BILL ROSE

SECRET PROJECTS

FLYING WINGSAND TAILLESS AIRCRAFT

SECRET PROJECTS

FLYING WINGSAND TAILLESS AIRCRAFT

BILL ROSE

MIDLANDAn imprint of

Ian Allan Publishing

SECRET PROJECTS:FLYING WINGS AND TAILLESS AIRCRAFTby Bill Rose

First published 20 I 0

ISBN 9781857803204

All rights reserved. No part of this book may bereproduced or transmitted in any form or by any means,electronic or mechanical, including photocopying,recording, scanning or by any information storageand retrieval system, on the internet or elsewhere,without permission from the Publisher in writing.

Ian Allan Publishing Ltd 2010

Published by Midland Publishing

an imprint of Ian Allan Publishing Ltd, Hersham, SurreyKTI24RG.

Printed in England by Ian Allan Printing Ltd,Hersham, Surrey KT12 4RG.

Mixed SourcesProduct group from weUmlln,gtdIOlnlsandolhtlCOnlro!ledsoUfCtl_h(,O'9 Certno ,SGS-COCOO5S16o 19'J6fo'ellSlll'Wll,dlh,p(ourKil

Visit the Ian Allan Publishing website atwww.ianallanpublishing.com

Distributed in the United States of America and Canadaby BookMasters Distribution Selvices.

CopyrightIllegal copying and selling of publications deprivesauthors, publishers and booksellers of income, withoutwhich there would be no investment in new publications.Unauthorised versions of publications are also likely to beinferior in quality and contain incorrect information.You can help by reporting copyright infringements andacts of piracy to the Publisher or the UK Copyright Selvice.

Photograph on half-title page:The first unmarked XB-35 seen as it is towedacross the runway at Northrop's Hawthorne facility.

orthrop Grumman

Photograph on title page:A B-2A strategic bomber of the 509th Bomb Wingin flight. USAF

ContentsIntroduction 6

Chapter One British Tailless Aircraft 10

Chapter Two German World War 2 Flying Wing Development 34

Chapter Three US Flying Wings (1935-1950) 70Chapter Four US Flying Wings (1950-1990) 95Chapter Five US Manned Tailless Aircraft (1980-2030) 109Chapter Six Soviet Tailless Designs 130

GlossalY 143

Index 144

Secret Projects: Flying Wings and Tailless Aircraft 5

The flying wing is the purest form in aviation.Millions of years before mankind existed itwas evolved by nature as a simple and effi-cient way of transporting seeds over consid-erable distances. When men first nviouslycast their eyes skywards and considered theidea of undertaking flight, birds were the obvi-ous inspiration; however, some pioneersstudied the way that certain trees used thewind to carry aerodynamic seeds to new loca-tions. This would encourage experimentswith flying wing models and large mannedgliders. However, various technical problemsand a resistance to new ideas restrained thedevelopment of powered designs, with rela-tively slow progress being made until WorldWar 2 when engine performance improveddramatically. The flying wing was then con-sidered for the role of advanced fighters andintercontinental bombers.

Th perfect flying wing is a shape reducedto the bare minimum, with few, if any protu-berances, no significant visible fus lage sec-tion and the efficient use of internal space.This type of aircraft promises high lift and lowdrag, although stability and handling werechallenges for early designers. Subsequently,many flying wing aircraft were aerodynamiccompromises. The early flying wings requiredvertical stabilisers for effective control, whilefuselage sections spread beyond the wing'sboundaries, and in some cases, engine pods

Introduction

were added. Becau e of this, many so-calledflying wings were more accurately defined as'tailless aircraft' and the wing could be sweptback, swept forward, swivelling in an obliquemanner, traight, of variable geometry, ordelta-shaped.

Typically, the flying wing and closelyrelated configurations are unsuitable fordevelopment as a supersonic aircraftbecause of their large wingspans. Neverthe-less, there are some interesting exceptions tothe rule and scope for supersonic perfor-mance with certain highly-swept taillessdesigns. That aside, the flying wing is poten-tially a very efficient aerodynamic configura-tion for carrying substantial payloads atsubsonic speeds over long distances.

Some historians will rightly attribute theearliest designs for a manned flying wing toLeonardo da Vinci (1452-1519) who pro-duced a detailed drawing of a man-poweredornithopter (flapping wing device) around1485. The de ign was little more than a fanci-ful idea, but intere t in the possibility ofmanned flight gath red momentum duringthe following centuries, eventually leading toexperiments with balloon and mall hand-launched gliding models.

