: first, because of the ir qui te inadequate coverage in previouswritings, and second, because of my int imate personal involvement witheach of them either as a researcher or as a supervisor. All of the programsfall in the category variously referred to as "general," "fundamenta l , "or "basic" NACA research . They are typical of w h a t was done in thiscategory; only one, the slotted tunnel, became a celebrated NACAachievement. (Each of the programs involved a number of differentresearch authorizations and none appears consistently in agency recordsu n d e r the titles I have used. The term "high-speed" is used here in thesame sense that it was used during those programs to mean high-subsonicand transonic speeds up to about M ach 1.2.) M ost of the work wascompleted by 1950 and all of it by 1958; an interesting renaissance ofth e airfoil program in the mid-sixties is also covered briefly.

In the prospectus for the study I proposed to attempt some hindsightanalysis, which is rare in NASA literature but a potentially useful devicefo r improving the R &D process (re f . 1) . M y experience in a previousstudy (r ef. 2) suggested tha t, insofar as possible, hind sight observationsshould be separated from the historical narrative. Accordingly, I havelocated them under the heading "Commentary" at the ends of theappropriate sections.


10/196

N81-15970C H A P T E R

The High-Speed Airfoil Program

B A C K G R O U N D A N D O R I G I N S ( 17 4 5- 19 27 )The f irst discovery of an aerodynamic anomaly near the speed of soundw as made over 200 years ago by the bri l l iant Bri t ish scientist BenjaminRobins, inventor of the bal l ist ic pendulum. He observed what we nowcall th e transonic drag rise by firing projectiles into this device andinferring the law of the i r air resistance as a funct ion of velocity fromthe def lect ions of the p end ulum (ref.3) . He states:. . . the velocity at w h i c h the body shif ts its resistance [law from a V2 to a V relat ion]is near ly th e same w i t h w h ich so u n d is propagated th rough air. I n d eed if the [V*re la t ion] is owing to a vacuum being l e f t b eh in d th e body, it is not unreasonab le tosuppose that th e celeri ty of sound is the very least degree of celer i ty w i t h w h i c h aproject i le . . . can in some w ay avoid th e pressure of the atmosphere on its h i n d e rparts . . . but the exact manner in w h i c h the greater and lesser resistances shif t intoeach other must be the subject of f u r t h e r ex p er im en t a l in q u i r i e s.By the end of the 19th century a considerable body of understanding ofthe differences between subsonic and supersonic flows for projectiles hadbeen built up by the work of Lamb, Ernst Mach, Lord Rayle igh, andothers, establishing the speed ratio V/a ( la ter "Mach n u m b e r " ) as thecontrol l ing nondimensional parameter, and clearly implying drasticchanges in the flow in the vicinity of V/a = 1.The flight speeds of the primitive a i rcraf t of the first two decades ofthis century were so low that compressibility effects were nil as far as theai rf rame was concerned . However ,x by the end of W orld W ar I enginepowers and propeller diameters had increased to the point where ti pspeeds as high as the speed of sound were being considered (r ef . 4 ). Thisappears to have been a matter of particular concern to the British who,


11/196

4 THE HIGH-SPEED F R O N T I E Rperhaps from f i rs thand acquaintance with Lord Rayle igh 's classicalstudies (r ef . 5), or perhaps from his direct personal advices as a memberof th e British Advisory Commit tee for Aeronaut ics , had become awareof a possible critical problem near the speed of sound. That the problemd id indeed exist was first demonstrated by L y n a m ( r e f . 4) in free-airzero-advance tests of a low-pitch propeller model at tip speeds up to1180 ft/sec, th e s t ructura l l imit for the "thoroughly well-seasoned blackwalnut" test blades. T he tests indicated loss of thrust and increase inblade drag, but provided no quan t i t a t ive data or detailed insight intothe phenom ena. W ind -tu nn el tests of a more representative modelpropeller at advance ratios in the range of flight operations were recom-mended .Contemporary wi th this early Brit ish work, the first American testspertaining to the propeller problem were und ertake n at M cC ook F ieldin 1918 by C a l d w e l l and Fales of the U.S. Army's Engineer ing Divis ion .Almost as though com ple m enta ry programs had been deliberately plannedand coordinated, the Am e r i c a n s chose to make high-speed wind-tunneltests of stationary propeller blade sections instead of propeller tests. T hemagni tude of the u n d e r t a k i n g was by no means less than that of theBritish, however, because no high-speed wind tunnel had ever been built,and Caldwell and Fales had to develop the world 's f irst such facility(ref . 6). Exploratory tests using an 8-inch diameter throat were madeat the Nat iona l Bureau of Standards where they were undoubted lyobserved w ith interest by a brillian t youn g Ph. D . in physics, H ug h L .D r y d e n , who had recently jo ined th e staff and who would shortly becomea pioneer investigator of high-speed aerodynamic phenomena. Afterexploratory experimental work on all components, a f inal configurationof th e Eiffel-type tunne l w as decided upon and constructed at M cC o okField.

The tunnel had a 14-inch diameter throat and was powered by a200-hp motor which produced a maximum speed with test model inplace of about 675 ft /sec ( M ac h . 64 ) . (This actual speed w as nevercalculated correctly by Ca ldwel l and Fales. N ot knowing how to deter-mine the true air density in the test section they used the ambient airdensity in the room to calculate an "indicated" airspeed from themeasured pressure drop between intake and test section of the tunnel.)


12/196


13/196

6 THE HIGH-SPEED F R O N T I E Rprogram of high-speed tests of thin-bladed metal propellers ( ref . 8) .R eed had inven ted a semi-rigid metal propeller formed from %-inch-thick duralumin billets, tapering to ^g-inch at the tips. The bendingmoments due to aerodynam ic th rust for the outer portions of the bladeswere balanced largely by the centr i fugal moments due to rotation andblade def lect ion . This design m ad e it possible to em ploy ex tremel y thinsections contrasting markedly with the very thick sections of the woodenpropellers then in universal use. In the in t roduc t ion of his paper, Reedm a d e the following revealing observation: "There has been a t rad i t iongeneral among aeronautical engineers that a critical point exists for tipspeeds at or near the velocity of sound ind icat ing a physical l imit . . . ,something analogous to what is known in marine propellers as cavita-tion." Eviden t ly th e expectation of the sonic anomaly was so widelyk n o w n as to be called a "tradition." Reed goes on to state, however,that th e only support ing evidence for this "tradition" that he couldfind were th e British propeller tests of L y n a m ( r e f . 4 ) . He notes thatLynam used blunt-edged, th ick b lades which, by inference from th epoor performance of bullets f ired blunt end forward, he postulated wouldhave poor sonic and supersonic performan ce . H e therefore conducted aseries of high-speed tests of his thin-bladed metal propellers to investigatethis postulate.A series of metal model propellers of 17-inch, 22-inch, and 4-footd iamete r were tested in still air at tip speeds up to nearly 1.5 times thespeed of s o u n d ; and 9-foot d iamete r full-scale propellers were tested inflight on a Curtiss airplane at near sonic tip speeds with help from theCurtiss Aeroplane and M o to r C o m p a n y . O n some of the test propellersth e very thin (of the order of 4 percent thickness) outer sections hadsharp leading edges. T he data showed no significant changes in thethrust/torque coefficient relationships in the region of sonic speed, andonly small deterioration at low supersonic tip speeds. The sound genera-tion became very loud and "penetrating" but had none of the "confusedand distressing violence" noted in the British tests. Reed concludes thatthe high-speed problems of the British propeller were due to "the useof [thick, blunt-edged] blades not adapted to high speeds." This remark-able investigation w as m a d e before any high-speed section data hadbeen obtained, and it preceded by over 30 years tests of "supersonic"


14/196

TH E H IGH -SPEED A IR F O IL PR OGR AM 7propellers by N A C A . R e e d a p p e a r s to have been unaware of the Cald-well/Fales program or perhaps he considered their highest test speed,V/a = .64, too low to be applicable. In any case, Reed had proved thatthe deterioration of propeller performance at near-sonic tip speeds couldbe avoided by the use of thin sections. The general fai lure to accordproper recognition to Reed's work in the subsequent l i terature may bepartly due to the cumbersome and mislead ing t itle of his report, andperhaps partly to the rather l imited amount of data and analysis itcontained.F ollowing L yna m 's init ial propeller tests in free air the British startedimmediately to develop a powerful turbine-dr iven propel ler dyn amom etersuitable for testing 2-foot diameter propellers at high ti p speeds in their7-foot low-speed w i n d tunne l . Douglas and Wood 's report of this investi-gat ion ( ref . 9) is one of the classical documents of the early years ofaeronaut ica l research. The t ip section of their wooden test propeller w as10 percent thick and compressibility losses started at about V/a = .78.A t their highest tip speed of 1180 ft/sec, V/a = 1.08, the propeller effi-ciency had dropped from 0.67 to 0.36. The British displayed greatingenuity in their deductions of blade section data from th e measuredpropeller data, aided by pitot surveys and optical measurements of bladetwist. The latter measu rements mad e it possible to d erive section mo men tcoefficients showing th e rearward movement of the center of pressure atth e highest speeds. T he inclusion of all of the test data and the detailedanalysis of results in the Douglas/Wood paper m ay account for the factthat it is widely referenced, whi le th e Reed paper, which contained onlyminimal test data and analyses, has seldom been cited in the subsequentliterature.