One particularly interesting idea thatstemmed from a simple paper dart was a fly-ing wing design produced by James WilliamButler and Edmund Edwards in 1867. They

envisaged a scaled-up man-carrying aircraft,powered by some form of steam propulsionand patented the design, although manya pect of the concept were ahead of its time.This was followed by the first demonstrationof sustained flight when Alphonse Penaud(1850-1880) flew a rubber band-poweredmodel for a distance of 131 ft (40m) on 18August 1871 in Paris. Having shown that pow-ered flight was possible, Penaud went on toconceive a series of astonishingly advancedflying wing aircraft that were never flown, butset th standard for all future de igns. Penaudwas succeeded by the aviation pioneer LouisPierre Mouillard (1834-1897) who, it isbelieved, made a short flight of 90ft (27m) atan altitude of 30ft (9m) in his fourth glidernear Cairo, Egypt, on 3 JanualY 1878.

While Otto Lilienthal (1848-1896) is oftencredited with flying the first succes fulmanned glider in 1891, it is widely claimedthat Mouillard and Lilienthal were followingin the foot teps of Engli h inventor Sir GeorgeCayley (1773-1857). He designed and builtseveral gliders, which were reportedly te tedby an unknown 10-year-old boy and Cayley'scoachman on several occasions between1799 and 1853. Whether or not Cayleydeserves to be credited with the first mannedglider flight remains open to debate. Anotherearly pioneer was Igo Etrich (1879-1967) whobuilt gliders based on the winged liana seedand this work proved to be influential to manyfuture designers. The application of an inter-nal combustion engine to a tailless designwas then adopted by pioneers such as Frenchengineer Rene Arnoux and British designerJohn William Dunne (1875-1949).

World War 1saw vast improvements in avi-ation technology with aircraft becoming avaluable military asset. After ho tilitiesceased in 1918, German engineer HugoJunkers turned his attention towards buildingan efficient aircraft known as the JG-1 that

The German glider pioneer Olio Lilienthalundertaking a short flight at Berlin on anunspecified date in 1896. He experienced manyaccidents during these trials and died as a resultof serious injuries sustained in a crash on 9 August1896. His last words were 'Sacrifices must bemade'. Bill Rose Collection

would seat passengers within the wing.Unfortunately for Junkers, the design wasdeemed to have contravened conditions ofaircraft construction set out by the Allies andthe JG- I was therefore scrapped. Junkers hadpatented a flying wing aircraft in 1910 andwas a trong supporter of this configuration,envisaging future giant transporters capableof carrying 1,000 passengers.

Although such a design never materialised,there would be significant progress made inGermany during the following decades byAlexander Lippisch and the Horten brotherswho were re ponsible for a number of highlysignificant tailles designs. World War 2 sawmajor developments of their work with Lip-pisch and the Horten designing advanced jetand rocket-powered aircraft that pushedtowards the boundaries of supersonic flight.In North Am rica, the main advocate anddeveloper of the flying wing was Jack (John)Northrop, although it should be noted thatVincent Burnelli (1895-1964), who is oftenoverlooked, made great strides with liftingbody designs.

At one stage during World War 2, Americanstrategists believed that Britain would fall toGermany and Jack orthrop was given thetask of designing a new US Army Air Corpsbomber capable of reaching Europe. Thiproject led to construction of the futuristicB-35 flying wing; however, by the time it flew,the propeller-driven warplane was virtuallyobsolete. The design was re-engin ered intothe jet-powered B-49, but the new Northropaircraft was t chnically flawed and could notcompete with the relatively straightforwardB-47 produced by Boeing. Subsequently, theB-49 never became operational and interestin flying wings sharply declined.

While designers continued to study the fly-ing wing, no significant aircraft of this typewere built for many years, although Britishpost-war interest in tailless aircraft led to thesuccessful delta-wing Avro Vulcan bomber.By the 1970s, advances in computer technol-ogy were making the flight control of unusualdesigns a reality and scientists recognised thepotential of the flying wing as an aircraft wilhan exceptionally low radar cross section.

The Burgess-Dunne Flying Wing Biplane of 1914.Bill Rose Collection

'Le Stablavion' (stable aircraft) was built by Frenchconstructor Rene Arnoux and demonstrated to thepublic in 1913. Powered by a 55hp (41 kW) Chenupiston engine, the performance and handling weresaid to be disappointing. Bill Rose Collection

Featuring a pusher configuration, Rene Arnoux'stwo-seat flying wing monoplane was demonstratedin Paris during 1914. Bill Rose Collection

6 Secret Projects: Flying Wings and Tailless Aircraft Secret Projects: Flying Wings and Tailless Aircraft 7

I -..

An illustration of the experimental Northrop NIMflying wing during a test flight. Bill RoselNorthrop Grumman

The futuristic Vickers Swallow supersonic nuclearstrike aircraft was designed in the 1950s and isshown resplendent in RAF markings and anti-flashwhite finish. The Swallow was too far ahead of itstime to be a realistic proposition.Bill Rose Collection

Northrop Grumman's proposal for the USAF's nextgeneration bomber expected to enter service in2018. Northrop Grumman/Bill Rose Collection

handling problems for civil versions andfew airports could accommodate BWBs atpresent.

Because there are many grey areasbetween true flying wings, tailless aircraft,BWB designs and deltas, I have tried to avoidsetting definite limits on what kind of aircraftare suitable for inclusion in this book. This hasproved to be quite a challenge, as it is oftenunclear where to draw the line despite con-siderable consultation with colleagues. As anexample, it now appears acceptable todescribe an aircraft in technical papers as a'tailed' flying wing and some authors regarddesigns such as the Mirage or Concorde asbelonging in the flying wing category. There-fore, I have set my own guidelines and haveincluded various interesting tailless aircraftranging from the British World War 2 experi-mental GAL gliders and the German Me 163rocket fighter to the advanced variable geom-etry of the Vickers Swallow and exoticoblique wing designs.

This book mainly deals with manned air-craft as the detailed additional coveragerequired for unmanned tailless designswould at least double the space and thereare practical limits in producing such a publi-cation. I have not attempted to write anexhaustive technical appraisal of the flyingwing or tailless design and this book is not adefinitive reference work containing exam-ples of every known design. That said, I havetried to provide background information onmany - often lesser-known - military con-cepts and designs that have (in a few cases)been built as prototypes and occasionallyreached production.

Many individuals provided assistance withbackground material for this book, but Iwould like to especially thank Chris Gibsonand Tony Buttler, Dr lain Murray, RobertBradley of the San Diego Air & SpaceMuseum, Martin Muller, Alexi Malinovsky andTony Chong of Northrop-Grumman.

Bill Rose, Norfolk, EnglandFebruary 2010

B-49 and Avro Vulcan proved hard to delectwith prevailing ground-based radar syslems.An ability to hide aircraft from enemy delec-tion would have many advanlages and thisresulted in highly classified US research dur-ing the 1970s 10 produce a low-visibility com-bal aeroplane. This eventually resulted in theLockheed F-117A that was developed from adiamond shape, and although it wasdescribed as a fighter, was designed as a spe-cialised subsonic attack aircraft that wasincapable of air-to-air combat. The ability tolaunch a small-scale surprise attack was obvi-ous, but the F-117A was not expected to farewell in a major confrontation with the SovietUnion, whose air defence capabilitiesremained significant and were under con-stant revision. Therefore, the USAF required anexl-generation bomber with an interconti-nental range, a substantial payload capabilityand relative invisibility to all existing andanticipated radar systems. This led to propos-als for an advanced bomber that resulted inthe Northrop-Grumman B-2A Spirit making itsfirst flight in 1986. Developed and built in greatsecrecy, the B-2A strategic bomber becameto be the most expensive combat aircraft inhistory, with its eventual use in small regionalconflicts widely regarded as overkill.

Nevertheless, the B-2A remains a technicalmasterpiece and is a true flying wing aircraftin every respect, although its shape wasselected for reasons of low visibility and thesleek appearance was almost a by-product.The US Navy also sought a new carrier-borne,multi-role combat aircraft with a stealth capa-bility and this led to the A-I2AAvenger II, a tri-angular-shaped flying wing developed byGeneral Dynamics and McDonnell Douglas.In some respects, this aircraft was moreadvanced than the B-2A. However, spirallingdevelopment costs brought the A-I2A pro-gramme to an abrupt halt.

A new US manned bomber is currently inits initial development phase and is widelyvisualised as a second-generation B-2. How-ever, the main application for low-observabletechnology at the present time is for increas-ingly sophisticated remote-control aircraftthat often feature flying wing configurations.Further tailless aircraft seem inevitable asdesigners look towards aerodynamically effi-cient blended wing body CBWB) configura-tions to satisfy demands for new tacticalairlifters, tankers and airliners. These aircraftare likely to become a common sight in ourskies during the coming decades. BWBs willnot be capable of higher speeds than existingaircraft in the same class, but will transportgreater payloads over longer ranges at lowercost. On the down side, there may be ground

Fig. I.