The Caldwell /Fales program had been accomplished under the gen-eral direction of Col . Thurman H . B a n e, C o m m a n d e r of M c C o o k F i el dand also th e A r m y A ir Service's member of N A C A from 1919 to 1922.Bane is believed to have apprised th e Committee of the results andarranged for their publ icat ion as a N A C A report ( ref . 6) . A lthough theneed fo r follow-on wind tunnel tests at higher speeds w as quite obvious,none w as at tempted by the M cC ook grou p; presumably they moved onto more pressing problems. T he seeds of interest had been sown, how-ever, in both N A C A and in the Bureau of Standards . I t is likely also


15/196

8 THE HIGH-SPEED FRONTIERthat NACA w as aware of the continuing British effort on the high-speedproblem. The personal relationship between Joseph S. Ames, Chairmanof the Physics Department at Johns Hopkins and member of the Execu-tive Committee of NACA and Hugh Dryden of the Bureau, one ofAmes' most outstanding recent graduates at Johns Hopkins, was probablya factor in NACA's negotiation of a contract for the Bureau to extendth e investigation of propeller sections to high speeds. Authorization fo rth e work w as signed in 1922 by George W . Lewis, th e recently appointedExecutive Director of NACA and also its "Budget Officer" (ref. 10).

Lyman J. Briggs, a senior official at the Bureau (soon to become itsDirector and a member of NACA), was in charge of the program. Hepersonally designed the compact balance used in the tests and also partici-pated in the testing. The curve plotting, analysis, and evidently the reportwriting w as done mainly by Dryden, aided b y G . F . Hull (ref. 11).

Primary emphasis was on extending the Caldwell/Fales data to near-sonic speeds. Rather than taking on the costly problem of designing anew wind tunnel or perhaps improving the one at McCook Field, theBureau of Standards group located a large 5000-hp air compressorcapable of continuously supplying air at 2-atmospheres pressure to a12-inch diameter nozzle. This provided them in effect with a ready-madefree-jet wind tunnel having about twice the test Reynolds number ofMcCook facility and a maximum speed of about Mach .95. A disad-vantage was that the airfoil testing had to be done incidentally to-developmental testing of the compressor at the General Electric plantat Lynn, Massachusetts. A nd thus i t was that Briggs and Dryden foundthemselves on Christmas Day, 1923, subjected to the rigors of airfoiltesting in an open jet. Shortly afterwards, as Dryden explained later,"We walked down the street in Lynn discussing the jet and noticedpassers-by staring at us strangely and shaking their heads. It was sometime before we discovered that we'd been shouting at each other at thetop of our voices, both temporarily deaf as a result of working with ourheads only a few inches from the large jet" (ref. 12).

The test models were 3-inch chord end-supported wings whichextended through the jet boundaries. It was not possible to determinethe boundary effects and thus quantitatively meaningful true sectiondata could not be obtained. Qualitatively, however, the results were


16/196


17/196

10 THE H I G H - S P E E D F R O N T IE Rbetween Mach .95 and 1.08, following the same pattern as the dragof projecti les . And, for the first t ime in history for an airfoil, the bowshock wave was seen standing about Va-inch ahead of the leadingedge at Mach 1.08.There was also one m a j o r misinterpretat ion of the pressure data. The

authors stated that the lowest observed upper-surface pressures corre-sponded approximately to the a t ta inment of the local velocity of sound,and that lower pressures could occur only in "dead air" spaces. "Thisobservation suggests that in an airstream obeying the law of Bernoullithe pressure cannot decrease indefinitely but reaches a l imit . . . near thecritical [sonic] value of 0.53." This is, of course, quite wrong. An ex-aminat ion of their pressure data actuall y shows q ui te clearly th e existenceof supersonic local velocities ahead of the probable locations of theupper surface shocks. Unfortunate ly , th e orifice spacing of 0.25 chordin the aft region of the upper sur face precludes any precise examinationof the flow and this m ay explain th e misinterpreta t ion .The pressure data underscored what was already evident from theearlier force datathat the burble phenomena were exceedingly com-plex, involving shock-boundary layer interactions quite beyond anypossibility of theoretical treatment. Future researches would be almostexclusively exper imenta l ; not unt i l th e later forties, when it was learnedthat th e shocks moved off the airfoil for M ach numbers greater thanabout 0.95, did v alid theoretical solutions appear for M ach 1 and above.In 1927 a conference of N A C A and the military services recom-mended a final extension of the Briggs/Dryden program to provide forcedata fo r additional more recent sections of interest to propeller designers.Inc luded was a typical 10-percent-thick airfoil used by Reed in his metalpropellers which was one of the best tested for that thickness ratio( ref . 15) . T he last extension was a series of tests of circular-arc sections,recommended by the authors for the outer regions of propellers fo r veryhigh tip speeds (ref . 16) . Unaccountably, they made no reference toReed's work of nearly a decade before suggesting a similar use of sharp-edged sections.Al though NACA cont inued to sponsor the Briggs/Dryden programuntil it ended in 1930, it had been decided in 1927 to develop a newhigh-speed tunnel at Langley and to embark on in-house NACA re-


18/196

THE H I G H -S P EED A I RF O I L P R O G R A M 11search at high speeds. T he ini t ial direct involvement of the staff withhigh-speed research was the Jacobs/Shoemaker investigation of thrustaugmentors for jet propuls ion ( ref . 17) in 1926. Al though the jet-propulsion connection w as m u c h ahead of its time, this study stirred inJacobs th e beginnings of a strong interest in high-speed aerodynamics.T he thrust augmentor inspired i n G . W . L ewi s not only keen interest,but also a display of t echn i ca l i mag i na t i on and inventiveness seldomseen in administrators at his level. He saw in this device a possible eco-nomical means of powering a large high-speed tunnel , using wastehigh-pressure air from the f r eq uen t blow-downs of the Variable Densi tyTunnel ( V D T ) ( r e f . 1 8 ) . D r. Ames , now N A C A C h a i r m a n , had alsofollowed the high-speed testing of Jacobs, Briggs, and Dryden with in-terest. A ll were aware that a m a j o r deficiency existed in the Briggs/Dryden invest igat ions, namely the u n k n o w n je t b o u n d a r y effects. T hein-house program was therefore l aunched wi th the i m med i a te ob ject i veof obta inin g accurate q ua nti tat iv e high-speed section d ata for propel lersto supple me nt the com parati ve results of Briggs and D ryd en (re f . 1 9).

Pre l i m i n a ry tr ials were made by Jacobs wi th a 1- inch d ia m eter throatwhich ind icated that th e je t -augmentor pr incip le could indeed be success-fully applied to drive a high-speed tunnel . Sufficient pressure was avail-able during V D T blow-down to induce supersonic f lows, and sonic con-di t ions could be m a i n t a i n e d for long periods. Even with a 12-inch throatJacobs' estimates showed several minutes test duration. T he dimensionsand configuration selected for the f i rst tunnel coincided with those ofthe firs t Briggs/Dryden tes t ing at L y n n : a 12-inch open throat with3-inch chord wings . T he proport ions of the open throat and i ts dif fuserinlet were similar to those employed in the N A C A V D T a n d Propel lerR esearch Tunne l (PR T) facilities. However , fol lowing Briggs' andDryden's design, the test wing spanned the jet and was supported atthe ends on a photo-recording balance designed by Jacobs and his group( re f . 19) . It is unclear now what the rationale was for obta in ing moreaccurate section data with this arrangement since i t dupl icated theBriggs/Dryden setup in all i mpor tan t respects except for the addi t ionof a diffuser. Several of those interviewed indicated that this was a "realwind tunne l w ith good f low" whi l e the fo rmer w as "only an open jet"and this m ay reflect the early N A C A attitudes. Or i t may be that the


19/196

12 THE H IG H - SPE E D FRON T IE Ropen throat w as in t ended to provide a direct comparison with the earliertest results , pr ior to the development of an improved closed throat con-figuration. But this could not be verified in the interviews. In any case,by mid -1928 N A C A was ready to begin using i ts f irs t high-speed win dt u n n e l ( r e f . 2 0 ) .C O M M E N T A R Y