Hugo Junker's influential 1910 patent for a newtype of aircraft that would contain its mostimportant features inside the wing.ESP Patent Office

Engineer, aviation pioneer and founder of thefamous aircraft manufacturing concern,Hugo Junkers was an advocate of the flying wingand did much to secure its widespread acceptance.Bill Rose Collection

It had been apparent that flying wings wereharder to detect with radar than conventionalaircraft and this was believed to be a featureof the Horten jet fighter that first flew at theend of World War 2. It is now suggested thatany stealth qualities the Horten jet fighter pos-sessed were more accidental than inten-tional, even when considering the significantadvances in radar countermeasures made byGerman scientists during the war.

It has been reported on several occasionsthat post-war aircraft such as the Northrop

Fig. 5

DlUfSCHf'-' PFICH

~EICIiSI'.-\lfNTA~lf

F(II(.6.

- Jft. 253788 -nASSE 77 h. GRUPPE :l.

Pat.alStrt I. Otll"("" R.le", "011 1. fdft" 1910 d.

PATENTSCHRIFT

HUGO JL;NKERS " AACHEN-FRANKENBliRC... Gleitflieger mit lur AurnahtDe yon nichl Auftrieb .n.u~enden Tellu dieJlenden

Koblk6rper.

r;;"""'@?

Chapter One

Baynes Bat

After leaving chool, Leslie Everett Baynes(1902-1989) became immediately involvedwith lh aviation industry. He worked for lheAircraft Engineering Company in Hendon,

orth London, became a designer for ShortBrothers at Rochester and then d vised thevery succe ful cud I light sailplane. This ledhim into a manufacturing partner hip withF.D. Abbot, forming the company Abbot-Baynes Sailplanes Ltd of Farnham, Surrey. In1935, a powered version of the Scud was builtto meet a request made by Sir John Cardenwho ran a company developing armouredvehicles. As a consequence of this order, anew partnership was formed, called Carden-Baynes Aircraft, with a factory located at He -ton, Middles x wh re powered aircraft weremanufactured. Baynes and Carden began

DO DO DO:1f,",:-,'r' "-,, lJ "\: c,:-"",'. 't}' ldr -,

-----1--1--- -- -- -- -- -- -- -- ----- ----.

controlled machine gun turrets weremounted outboard of the propellers on eachtrailing edge and a single forward firing gunturret was positioned on the leading edge, tothe port side of centre. The control surfacesappear similar to the larger de ign.

These designs were undoubtedly submit-ted for official consideration, but went no fur-ther as they were probably considered far toounconventional. Undeterred, Professor Hillcontinued to work on the concept, which henow envisaged as a large trans-Atlantic air-liner suitable for post-war operation Thisproposal was known as the PterodactylMkVlll and was to be powered by five Rolls-Royce Griffon engines driving pusher pro-pellers. The airliner was to have a cockpitsection protruding from the central leadingedge. Despite interesl from Short Brothers,nothing came of this project.

A proposed six-engine, three contra-propPterodactyl flying wing bomber, with an estimatedweight of 85,000 Ib (38,555kg). Bill Rose Collection

Professor Hill's wartime proposal for a four-engine, propeller-driven flying \\1ng PterodactylBomber. Bill Rose Collection

also considered and Hill believed this designcould easily be modified when jet propulsionbecam available. Estimates of performanceare not quoted in any available documenta-tion. The Pterodactyl bomber would be sup-ported on a rather unusual fully retractableundercarriage consisting of four larg mainwheels, positioned across the underside ofthe wing at the centre of gravity and a singlenose-wheel. During take-offs and landings,the pilot would sit in a capsule locatedtowards the centre of the aircraft. This wouldbe exlended below the wing to afford thepilot a better view of the runway and it wouldbe possibl to use the front wheel as a posi-tioning aid. For normal night, the pilot wouldsit behind a windshi ld in the leading edge ofthe wing, positioned slightly starboard of cen-tre. Hill also planned to seat the navigatoralongside the pilot at the front of the aircraft,with extra windows allowing downwardobservation.

The bomber would be equipped with fourremote-conlrolled machine gun turrets, alloutboard of the engine bays and propellers.Details of bomb carriage are unknown,although it seems certain that at least twoseparate bays w re envisaged on either sideof the cabin area. There were no vertical con-trol surfaces, although in addition to elevonsand naps, Hill proposed movable wingtips.One development of this design was an air-craft carrier, able to transport up lo thr e pis-ton-engine fighters (presumably Spitfires orHurricanes) on the upp I' wing. This wouldmake il possible to operale at ranges muchgreater than might normally be available. Pre-sumably the fighters would return to baseunder their own power, or the pilot wouldbailout at a pre-arranged location.

Hill also designed a smaller version of thePterodactyl bomber with a wingspan ofapproximately 100ft (30Am) and a length ofabout 38ft (ll.5m). It would have an antici-pated weight of approximately 45,0001b(20,411 kg). This design differed in having fourseparate unspecified piston engine , eachdriving a tail-mounted propeller. The fUlly-retractable undercarriage would be slightlysimpler than the larger bomber, using threeseparate main wheels and a single nose-wheel. It appears that Professor Hill did notconsider th retractable pilot's cap ule nec-essary for this aircraft. Rear facing remote-

Designed by Geoffrey Hill, the MkV Pterodactylwas a military variant of this pre-war series ofaircraft developed to meet Specification F.3/32.A sesquiplane in configuration with a small,unsweptlower wing, the Pterodactyl MkV was filledwith two forward-firing machine guns and carrieda rear gunner who enjoyed an unrestricted view.Bill Rose Collection

The Pterodactyl BombersIn 1939, Professor Hill was recruited by the AirMinistry to oversee various specialised mili-lary research projects and his first assignmentinvolved a programme to assess various waysof allowing an aircraft to cut enemy barrageballoons' cables without slalling. In additionlo a range of consultative work, Hill becamethe chief cientific liaison officer between theBritish and Canadian governments in 1942.At the tart of this period, he began a privatedesign project for an advan ed taillessbomber with initial plans being completed on22 July 1942.

Hill's fir t proposal was a boomerang-shaped nying wing with a straight c ntralleading edge and the outer wing sectionswept (at the leading edge) by approximately44. The wingspan was 140ft (42.6m) with awing area of 3,350ft' (311 m') and the overalllength appears to have been around 50ft(15.24m). The wing would have a centraldepth of 7ft 6in (2.28m), making a substantialpressurised crew compartment possible.Quoled weight is 85,0001b (38,555kg) andwas presumably an all-up estimate.

Hill envisaged six forward-posilioned pis-ton engines mounted in side-by- ide pairs,driving three rear-mounted pusher configura-tion contra-rotating propeller assemblie vialengthy drive shafts. The use of turbojet was

bomb load. The prototype was ready to ny in1933. During tests the performance was saidto b comparable to a Hawker Demon andlongitudinal stability was certainly superior to

arlier Pterodactyl designs. Despite itspromise, the MkV failed to win support for fur-ther development. Although disappointed,Hill ontinued to work on the Pterodactyldesign producing further tailless aircraft con-cepts, but the project was effectively at anend.

British Tailless AircraftCherub engine and new towards the end of1925. In 1926, Hill joined Westland Aircraft,which secured Air Ministry backing for furtherdevelopment of the Pterodactyl design andfiled a patent for a wing adjustment design in1930

A series of small aircraft followed with Hilldesigning a two-s at fighter version of the air-craft for Specification F3/32, known as theMkV. Construction began in 1932 with duralu-min and steel being us d for the majority ofthe fuselage framework and main wing.From the outset, Hill decided to add short,unswept lower wings for strengthening pur-poses, making the MkV a sesquiplane. Theundercarriage consisted of tandem wheelslocated beneath the centre of the fuselageand skids below each lower wingtip withsmall trailing wheels. A very troublesomeROils-Royce Goshawk I engine in a forwardmounted position powered the MkV, provid-ing a maximum speed in level night of around190mph (306kph) and a service ceiling of30,000ft (9,144m). When empty, the MkVweighed 3,5341b (1,602kg) and around5, I 00 lb (2,313kg) when fUlly loaded. Thewingspan was 46ft 8in (14.2m) with a wingarea of 396ft' (36.78m') and the overall lengthwas 20ft 6in (6.24m). As a fighter, the aircraftwould be equipped with several machineguns and was capable of carrying a small

Geoffrey Hill

In late 1942, the K Director of ScientificResearch decided to form the Taille sAircraftAdvisory ommittee (TAAC) to over eedevelopment of new designs. German air-raft designers were known to be int rested

in developing tailless military aircraft andconsiderable research was also taking placin America. orthrop was working on a long-range, nying wing bomber and the ationalAdvisory Committee for Aeronautics (NACA-the forerunner of ASA) was underlakingextensive research into laminar now aero-foils. Therefore, the British hoped to takeadvantage of US developmenls for a future jetpropelled tailless bomber.