T he c o m b i n a t i o n of the British tests of model propellers at high ti pspeeds, Reed's tests of th in metal propellers , and the American invest iga-t ions of blade sect ions by Caldwell and Fales and by Briggs, Dryden, andHull constitutes one o f the first concerted efforts of the fledgling aeronauti-cal c o m m u n i t y to solve what w as feared to be a serious obstacle toprogress. By any s tandar ds , th e array of talent mustered was t ruly ex-ceptional . W ith in the short t ime of abo ut five years, the problem wasaccurately del ineated and pract ical solut ions had been found . The useof thin sect ions at low angles of attack in the tip region was the basicprescript ion, and this was readily pract ical for the new metal propellerdesigns that were be gi nn in g to appear . Beyond tha t , how ever, the useof gearing, and finally variable-pitch and constant-speed propellersel iminated the problem entirely for the airplane speeds foreseeable in1925. Accordingly, most of the researchers initially involved moved onto more pressing problems in other areas. Briggs and D r yden had de-veloped sufficient scientific and personal interest to carry on for a t imeund er thei r own m om en tum , but they both b ecame increasingly involvedwith other pursuits. The pressure for blade-section research was fur therdimin ished w h e n N A C A ' s new " P R T " w as placed in operat ion in 1927.

Certa in ly there was little comprehension in 1927 that the airframe aswell as the propeller would become subject to compressibi l i ty problems.A d v a n c e d pursuit planes reached speeds of only about 200 mph and i twould be six or seven years later before ser ious speculat ions regardingth e "500-mph airplane" would appear . A scan of the l i terature of themid -twenties show s only rare suggestions of very high f u tu r e speeds. (O neoverly sanguine predict ion found in a NACA re-publ icat ion of a 1924F rench do cum ent (re f . 21) envis ioned aircraft flying at Mach 0.8 ormore by 1930, in cl ud in g develop men t of some wholly new but unspecified


20/196

THE H IG H -SPE E D A IR FO IL PRO G RA M 1 3type of propulsion plus appropriate new high-speed wind tunnels tosupport these developments .)

The init iat io n of in-hou se N A C A research in high-speed aerodynam icsin 1927, coming in a period wh ere in d us try pressures for such work werenonexistent (except for extend ing the Briggs/Dryden program to a logicalc o n c l u s i o n ) , has been called an act of "great fores ight" ( ref . 20 ) . M oreprobably, th e start at this part icular t ime was a natural consequence ofJacobs' 1926 invest igat ion of jet augmentors . This provided both th ebasis for Dr. Lewis ' imaginat ive suggest ion to use VDT blowdowns toactuate a "large" t u n n e l , and a sufficient level of interest in both m ento take on such a project. Jacobs and Lewis also realized in tu i t ive ly thatthere was a place in Langley's burgeoning stable of wind tunnels for onethat could deal with high-speed problems, el iminat ing continued d e-pendence on the Bureau of Standards and outside test facilities.THE QUEST FOR U N D E R S T A N D I N G ( 1 9 2 8 - 1 9 3 5 )

O n J u l y 16, 1928, the man who was to dominate Langley high-speedaerodynamics for the next 30 years reported for duty. John Stack wasthe son of Irish-born parents, a heritage which may have accounted forhis personal charm, garrulousness, love of horses, and ability to absorblarge quant i t ies of whiskey. Educated at the Chauncey Hall School andM assachusetts Inst i tute of Technology, his distinctive accent retainedlittle to suggest an Irish background (it can be described as upper-classBostonian wi th var iat ions) . S tack was at his best in the midst of conflict,crusading passionately for some cause such as a new wind tunnel againstth e forces of reaction and stupi d ity (w hic h in his view was anyone andeveryone who had any objection to the project ) .He had applied for N A C A e m p l o y m e n t d u r i n g his senior year at M I T ,where several of the faculty were involved in var ious ways wi th NACAactivities. O n h i s arrival there were fewer than 60 professionals atL angle y, loosely organ ized in "sections" attached to the research facili-ties they operated. A s was custom ary, Elton W . M il ler , the father ly, mild-mannered Chief of Aerodynamics, escorted Stack around th e Laboratoryi n t roduc i ng him to vir tual ly the ent ire staff. After the tour , "M r . Miller,"as he was unive rsal ly cal led, ind ulged himself with a final quest ion that


21/196

14 THE HIGH-SPEED F R O N T I E Rhe invariably directed to new engineers with pr ivate enjoyment, ''Wherewould you prefer to be assigned?" Believing he had a choice Stack said,"the VDT." "Very good, I had already decided to put you there,"M i ll er r ep li ed . ( M o r e often than not , as in my own case, the newarrival's choice did not agree with M r. M i ll er 's and he was told, "WellI have decided to place you elsewhere. L e t m e know in a year or twohow you l ike it .")

Stack w as assigned immediately to the 12-inch high-speed tunnelproject which w as then under cons t ruct ionthe lone NACA researcherin this field. For the next decade his work would be closely followed byEas tman N . Jacobs, V D T section head, a man for whose technicalsagacity Stack had enormous respect . Both men had the same kindof restless energy and pr agm at ic appr oach to research problems. Neitherwas a theoret ician, al though both of them f requent ly supported theoret i-cal work by others and f requent ly m a d e use of such work. Their ow nactivity in this area w as l imited to apply ing th e usual analyt ical toolsof th e engineer .

In his first years at Langley, Stack w as quite modest about his knowl-edge of aer odynamics and was eager to learn. A s W . F . Linds ey , w hoarrived in 1931 and was a major cont r ibutor thr oughout th e high-speedairfoil program, puts it, "Pract ical ly all we knew about compressibleflow theory at that t ime w as w h a t was writ ten in five or six pages inGlauer t ' s 1926 textbook." Among the five professionals in the VDTgroup in 1930, Stack was chosen to act as section head in Jacobs' absence.( In those days, there was no formal a p p o i n t m e n t to the assistant sectionhead pos i t ion. ) Apparent ly Stack 's general depor tment as a junior engi-neer w as e x e m p l a r y ; the tough assert ive character is t ics mentioned earl ierbegan to show themselves slowly at first, not reaching full f lower untilafter Jacobs depar ted Langley in the mid-for t i es (refs. 22, 2 3 ) .The f irs t attempts to operate the 12-inch tunnel with i ts unique jet -augmentor induct ion dr ive produced such violent f low oscillations thatit w as soon decided to convert to a closed throat. Stanton's small super-sonic tunnel in E n g l a n d , in which the test airfoil spanned th e throat( ref . 2 4 ) , m ay have suggested the configurat ion. This configurat ioneliminated the pulsat ions and the uncertain large boundary effects of theopen-tunnel setup, but suffered large constriction effects which were not


22/196

TH E H I G H -S P E E D AI R F O I L P R O G R A M 15understood at that time. Pressure distributions on the 3-inch chord air-foils were found to be similar in character to the Briggs/Dryden resultsbut different in deta i l . There was no way to tel l whether either set ofdata w as correct at the higher speeds. T he renowned British theorist,G . I . Taylor, visited L angley in late 1929 and examined th e data. R e-sults of his recent studies of subc rit ical compressible flows by the electricalanalogy method seemed, by inference and extrapolation, to cast doubtson the 12-inch tunnel data. Discouraged, Stack and Jacobs set the dataaside and dec ided to go back to the openthroat conf igu rat ion, with thefirst objective of achieving stable flow. ( It is now believed that th e closed-throat data were valid at speeds below the onset of tunnel choking. Un-for tunate ly they were never published and were later disposed of .)

Another famous vis i tor , Amel i a Earhart , came to view a test run inthe high-speed tu nn el at this same t ime period. She was clad in a raccoonfu r coat . When the tu n n e l started she leaned forward to feel the flowof air into the entrance bell and her coat was instantly sucked into thebell, causing a large tear and terrifying i t s owner ( re f . 2 2 ) .Stack has reviewed the laborious succession of design changes to thet u n n e l ( re f . 20) that followed Taylor's visit: reversion back to the openthroat modif ied by incorporation of a %-inch annular enlargement atthe entrance to the diffuser and a large reduction in length of the opensection; rejection of the open thro at, pri m aril y because of windage effectson the balan ce and secon da rily because of flow pulsation s; a second re-version back to the closed throat 11 inches in diameter but virtually thesame arrangement at the 12-inch tunnel except for the %-inch step atth e entrance of the dif fuser and the use of 2-inch chord test models. Byth i s t ime (1931) a high-tip-speed propeller test had been made in theP R T w h i c h afforded a basis fo r comparison and evaluat ion of the closed-throat wind tunnel data. Stack applied Goldstein 's method to calculatethe performance of the test propeller using the new high-speed sectiondata from the 11-inch tunnel ( re f . 25 ) . H is results agreed with t h e P R Ttests except that th e onset of performan ce deter ioration in the calcula-tion occurred at a somewhat lower M ach num ber . W e n o w know thisshift in speed was due to a combinat ion of constriction effects in thetunnel , Reynolds number differences, and three-dimensional relief at thepropeller tip. Still, th e comparison w as close enough to conf irm that the


23/196


24/196

THE H IGH -SPEED A IR F O IL PR O GR AM 17sound was reached locally and that f low separation was induced by effectsof th e shock. This emphasized th e idea that shapes should be sought withth e least possible induced velocities. Stack has described this concept as"the inspiration . . . w h i c h led immed ia te ly to a new approach to theproblem of developing better shapes" ( re f . 20) .

Shortly after the first dramatic results of the schlieren tests had beenobtained, Jacobs came back from a meeting with Reid and announcedthat $ 10 000 of Publ ic Works Adminis t ra t ion funds would be madeavailable to build a 24-inch high-speed tunnel, provided that a designcould be accomplished in a few weeks. Justification for the larger tunnelrested entirely on Jacobs' argument that it was the low Reynolds numberof the 11 - inch tu nn el data which was responsible for the discrepancywith the PRT propeller data mentioned previously. Jacobs' idea was tobuild a 24-inch tunnel exactly similar in all respects except size andReynolds number to the 11-inch tunnel, and this was the basic designspecification. A number of improvements were included however: a new5-inch schlieren system, an improved balance, and a photo-recordingmult ip le- tube manometer .

The tunne l was erected outside the VDT bui ld ing on a reinforcedconcrete base which also formed the entrance section and the test cham-ber surrounding the tunne l throat. I ra Abbott qu ick ly became an expertin reinforced concrete. Dick Lindsey and Ken Ward were instructed byJacobs to design the entrance section ind epend ent ly and bring their resultsto him for comparison. (They were sufficiently similar to merit Jacobs'q u i c k app roval .) Stack specialized in aero d yna m ic issues and coordinatedth e design project. T he design w as completed as scheduled and the tunne lw as bui l t approximate ly with in the cost limitation in about one year'st ime . F igure 1 shows the two principal operators of the 24- inch tunnelinvolved with a survey rake installation in a scene typical of the mid-thirties.The f irst test in the new tunnel involved a much more important issuethan the Reynolds number-effect question for which the tunnel had beenbuilt . Jacobs had been invi ted to present a paper at the fo r thcomingF i f th Volta Congress on High Speeds in Avia t ion in I ta ly , and he realizedthat an e luc ida t ion of w h a t w as actual ly happening in the compressibil i tyburble phenomena would be most appropriate and im porta nt, especially


25/196

18 TH E HIGH-SPEED FRONTIER

F I G U R E 1.John Stack and W. F. "Dick" Lindsey (standing inside the24-Inch High-Speed Tunnel) in the thirties.in view of the possibility now of presenting flow photographs in additionto pressure distributions and forces. Accordingly, a 5-inch chord 4412airfoil model built for the VDT with 5 6 small pressure holes was testedin the 24-inch tunnel and simultaneous pressures and flow photographswere obtained for the first time. After describing the new understandingof th e burble phenomena achieved in the Langley program, Jacobs wenton to derive for the first t ime th e relation between th e low-speed suctionpressure peak on an airfoil and the speed ratio (Mach number) at whichthe local speed of sound would be reached. That is, the critical Machnumber could now be related to or estimated from the low-speed pressure

ORIGINAL P A G E IfcO F POOR QUALITY


26/196

TH E H IG H - SPE E D A IRFO IL PROG RA M 19signature of the airfoil . Obviously this relat ion contained a powerful im-pl icat ion: th e cr i tical M ach num ber could be increased by shape changeswhich could be deter mined by simple incompressible theory or low-speedtests.

A N A C A Technical Note covering some of the same ground as theVolta paper w as written by S ta ck ( r e f . 2 7 ) , and a more elaborate Tech-nical Report (ref . 28) was issued later in which Stack credits Jacobswith the cr it ica l M ach num ber der iva t ion . Together with Jacobs' paperthese publ icat ions proclaimed the f i r s t major contr ibut ion of NACA in-house high-speed researchthe fu nd am en tal und ers tand ing of the burblephenomena der ived in large part from the revelations of the schlierenphotographs.

C O M M E N T A R YThroughout the history of N A C A new er types of test facilit ies were

often placed into service somewhat prematurely in order to capital ize ontheir advanced capabi l i t ies . This f requent ly resulted in some unfore-seen difficulties. In the case of the first N A CA high-s peed wind tunnelthese difficult ies were compounded by strong interactions between th etunnel f low and the test airfoil f lows at high speeds. Furthermore, th ehigh-speed airfoil problem was so new that no cr i ter ia existed for judgingwhether val id data were being obtained, a s i tuat ion which had its rootsin th e lack of knowledge of w h a t actua l ly happened in airfoil flows w h e nthe compressibi l i ty burble occurred. I t seems obvious now that the firstgoal in such circumstances should be to a c q u i r e at least a q ua l i t a t iveunder s tand ing of the basic flow phenomena, and that this should alwaysprecede any program to pr oduce force data for use by designers. T heclosed-throat 12-inch tun ne l of 1929 could have been used to providethe great enl ightenment from combined pressure and schlieren pictureswhich did not come until some five years later in the program actual lypursued. I t w as the eventual achievem ent of this f u n d a m e n t a l u n d e rs ta n d -ing that now stands out as N A O A ' s first major accompl ishment in high-speed aerodynamics . I t also formed th e solid base on w h i c h th e advancesin critical speed discussed in the next section could be made . By com-parison, perfect ion of the testing technique so as to acquire improved


27/196

20 THE HIGH-SPEED F R O N T I E Rforce data fo r designers, which was the goal of the early program (ref.19 ) , produced only relatively unimportant data pr ior to 1934.

A principal factor in the long delay in acceptance of the closed-throatdata was the doubt engendered by G. I. Taylor in 1929. W. F. Lindseypoints out tha t Taylor's real expertise extended only to the critical speed,and beyond th at point his speculations should not have been taken asseriously as they were (ref . 23) . E . N . Jacobs also feels that th e cautiousconservatism often displayed by so-called "experts" when they are askedto j u d g e new phenomena beyond their previous experience has been acause of undue delays ( ref . 19) . A s another example he cited his 1926investigation of thrust augmentors ( ref . 17) . Lewis turned th e report ofthis work over to Dryden for review. Dryden expressed some doubts aboutit based on momentum considerations. A s a result, publication was heldup for several years, until 1931. Another obvious example was Theo-dorsen's off-hand "optical- i l lusion" pronouncement, but by that timeJacobs and Stack had acquired enough confidence and momentum toproceed on their ow n judgments . A s a general rule, th e speculations anddoubts of experts in view ing new phenom ena should not be overrated.T he essence of the idea that th e critical speed could be related to thelow-speed velocity profile of the airfoil was first stated by Briggs andD r y d e n in 1925 (re f . 11 ) . H owever, th e only use they made of it wasto show th at the trends in the ir observed critical speeds w ere q uali tativel yconsistent with th e concept. They never considered applying th e idea asa tool to develop improved shapes. It remained for Stack and Jacobs torecognize the potential of this conce pt and to put it to q ua nti tat iv e use.They established the mathematical relat ionship between Mcr and thelow-speed peak negative pressure coefficient, thereby making it possiblefor designers to estimate from low-speed theoretical or experimental datathe critical speeds of their designs, and providing high-speed researcherswith a practical theoretical tool fo r achieving improved forms. Stackclearly felt a sense of exci tement and fresh "inspiration" from thisaccomplishment (r ef . 20 ) . I n his view the "new" concept was one of thefruits of the combined pressure and schlieren study for the 4 4 1 2 airfoil in1934. W hethe r previous reading s of Briggs and D ryden had p lanted theseeds of the idea matters little; the revelations of the 1934 research gaveth e concept real meaning and inspired its useful application.


28/196

THE H IGH -SPEED A IR F O IL PROGRAM 21I t will be difficult for today's researchers to comprehend the procure-ment story of the 24-inch high-speed tunnel. That kind of q uick actiondesign by the research staff in three or four weeks and construction fo rsome $ 12 000 in less than a yearis rarely seen in the present complex

organization. F acility procurem ents follow a complex process of reviewsand approvals and m an y stages of design and construction involvingseveral inhouse and outside agencies. Procurement of test models hasfollowed a similar pattern. Of perhaps even greater concern than t imeand cost is the di scou raging effect of these long and costly procurementson the interest and in i t ia t ive of researchers.