Tailless aircraft were nothing new in Britainwith the early pioneering work of JohnWilliam Dunne inspiring more advanceddesigns by Geoffrey Terence Roland Hill(1895-1955). Hill had become preoccupiedwith the problem of stalling and this led himto design the Pterodactyl Mk 1, a nying wingwith a 31 0 leading edge sweep and all-movingwingtips. The aircraft was built during an 18-month period with Air Mini try support andn was a glider in December 1924. It was thenequipp d with a 33hp (24.6kW) Bristol

10 Secret Projects: Flying Wings and Tailless Aircraft British Tailless Aircraft II

Mn

Ft 10

~~-Siiiltri"T~-~13

The quite distinctive central fuselage sec-tion accommodated a two-man crew withthe flight observer sitting directly behind thepilot and facing rearward. The Manx wasequipped with a tricycle undercarriage withthe rear wheels able to retract, but the nosewheel fixed. This was normally fitted with astreamlined fairing to reduce drag. Taxiing tri-als began at Radlett in late 1940, althoughtests became long and drawn out, with morepressing wartime work often taking priority.

The company's senior test pilot, JamesRichard Talbot (1909-1945), was nowinvolved with these trials, along with EdgarAlexander 'Ginger' Wright (1914-1945), a Han-dley Page test pilot who headed theObserver's Office. On 12 September 1942, dur-ing high speed taxiing trials, the Manx becamemomentarily airborne, achieving a height ofabout 12ft (3.65m), but after touchdown, itwas evident that the nose-wheel had beendamaged, so repairs and modifications weremade. Further attempts to fly the Manx fol-lowed and it finally became airborne on IIJune 1943, with Talbot at the controls. How-ever, the cockpit canopy became detachedsoon after take-off and the flight was immedi-ately terminated. As trials continued on anintermittent basis, the renowned glider pilotRobert Kronfeld was invited to test fly theManx. Kronfeld was now an RAF SquadronLeader, assigned to General Aircraft Ltd (GAL)ofFeltham, Middlesex, which was working onits own experimental tailless glider project.

General arrangement of the taillessHP.75 Manx experimental aircraft.Chris Gibson

found to be substantially overweight andthere were serious problems with the type ofglue used during assembly. As a conse-quence, the aircraft underwent a programmeof modifications and the main spar waslargely rebuilt. It is possible that this had somebearing on the fact that Dart Aircraft hadceased trading by the end of 1939.

Built mostly from wood, this aircraftreceived the semi-official name Manx (afterthe tailless cat). It has been claimed that thecompany designation HP.75 may not havebeen applied until 1945, although the officialregistration H-0222 was issued at an earlystage in this project. The Manx was fitted withtwo rear-mounted de Havilland Gipsy Major IIfour-cylinder piston engines, each rated at140hp (l04kW) and fitted with two blade pro-pellers acting as pushers. The overallwingspan was 39ft lOin (l2.lm), wing areawas 245ft' (22.76m2), the length was 18ft 3in(5.5m) and the empty weight was 3,0001b(I ,360kg). Maximum take-offweight was esti-mated at 4,0001b (I,814kg) allowing a maxi-mum speed of approximately 146mph(234kph) and a service ceiling of about 10-12,000ft (3,000-3,650m). The central sectionof the wing on either side of the cockpit wasstraight and the outer section was swept backby 35 with elevons and leading edge slots.An upright fin and rudder was attached toeach wingtip and an additional fin waslocated on the upper rear of the fuselage toenhance flight control.

Melrose 'Winkle' Brown was less enthusias-tic about the Baynes BAT when he tested it,finding the controls disappointing.

Why this difference of opinion arose is hardto say, but the tests were still judged success-[ul. Unfortunately, there were ongoing prob-lems concerning tank suitability and it hadI> en realised that landing faster than thetank's maximum speed was a serious issue.I'his led to the idea of fitting an undercarriageto the BAT, but it was considered over-com-pi x and the Baynes BAT was then movedonto the back burner. At the same time, aheavy glider was being developed by Generalo\ircraft called the GAL.49 Hamilcar, whichhad been designed from the outset to carry asubstantial payload that included a Tetrarchor the US-designed M22 Locust light tank.

While the Hamilcar eventually entered ser-vice, the Baynes BAT never progressed to ahill-sized carrier wing glider. It was thentransferred to the Royal Aircraft Establish-III nt at Farnborough, which made good useo[ it for stability tests that continued after thewar. In late 1946, the Baynes BAT was for-lI\ally disposed of as surplus equipment,Il(,jng purchased by British Light Aircars ofHl'dhill, Surrey. It seems that the BAT wasthen moved to Croydon Airport where it was1,Ist seen behind a hangar in 1958, apparentlyIII poor condition. Soon after this, the gliderwa scrapped. Baynes, known to his friends

~I \ '..'\ I

.........

Top: The third GAL56 glider 'Maximum V' (TS513),which made its maiden flight on 30 May 1947at Lasham. Bill Rose Collection

Above: Three-view drawing of the GAL56'Medium V'. Bill Rose Collection

Below: The GAL56 'Medium U' glider (TS510)piloted by Fit Lt Robert Kronfeld, after releasefrom a Halifax tow aircraft. Bill Rose Collection

with it, describing it as very difficult to controland noting that he 'could not relax for a sec-ond, beginning right away with takeoff.

The second aircraft known as GAL56/04(TS51O) was designated 'Medium U'. It dif-fered from the first aircraft in having a parallelchord centre section and swept outer leadingedge of 28.4. The wingspan was 51ft(I5.54m, height 21ft (6.4) and length lOft

245ft 4in (I3.8m)28.4' (leading edge)317ft' (29.45m')350ft' (32.5m')5.819ft (5.79m)8ft 9in (2.66m)RAF 34 Modified11ft 4in (3.45m)43~in (1.1 m)0.5'

4,4001b (I,996kg)58mph. E.A.S. (93.3kph)150mph. E.A.S. (24Ikph)

2Pilot and space for prone observer51ft Ilin (15.8m)36.4' (leading edge)24ft 8in (7.5m)5, 1741b (2,348kg)

Crew

WingspanWing sweepLengthFlying weight

GAL61

CrewWingspanWing sweepWing area (net)Wing area (gross)Aspect ratioLengthHeightAerofoil sectionChord-rootChord-tipDihedralnormalFlying weight(crew +max ballast)

Stalling speedMax towing speed

GAL56 'Medium V'

the wingtips and it was decided to add anti-spin parachutes in wingtip containers foremergency use.

No overall name was assigned to the glider,but the models were referred to by their wingconfiguration. The first to be completed dur-ing autumn 1944 was the GAL56/01 (TS507)'Medium V' which identified its 28.4 leadingedge sweep. An RAF-34 aerofoil section wasused and it had a modified contour toenhance the elevon performance. Thewingspan was 45ft 4in (I3.8m), length was19ft (5.79m) and it was 8ft 9in (2.66m) high.The aircraft undertook taxiing trials at Alder-maston during November 1944 and made itsfirst flight on 17 January 1945 at RAF DunholmLodge in Lincolnshire. This location had beenassigned to GAL as a facility to store Hamilcarheavy gliders and it was chosen for testingbecause the GAL56 was a secret project andthere was less chance of the trials beingobserved by anyone.

For the flight at Dunholm Lodge, an Arm-strong Whitley V bomber was used to tow theglider into the air and test pilot Robert Kron-feld was at the GAL56's controls. He reportedthat the aircraft handled fairly well except inextreme conditions and after making 48flights at Dunholme Lodge, TS507 was trans-ferred to RAE Farnborough in June 1945. How-ever, when RAE test pilot Capt Eric Brownflew the GAL56, he was far from impressed

GAL Gliders

Soon after it became established at the end of1942, the Tailless Aircraft Advisory Committee(TAAC) invited a number of aircraft contractorsto submit plans for experimental tailless air-craft that could be used for research purposes.Many companies (including de Havilland,Saunders Roe, Hawker and Fairey) declined totake part, leaving Armstrong Whitworth andGeneral Aircraft Ltd (GAL) as the principal par-ticipants, followed by Handley Page which hadalready built the Manx. Designers at GAL hadbeen interested in developing tailless aircraftsince the mid-1930s and proposals were sub-mitted to the TAAC, which swiftly approvedseveral swept-wing prototypes that would beused to test low-speed handling characteris-tics. Initially, it was planned to build six proto-types, but this was reduced to four. Twowould be gliders designated GAL.56 and twopowered versions referred to as GAL57, usinga single Lycoming R-680-13 radial engine in apusher configuration.

Construction would have to be cost effec-tive, utilising common parts wherever possi-ble. This was necessary to avoid diverting toomany resources away from war production.It was decided from an early stage to use astandard centre section with commonattachment points for wing anchorage. How-ever, it became apparent that there would beproblems interchanging wings with differentprofiles on a standard body and a decisionwas taken to abandon the powered versionsand build four different gliders, taking thesimplest route to keep the project on course.It was also agreed that GAL test pilots andobservers would conduct most of the initialtest flight. Refinement and construction werethen passed to an engineering team at theHanworth factory, Feltham, Middlesex,which was headed by a Czech engineercalled Otto Wels (no connection to the pre-war German politician), who specialised inflight control systems.