Periodically throughout th e history of NACA s i tua t ions would arise,in the research programs as well as in facility procurements, where it wasobvious that the normal agency procedures could not accomplish thejo b effectively wi th i n time or cost limits. Small teams or task groupswould be set up in these cases, relieved of their normal duties and ex-empted from normal lines of authori ty, burdens of paperwork, etc.that is, freed from th e restraints of the large parent organization, whiletaking advantage of its services and facilities whenever possible. Almostinvariably these special groups did an impressive job.The use of this special-group technique, not only in emergencies butas a regular device in R&D and procurement programs for recaptur ingth e benefits of the small organization, offers partial salvation to today'senormous bureaucracies, industrial as well as governmental .I N C R E A S I N G T H E C R I T I C A L S P E E D (1936-194 4)

O n the mornin g of A ugust 31, 1936, I boarded a street car inHampton, Virginia , and traveled to Langley F ield to report for d u t y as aJunior Aeronaut ical Engineer at $2000 per annum. Af ter the usual shortindoctr inat ion in his office, M r. M iller escorted me to the 8-foot high-speed tunnel and introduced me to Russel G. Robinson who would bemy boss. Robinson had been project engineer for this new facilitysince its conc eption in 1933 at abou t the time the 24 -inch tu nn el wasstarted. Fol lowing th e usual practice of that period, he had more or lessautomat ica l ly become head of the small group of researchers w ho wouldnow operate th e facility. T he basic idea fo r this large tunnel is believed


29/196

22 THE H IGH -SPEED F R ONT IERto have been first suggested by Jacobs. It was to be a "full-speed"companion to the "full-scale" tunnel, using the same drive power (8000hp) to produce 5 00 mph-p lus in an 8-foot throat as the full-scale tun nelused for its 100 mph-plus speed in a 30- x 60-foot throat. The namewas later changed to the less vague "500-mph tunnel," and finally to the"8-foot high-speed tunnel." T he very large power input in this closed-circuit tunnel had introduced an unprecedented heating problem whichRobinson had solved by an ingenious air exchanger in which part of thehot air was cont inuously and efficiently replaced with cool outside airwithout the need for any aux i l ia ry pump ing or air cooling equipment.

We spent the rest of the morning examining the new tunnel and thenwalked down to the lunchroom on the second floor of the adminis t ra t ionbui ld ing . T he entire professional staff and some of the support people,except for a few "brown-baggers," assembled here everyd ay for a simplebut excellent plate lunch costing 25 or 30 cents (35 cents on s teak days) .W alter R eiser, in charge of "M aintenance," and also head of the em-ployee's organization which operated the l u n c h r o o m , the Langley E x-change, personally mark ed d own everyone's charges as they passedthrough the line and once each month collected payment. The lunchtables had white marble tops, a feature which was a great boon totechnical discussions. O ne could draw curves, sketches, equations, etc.,direct ly on the table, and easily erase it ah1 with a hand or napkin . Thisgreat unintentional aid to communica t ion w as lost in later years when th elunchroom w as replaced with a much larger modern cafeteria .I t was excit ing and inspir ing for a young new arrival to sit down inthe crowded lunchroom and f ind himself surrounded by the wel l -knownengineers who had authored th e N A C A p a p e r s he had been studying asa student . I well remember that first day at a table that included StarrTruscott , Ed H a r t m a n n , and Abe Silverstein. There were no formalpersonnel development or training programs in those days, but I realizenow that these daily lunchroom contacts provided not only an in t imateview of a fascinating variety of live career models, but also an unsurpassedsource of stimulation, advice, ideas, and amusement. A n interesting con-sequence of these daily exchanges and discussions was that often no oneoriginator of an important new research undertaking could be ident i f ied .The idea had gradually taken form from many discussions and in t ruth it


30/196

TH E H I G H - S P E E D A I R F O I L PROGRAM 23was a product of the group. At the same t ime there were undoubtedlyinstances where perceptive individuals picked up new ideas from some-one else's off-hand remarks and went on to develop them successfully,perhaps not remember ing where the in i t ia l s t imulat ion had come from.

Frequent references to these lunchroom contacts can be f o u n d . R. T.Jones tells of his first indoctrinations into the mysteries of supersonicflow by Jacobs and Arthur Kant rowi tz in 1935 in "lunchroom conversa-tions" ( r e f . 2 9 ) .

After lunch that first day, Robinson took me on a tour of the varioussections. I have a vivid memory of the 24-inch high-speed tunnel office.Stack and L indsey were working up some test data which Stack discussedwith characteristic intensity and impressive profa nity .

The fol lowing morning Robinson out l ined the N A C A o u t l o o k at thatt ime for high-speed aeronautics, what w as expected of the 8-foot high-speed tunnel, and what part he wanted me to play. He said that it hadbeen determined that about 550 mph was the probable upper l imit ofairplane speeds. Beyond this speed th e occurrence of the compressibilityburble would cause the drag to increase prohibit ively "like throwing outan anchor." Our first job with the new tunne l would be to determine indetail what the high-speed aerodynamic characterist ics for componentsand complete configurations actual ly were . Our next goal would be todevelop improved shapes with higher crit ical speeds so that aircraft couldapp roac h as closely as possible to the ul tim ate lim itin g speed, perh apseven a bit higher than 5 5 0 m p h . W e w o u l d not invent advanced aircraf tbut would provide designers with accurate high-speed component data.

O ur work in the 8-foot tunnel w as necessarily mostly experimentalbecause flow problems involving shocks held l ittle possibility of theoreticalsolution. In effect the tunn el was used as a giant analog co mp uter prod uc-ing specific solu tions to the com plex flow problem s posed by each testmodel . M any other L angley programs generated impo rtant theoreticaladvances, among them airfoil and wing theory, wing flutter, propellernoise, nose-w heel dy nam ics, stability, control, spin ning , compressiblef lows, heat transfer and cooling, and others. L angley's princ ipal theoreti-cians and analysts of the thirt ies inc luded T . Theodorsen, I . E . Garrick,C. K a p l a n , R. T. Jones, B. Pinke l , A. Kantrowi tz , H. J. Allen, S.Katzoff , E. E. L u n d q u i s t , and P. K u h n .


31/196


32/196

THE H IGH -SPEED A IR F O IL PR OGR AM 25mechanical aspects of the operation were supervised meticulously by oneJohnny Hus ton , a sharp- tongued veteran N A C A shop mech anic w hoseemed to relish catching and correcting th e not- inf requent mis takes ofneophyte engineers in the m echa nica l operations of the tunn el . I won deredif my talents would prove worthy of this impressive an d d e m a n d i n gfacility.

T he acceptance testing had to be done late at night when the H a m p t o npower plant w as able to provide us with th e necessary 5500 kw . Airfoi lforce tests and test-section flow surveys were made concurrently withthe motor tests (fig. 2) . In those days the ent i re operat ion w as conductedby one engineer and one mechan ic in the igloo-shaped test chamber.(O ne other engineer involved in the electr ical drive measurements w aspresent only during the acceptance tests in the dr ive equ ipment room.)Dur ing a test, the engineer controlled the tun ne l speed, changed angleof attack, pushed the "print" button for the scales at selected times,recorded visual data readings from the scales, made quick slide rulecalculations of the coefficients, and plotted the results to insure that gooddata were being obtained. (A recent visit to a comparab le NASA tunne ld u r i n g a test ru n revealed a test crew of no less than tw o engineers andtwo engineering aides plus three mechanics, for a similar type of operationexcept that the pre l iminary coefficient plots were produced by an auto-matic computer an d data p lot ter . )O ne night dur ing m y second week on the job just before I closed th eairlock doors at the entrance to the test chamber for a test run, anunusual- looking stranger dressed in hunting clothes came in and stoodthere watching m y preparat ions . Robinson had advised me not to allowvisitors in the test chamber during a high-speed run primarily because th epressure dropped q u i c k l y to about two-thirds of an atmosphere, theequivalent of about 12 000-foot alt i tude. Assuming that th e visitor hadcome in from one of the num erous d uck bl ind s a long Back R iver, Isaid firmly, "I wil l have to ask you to leave now." M aking no move hesaid, "I am Reid," in such ponderous and authoritat ive tones that Iqu i ck l y realized it was Lang ley ' s Eng ineer - in -Charge whom I had notyet met. No one had told m e tha t Re id , w ho l ived only a couple of milesfrom Langley Fie ld , often came out in the evening, especially when testsof electr ica l equipment were being made (he was an electr ical engineer) .


33/196

2 6 T H E H IG H - S P E E D F R O N T IE RWhen I came to know Reid better, the memory of this incident softenedinto proper perspective.

A b o u t a year later at 3:06 a .m. on October 8, 1937, I was r u n n i n g thetunnel at full power and had just promised the operator at the Hamptongenerating plant that I would reduce power gradual ly when, withoutwarnin g, there was a sickening break in the stead y roar of the 55 0-mphwind (ref . 30 ) . A crid smoke rilled the test chamber as I pushed the redemergency stop button, no doubt blowing the safety valves in Hampton.O n entering th e t u n n e l w e found the huge m ul t i -b laded dr ive fan twistedand broken. T he cast a luminum al loy blades had failed in fat igue fromvibrations induced by their passage through th e wakes of the supportstruts. Operations were suspended un til M arch 1938, and the staff wastemporarily dispersed to other sections.

Demonstration runs of the 8-foot tunnel were made for the last of theNACA Annual Engineer ing Conferences , he ld in M a y 1937. Natural ly,we wanted to dramat ize th e compressibili ty burble and to do so wemounted one of the worst ( lowest-cr i t ical-speed) N A C A cowling shapesin the tunnel with a static pressure orifice near the suction peak and atotal-pressure tube on the surface of the cowl afterbody which provideda qual i tat ive indicat ion of drag. There was no way to actually see theshock wave on the cowl, but at Robinson's suggestion, we set up a largechart with a red l ight bulb directly behind a line of small slots at thepart of the cowl drawing where the shock was located. During thedemonstration the tunnel speed was advanced rapidly to the crit icalspeed, about 400 mph. At that point the suction-pressure tube indicatedlocal sonic conditions on the chart . At a slightly higher speed the totalpressure tube showed a dramatic increase in drag and the red l ight wasflashed on ( m a n u a l l y by the tunnel operator) showing th e presence ofth e shock. Runs were made for six groups of visitors on each of the threeconference days and we received many compliments. Orville Wrightand several other pioneers were among the visitors. I had t ime for achat with Alexander Klemin, my college mentor, who perennially re-ported on these N A C A affairs for A ero Digest.T he desire to dramatize compressibili ty effects in that period reachedits peak with our high-speed testing of a model of the DC-3 conf igurat ionin 1938. Although that stolid vehicle cruised at only about 160 m p h , we


34/196

TH E H I G H - S P E E D A I R F O I L P R O G R A M 27tested i t up to 450 mph to show th e speeds at which the various com-ponents, designed without regard for compressibi l i ty, became afflictedwith shock wave problems. T he tests showed th e drag rise for the enginecowls started to develop at speeds as low as about 35 0 m p h . For thefirst t ime we noticed th e adverse effects of interference between com-pon ents; the cr itic al speeds of the cowls and of the wing were red uce dabout 20 mph by the presence of the fuselage ( r e f . 3 1 ) .

T he predicted cr i t ical speeds of a large n u m b e r of existing airfoilsand bodies were determined by Robins on a n d R a y H . W r i g h t from t h e i rlow-speed pressure distr ibut ions as a necessary prelude to the develop-ment of improved shapes (ref . 32) . S tack led the effort in the period1936-1940 to find airfoils with higher cr i t ical speeds, aided by Robinson,L i ndsey , and others. I t was a relat ively s imple mat ter to d e t e r m i n eanalytical ly f r om th in airfoil theory the uni for m- load camber l ines w h i c hwould give th e lowest possible induced local velocities fo r airfoils ofzero thickness. There was no way then, however , to ca lcu la te the opti-m um thickness d is t r ibut io n, and a cut -and-t ry process had to be resortedto. A cons iderable number of system atical ly varied thickness d i s t r i b u t i o n swere analyzed by the Theodorsen method to obta in the theoret icalincompressible pressure distr ibut ion, unt i l one giving a near ly un i for mdistribution w as found . Cur ious ly , it was almost identical to one of theNACA family of airfoils previously defined, th e 0009-45 ( r e f . 2 0 ) .C ombining this thickness d i s t r ibut ion w i th the uniform -load m ean camberl ine gave what was cal led the "16-series" fami ly , the first of the high-cr i t ical-speed low-drag famil ies (r ef . 33 ) . Selected m embers of thefamily were tested at high speeds and firs t reported in the general l i tera-ture in 1943 (re f . 34 ) . (A n extended and improved series of tests w asreported in 1948 ( r e f . 35 ) , and in 1959 tests at transonic speeds up toMach 1.25 were reported ( r e f . 3 6 ) ) .T he 16-series sect ions found immediate acceptance by propellerdesigners, not only because of their high critical speeds but also becauseof their relat ively thick convex shape in the trai l ing edge region whichw as desirable from the s t ructural s tandpoint . A remarkable tes t imony tothese sections was heard at the - N A S A Airfoil Confer ence of M a rc h1978, some 35 years af terward, when a spokesman for propel ler manu-facturers said that the 16-series sections, still used in modern propellers in


35/196

28 THE H I G H - S P E E D F R O N T I E Rthickness ratios from 2 to 10 percent, provided excellent p erform ance .

Although i t was original ly thought that the 16-series sections wouldbe desirable also for high-speed wing applications, it rather quickly w aslearned that they were not suitable. T he problems included: a lowm a x i m u m lift coefficient, a narrow operating range for h ig h M c r, atendency for f low separation in the tra i l ing edge region for the thickersections, and laminar f low characterist ics inferior to the 6-series sectionswhich also had high cr it ical M ach num bers . I t was also found that theuniform-load camber line used in the 16-series family , while it obviouslygave the highest possible critical speed for zero thickness, did not givethe highest possible M cr for finite thickness. Slightly higher M cr couldbe obtained with a camber l ine which concentrated the lift loadingtoward the rear ( r e f . 3 7 ) , but the small advantage is obtained at theexpense of an und esirable rearward shift in center of pressure. A n inter-esting later attempt to develop high-critical-speed sections with largeleading-edge radii and g o o d m a x i m u m lift characteristics w as m a d e byLof t in ( r e f . 38) with some success, but unfortunate ly this program w asterminated in mid-course when NACA management decided to phaseout the airfoil program in the early fifties.In late 1939, we und ertook an un usu al project for H ow ard H ughesth e only pr iva te ly - funded testing ever done in the Langley 8-foot high-speed tunnel. Hughes was represented by his aerodynamics consultant,Col . Virginius E . Clark, an old- t imer in aeronautics and designer of thewell k n o w n "Clark Y " airfoil. Carl Babberger, a former Langley engi-neer, was Hughes' C hief A erod ynam icist and he was also present forth e tests. ( C lar k explained the simple philosophy behind th e "Clark Y "section: it was simply th e thickness d is tr ibu t ion of a Goettingen airfoildeployed above a flat u n d e r s u r f a c e t h e flat feature being highly desir-able in the m a n u f a c t u r e and operation of propellers as a reference surfacefo r applying the protractor to measure or set blade angles. An unhappyproblem in using the C lark Y was the interdep end ence of camber andthickness ratio.)T he most remarkable aspect of this Hughes program, however, w asth e fact that th e test models were not actually representations of theconfigurat ion Hughes was designing. To preserve company secrecy, thetest models had been designed to answer questions relative to nacelle place-


36/196

THE H IGH -SPEED A IR F O IL PROGRAM 29ment, etc., without revealing the real configuration to NACA eng ineers .

T he under ly ing theme for much of our work in the first few years of the8-foot high-speed tunnel was "to provide accurate component data fordesigners." O ften plans for a forthcomi ng test program would inc ludesketching the anticipated data plots in advance, so that r u n n i n g the testseemed more a matter of nicely filling in the data points rather than asearch for anything new. Our Chief of A erodynamics , M r . M i ller, encour-aged this conservative philosophy, telling th e staff at one of the monthlydepartment meetings, "Our aim is to produce good sound research datanothing spectacular, just good sound data." I can provide this quotewith confidence because, even in those days when there was little thoughtgiven to R&D philosophy, agency goals, etc., it provoked some negativereactions among th e more lively members of the staff after th e meeting.

Dr. Lewis had a broader outlook and a willingness to invest occa-sionally in speculative new ideas such as the thrust-augmentor workwhich led to the induct ion dr ive scheme for the first high-speed tunnels.A specific ins tance occurred dur ing a 1938 visit of L e w i s to our officeat the 8-foot tunnel to review recent results and forthcoming test plans.He approved our plans but advised us to "take some shots-in-the-darknow and then."T he Langley of the thirties did not think of itself as a part of thefederal bureaucracy. Broadly d irected by a committee whose distin-guished members served without compensation, and managed by aminuscule Washington office, the Langley operation was spir i tually aswell as physically separated from Washington. The youthful staff hadbeen largely handpicked in one way or another to form an elite groupu n i q u e in the federal system. I t was possible for the entire staff of thissmall organization to become personally acquainted all the way upthrough Lewis, and this resulted in a beneficial sense of family. What-ever their personal foibles, the senior managers, all of whom held careerappointments, were intensely loyal to the organization. They could berelied on for cont inu ing interest in and und erstand ing of our researches,and for cont inuing support and advocacy. These important intangiblesare missing in large agencies whose to p managers come and go at four-year intervals with changing presidential polit ics . T he costly, cripplinginternal friction common in today's large agencies, in the form of


37/196

30 THE H IGH -SPEED F RO NT I ERvoluminous paperwork, repetit ious program reviews and justif ications,lengthy procurements, u n e n d i n g staff meetings, etc. , were virtually non-existent in the L a n g l e y of the thirt ies. W e were also blessed in thosedays with relatively simple research problems which yielded to straight-forward pragmatic research methods. But this happy situation was soonto deteriorate in the enormous expansion and other changes wroughtby World War II.

Crossing th e Atlant ic on the d i r ig ib le Hindenburg in the fall of 1936,Lewis visited Germany and Russia and saw m a n y of their aeronauticalresearch installations. On his re turn he spoke to the L angle y researchstaff in the large room on the second floor of the Engine L ab bui ld ingused fo r such convocations. H is principal impressions were of majorexpansions, especially in Germany. Several large new centers fo r aero-nautical research were under construction, and Lewis was even moreimpressed with the huge new staff, m a n y t imes larger than NACA andpopulated by a larger proportion of advanced-degree holders. He hadlittle or nothing to say, however, about any new aerodynamic or propul-sion concepts or any new research results ( re f . 39 ). H e m ad e a secondsimilar visit to Germany in June 1939 which further impressed himwi th Germany's preparat ions for war. But again he learned little oftheir advanced programs. (The Heinkel He 178, the world 's firstturbojet-powered airplane, was then being readied for its first flightwhich occurred on August 27, 1939.) These Lewis visits to Germanytogether with those of Lindbergh provided th e justifications needed formajor expansions of facilities and staff at L angley star t ing in 1938, andfo r th e establishment of two major new NACA centers at Cleveland,Ohio, and Sunnyvale, Cal ifornia, wel l before December 7, 1941. Sig-nificantly, however, there w as little effect of any of these visits on thenature of our research programs or the problems being tackled priorto the actual start of the war. We were increasingly conscious that awar was coming, but considered all of our existing programs aproposto the improvement of mili tary aircraft .Although there was considerable advocacy of "military preparedness"in the press at that t ime there w as little pressure on us by N A C A m a n -agement to do anything different in character from what we had beendoing. There was no real sense of emergency or war peril to motivate


38/196

TH E HIGH-SPE E D A IR FO IL PRO GRAM 31a search for radical new weapons or bold new concepts in a i rcraf t .Avia t ion had been making rapid progress and the N A C A c o n t r i b u t i o n shad been substant ia l . A lthoug h there was a m ino ri ty group of vocaldetractors, the major i ty opinion was clearly that the United States ledthe world in technical develo pmen t . N A C A bel ieved tha t cont inuedsupremacy could be assured by expansion of its existing programsthrough increases in manp ow er and conv ention al test facilities. M ostN A C A veterans bel ieve that i t w ould hav e been q ui te impossible in thepre-war period to have obtained any major support from the mi l i ta ry ,industry , or from Congress for research and development aimed at suchradical concepts as the turbojet , the rocket engine, or transonic andsupersonic a ircraf t ( ref . 40) .

A noteworthy except ion to the generally conservative pattern w asE . N . Jacobs' investigation of a full-scale Campini sys tem of jet propul-sion in the 1939-1943 period . Ini t ia l ly, Jacobs was motivated more byhis penchant for new ideas than by a sense of war emergency. A greatdeal of effort went into this project , but l ike many hybrid concepts ithad m ajor l imitations, and i t fell by the wayside in 1943, yielding tothe pure turbojet . The Jacobs group harbored a misconcept ion in thisproject which w as shared by the American engine companies at thatt ime; they bel ieved the gas turbine ( turbojet) engine would be imprac-tical for aircraft because of prohibi t ive s tructural weight (refs. 4 1, 4 2 ) .

N ot really in the same category as the C a m p i n i effort but worthy ofspecial note because of its important implicat ions fo r turbojet develop-m e n t was the axial-flow compressor designed and tested in 1938 byE. W. Wasielewski and E. N. Jacobs. In tended for the piston-enginesupercharger application, this machine, designed on the basis of airfoiltheory, developed an efficiency of 87 percent at a pressure ratio of 3.4,a convincing early demonstration of the high performance potential ofthis type. This result is believed to have later influenced Amer icanturbojet designers to favor the axial over the centrifugal compressor( ref . 41) . In teres t ing ly , Jacobs himself w as left with serious doubts aboutthe axial design when the blades of the test machine were destroyeddur ing a run in w h i c h th e compressor stalled. H e believed this might bean inherent w eakness preventin g practical applications (r ef . 4 2 ) . I t issignificant that both this early misconception and the one relating to


39/196

32 THE HIGH-SPEED FRONTIERexcessive turbojet weight involved structural considerations outside thefield of expertise of the Jacobs group.In 1939 R. G. Robinson became an assistant to Lewis in Washingtonand Stack replaced him as head of the 8-foot high-speed tunnel section.Stack's upbringing under Jacobs, plus his natural inclinations, relaxedand enlivened the atmosphere at 8-foot. There was a lessening of theemphasis on data-gathering and chore-doing for industry. There wasalso a pronounced increase in the level of talk, badinage, and practical-joke playing. Although his entire background had been in high-speedairfoils, Stack rather quickly became interested in the other areas of ourworkhigh-speed cowlings, internal f low, interference effects, and air-craft configuration problems.

The war period at NACA has often been described as a time when"fundamental" or "general" research was largely forsaken and replacedby war work fo r specific aircraft . This is inaccurate. Virtually all ofthe general programs underway in 1939, together with their naturalextensions and many new programs as well, were completed during thewar years, subject to occasional delays due to the specific work. Muchof the burden of specific configuration testing fell upon the horde ofnew employees; extended facility test periods were obtained by multipleshifts and the 48-hour week. The involvement in general researchundoubtedly declined on a per capita basis, but in absolute terms mybelief is that it increased.

The long exhausting hours which NACA employees generally aresaid to have put in during the war is another myth. Only a smallminority at Langley worked more than the 48-hour week except forinfrequent stints of additional overtime. O f course there were somenotable exceptions, one of the more interesting occurring in E. N. Jacobs'section. About a year before U.S. entry into the war Jacobs unilaterallyimposed a 48-hour week on his men, with no increase in pay, in orderto expedite their growing programs. He also let it be known that leaverequests were not likely to be approved unless the applicant had putin considerably more than the 48-hour minimum. Surprisingly, therewere only a few protests. The fact that a strong section head couldget away with a high-handed move of this kind implies both patrioticmotivations in the staff and relaxed flexibility in Langley personnel


40/196

THE HIGH-SPE E D A IRFO IL PROGRAM 33operations at that t ime. Such a move would be unth inkable by any federalagency in today 's world .T he inf lux of thousands of new employees during the war periodcaused irreversible changes. The selective standards which had providedth e exceptional talent of the twenties and thirt ies had to be abandoned .Both the quality and the per capita yearly output of reports decl ined.Not a few of the newcomers hinted openly that immunity from th edr af t was the reason they had come. T he increased wind-tunnel testingof specific military designs provided convenient undemanding assign-ments for the less-talented new engineers. T he term "wind-tunnel jockey"was coined d urin g the w ar and is sti l l used to describe inve terate tunn eloperators.

A distinctly pleasant aspect of the large expansion of the 8-foot tunne lstaff was the addit ion of a female computing group. They not only tookover most of the slide-rule work and curve plotting formerly done by theengineers, but also added an interesting social dimension.

T he staff relaxed through all of the usual sports and social events withlittle apparent effect of wartime pressures. Five of us had formed aninformal golfing group consisting of Donald Baals, Henry Fedzuik, CarlKaplan, Stack, and myself. Stack called it the "Greater H a m p t o n R o a d sImprovement Society and Better Golfing Association." I well recall thefirst afternoon w e played at the Y orktow n course . S tack had never playedbefore and had no clubs of his own, but we offered to loan him anold bag with a broken strap and some of our spare clubs. Fedzuik, w howas the chief humorist of the group, had often been th e butt of Stack'spractical jokes and saw here a welcome chance to turn the tables. Withenthusiastic help from some of the rest of us he lined the bottom ofStack's bag with some 10 pounds of sheet lead. We also made sure thebag had a full complement of clubs, and we told Stack that caddieswere used only by the rich and d ecrepi t. By the start of the bac k nin e,with a score card showing well over a hundred in spite of considerablecheating, Stack was seen to start dragging the bag along behind him,his expletives becoming louder and more colorful, and a short time laterhe discovered what had been done. Understandably, he always examinedhis equipment very suspiciously at subsequent sessions.

In mid-1944, Stack was notified that he had been chosen to present


41/196

34 THE H IGH -SPEED F RO NT I ERth e Wright Brothers Lecture of the Inst i tute of Aeronautical Sciencesfo r that year, the first of m an y honors he w as to receive. H e was nowrecognized not only as NACA's lead ing exper t in high-speed aero-dynamics but also as an unusual ly colorful character. This l ec ture ( re f .20) was in essence an updated broader version of Jacobs' Volta paper.T he compressibili ty burble phenomena were i l lustrated and discussed infull detail, results of the systematic efforts to obtain improved componentswith high critical speeds were reviewed, and the stability and controlproblems of advanced aircraft in dives through shock stall were discussedfo r the first t ime in the open literature. S everal of us, par ticu larl y W . F .Lindsey, participated in making new flow photographs and a schlierenmovie of shock-stalled flows in the 4 x 18-inch tunnel which had beenplaced in operation in 1939, sup ersed ing the 11-inch high-speed tunnel .It had higher choking M ach num bers and the 4- inch wid th made itbetter suited for airfoil f low photography. The movie proved to be thehighlight of the lecture. H. L. Dryden, commenting on the talk 20years after his pioneering high-speed tests, said, "We did not under-stand these [high-speed flow-breakdown] results at the t ime [1925] . Thelecturer and his associates have now given us a complete interpreta-tion. . . . The direct shock loss is much smaller than the loss due to[shock-induced] separation (ref . 20)."

In the course of producing these pictures a my sterious oscil lating shockstructure w as observed in the w a k e of a circular cylinder which engen-dered much discussion. Stack dubbed the appari t ion "Yehudi" but thisappellation was edited out of the text. (A m ong other names he coinedwere "Reichenschmutz" for a ducted propulsion scheme, and "Rumble-gut-whiz" for an unsuccessful noise-making device considered by theArmy during the war; i t was to be attached to diving airplanes in thehope of terrifying th e e n e m y . )A figure was included in the lecture emphasizing th e inadequacy ofthe critical Mach number as an index of force breakeither the M achnumbers of force break or the severity of the subsequent changes. Along discussion of "supercritical flows" was included but unfortunate lythis covered only the speed range up to about 0.83, the highest speedat which reliable results could be obtained with the 4 x 18-inch tunnel.Interest in the entire transonic range up to low supersonic speeds was


42/196

TH E H I G H - S P E E D A IR F O IL P R O G R A M 3 5only starting to build up at this time, as a consequence of the excitingnew propu lsion possibilities opened up by the turboje t and rocket engines.Stack's Wrig ht B rothers L ecture brough t to an end the period in w hi chthe true nature of the shock stall had been exposed in detail and theconcept of designing for the highest possible critical speeds to avoidshocks had been fully exploited. For the next decade or more, theemphasis w o u l d be on developing airfoils and wings capable of efficientperformance through the entire transonic speed range. Only the thresholdphenomena had been treated so far , and what happened beyond shockstall in the transonic zone from about M ach 0.8 to up to 1.2 was sti l lunrevealed.C O M M E N T A R Y

In the l ight of later events, NACA's 1936 vision of the "550-mph"propeller-driven piston engine airplan e as the ult im ate goal of high-speedaerona utical research w as obviously too shortsighted and restrictive.Focusing th e effort total ly on the immediate problem of increasing th ecritical Mach number of convent ional aircraft components denied con-sideration of the broader and far more important "barrier" problemareas of transonic f l ight, incl ud in g new prop ulsion concepts, radicalconfigurat ions, t ransonic facilities, etc. A small cadre of the moreimaginat ive thinkers could have been separated from the m a i n effortto provide high-critical-speed data for industry, and encouraged to lookbeyond th e speed range of the existing high-speed tunnels at these"barrier" problems. Even in 1936, it was predictable with certainty thatwith in a few years th e approach of improving th e crit ical speed wouldreach a point of zero re turn , leaving th e barriers still to be reckoned with .T he 550-mph airplane w as achieved in the ear ly forties by the G e r m a n sin the form of the turbojet-propelled M e 2 62, which went into serviceabout the t ime of Stack's Wright Brothers lecture in 1944. In January1945 every airplan e in a 12 -plane A me rican bomber squadron wasdestroyed by M e 262 's . O nly the German failure to produce them inlarge numbers mad e possible cont in ued A ll ied bombing (r ef . 4 1 ) . T heG ermans were also appl yin g variable-sweep ( w i t h outboard pivot loca-tions) to more advanced aircraft as the war te rminated ( re f . 4 3) . These


43/196


44/196

TH E H I G H - S P E E D A I R F O I L P R O G R A M 37increases, permit t ing a continuous increase in power in contrast to thefixed power of the piston engine. This power increase is significantlyaugmented at high speeds by the "ram" pressures of the air whichprovides supercharging and improves the cycle efficiency.W e understood th e principles and enormous potential of the turbojetonly vaguely at the t ime of the XP-59A demonstrat ion. Very l i ttle datawere avai lable to us on the details and per formance of the G.E. 1-16engine. Several of us spent the next few days in exciting speculationsof possible jet-engine thermodynamic cycles, airflow characteristics, andcrude performan ce es t imates , w hich gave us a bet ter unders tanding. K . F .R u b e r t , who had taught in ternal combust ion engines at Cornel l , under-took a more careful systematic analysis, published in 1945 in a paperw h i c h I reviewed as cha i rman of the Lang ley Ed i to r ia l Commit tee ( re f .4 4 ) . (P eriod ic editoria l duties of this kin d were of great valu e as a meansof educa t ion and s t imulat ion of all involvedin addi t ion to their obviousdirect benefit to the q ua lity and accuracy of L angley reports .)

By now our limited goal of the 550-mph subcrit ical airplane of themid-thirt ies had become meaningless and we could foresee the imminentachievement of supersonic f l ight. Few doubted that operational super-sonic military aircraft would soon follow th e research airplanes. T heneed to acq uire ac curate supercr i tica l and t ransonic aerodynam ic data hadbecome acute, and L angley researchers responded to the challenge withconsiderable inventiveness. Eight innovative techniques were eventuallydevised and explored in var ious forms by NACA, ending with genera lacceptance of the semi-open tunnel for two-dimensioned airfoil testingup to M ach 1, and the slotted transonic tunnel for wing and a i rcraf tconfigurat ion testing as the most satisfactory devices. (These develop-ments are listed in the A p p e n d i x and discussed in detail in C h a p t e r I I I . )U n k n o w n to us, the I ta l ians had already succeeded in obtaining air-foil force data through the supercrit ical range up to about M ach 0 .94 ,and the Germans to about 0.92. We first learned of this in 1944before any of the new Langley schemes mater ia l izedupon th e arrivalof A n t o n i o Ferri, formerly of the I tal ian aeronautical labo ratory atG u i d o n i a , and recently an I tal ian Part isan in the war . Ferr i brought withhim extensive airfoil data from their tests in a semi-open high-speedtunnel in the early forties. H e completed analysis of the data at Langley


45/196

38 THE HIGH-SPEED F R O N T I E Rand we published the results i n a N A G A wart ime report ( r e f . 4 5 ) . H isEnglish at that t ime w as negligible, and I wrote the f inal text after muchheated consultat ion with F erri and help from Lou Nucci who acted asin te rpre te r. (M a jo r confusions arose from Ferri's pronounciat ion of"subsonic" and "supersonic," both of which sounded to me like "soup-sonic.") T he proportions of Ferr i ' s tunnel (1 .31 feet across th e opentop and bottom and 1.74 feet on the closed sides) corresponded to 43percent of the perimeter being open. This closely approached th e valueof 46 percent suggested by a theore tical an alysis of Wieselberger inG erm any (re f . 46) as the correct proportion for zero "blockage" (zeroaxial-velocity correct ion) appl icable to a three-dimensional test model.However, this large degree of "openness" had a serious drawback in thelarge pulsations which occurred at high speeds. None of us was quitesure of the validity of the semi-open tun nel te ch niq ue at th at time.

Despite th e quest ions of technique, Ferri 's data revealed a mostimportant new finding: the loss in l i f t associated wi th the compressibilityburble did not persist indefinitely. A t about M ach 0.9 a marked recoveryin lift occu rred , suggesting that the separated ("shoc k-stalled ") flowtended to disapp ear as M ach 1 was approache d. L ater that year supportfor this result was indicated in tests of small wings by means of the"wing- f low" techn iq ue ( re f . 4 7 ) . I n 1946 w e obta ined German airfoildata from their large 2.7-m ete