Each central section utilised a tubular steelframework covered in plywood and thewings were mainly of wooden construction,supported by a single spar. Off-the-shelf land-ing gear was used for the GAL56, with themain wheels being Dowty units designed forthe Lysander and the tail wheel normallyused on a Bisley. The serial numbers TS507,TS510, TS513 and TS515 were issued on 21October 1944, although it was finally decidednot to proceed with construction of TS515.Each GAL56 accommodated a pilot andobserver in the centre section, with the maindifference between each model being thewing profile. Fins with rudders were fitted to

239ft lOin (I2.lm)245ft' (22.76m')18ft 3in (5.5m)3,0001b (I,360kg)4,0001b (I ,814kg)Two de Havilland Gipsy Major IIfour-cylinder in-line piston engineseach rated at l40hp (I04kW)146mph (234kph)I0-12,000ft (3,000-3,650m)

Below: The HP.75 Manx in flight. Bill Rose Collection

Above: Receiving the semi-official name of Manx(after the tailless cat), the experimental HP.75 wasmainly built from wood and powered by two rear-mounted de Havilland Gipsy Major II four-cylinderpiston engines. Bill Rose Collection

Maximum speedCeiling

HP.75 Manx

CrewWingspanWing areaLengthEmpty weightGross weightPowerplant

He made at least two flights in the Manxand reported that the aircraft handled well incomparison to other tailless designs he hadflown. By late 1945, the Manx had logged upto 30 test flights lasting for a total of around 17hours. Talbot and Wright had flown many ofthe missions, but both were killed on 3December 1945 when a prototype Hermes Iairliner they were flying crashed shortly afterleaving Radlett Aerodrome. This loss had adirect impact on the Manx programme andthe aircraft made one more test flight in April1946 and was then placed in storage.

The Manx was an interesting design, whichwas briefly considered for development as atwin-engine transport aircraft, or possibly alight bomber, with a canard configuration.Nothing came of this and the Manx remainedin storage until 1952 when it was broken upfor scrap.

14 Secret Projects: Flying Wings and Tailless Aircraft British Tailless Aircraft 15

6120ft (36.6m), later reduced to112ft 6in (34.3m)2,000.04fl' (1858m')45ft (13.7m)49,7651b (22,573kg)

470mph (756kph)480mph (772kph)3,000fVper min (9 I4m/per min)Unknown1,500 miles (2,414km)4xMetropolitanVickers(Metrovick) F.2/4 Beryl axialflowturbojet engines, initiallyproducing 3,500 Ib (155kN)static thrustProvisionally, 2x20mm forwardfiring cannons and one rear firingremote controlled turret with2x20mm cannons.12,000 Ib (5,443kg)

CrewWingspan

Armstrong Whitworth AW.50

Armament

Wing areaLengthGross weightMaximum speedat sea levelat 30,000ft (9, 144m)

Rate of climbCeilingRangePowerplant

Bomb load

cated from Plymax (a thin sheet of duraluminbonded to a thicker sheet of plywood). Thisallowed a very smooth finish that met therequirements for effective laminar flow. Thecentral cockpit area accommodated a pilot

early design phase, but engine problemsappear to have ruled this out.

However, it was decided that a one-third-scale size glider should be built to test theaerodynamics, as there was little data to drawon regarding the behaviour of flying wing air-craft at low speeds or in unusual attitudes. Ini-tially assigned the company reference AW.51,design work started in May 1943, with the pro-ject altering to AW.52, which finally becameAW.52G, with the last letter referring to its sta-tus as a glider.

Assembly started at the beginning of March1944, although proceedings were relativelyslow as priority was being given to morepressing war production work. Wind tunneltesting continued and AW.52G was finallycompleted in early 1945. It was then assignedthe official registration RG324, but by thistime, the original flying wing bomber hadbeen cancelled. Nevertheless, AW.52G wasseen as a useful research tool and on 2 March1945 was towed into the air behind a Whitleybomber. This first flight was undertaken bythe company's test pilot Charles K. Turner-Hughes, who reported no unexpected han-dling difficulties.

Spruce and plywood (used for the singlebox spar and ribs) were utilised in much ofthe construction, with the outer skin fabri-

Right: One of John Lloyd's early design for anexperimental twin-jet aircraft, which appearedin a UK Patent (2,474,685) filed during 1944and released in 1949. This would eventually leadto the AW.52. Bill Rose Collection/UK Patent Office

Bottom: AW.52G is towed into the air during a testflight. Bill Rose Collection

work, which was supported by the Ministry ofAircraft Production (MAP). Lloyd envisagedthe use of a smooth surface for the aircraft'sskin, fabricated from a fairly thick alloy,strengthened by rolled corrugations of lightergauge metal to maintain adequate rigidity. Atthe centre of the aircraft was a cockpit afford-ing good visibility, with a slightly raised crewsection and a tail-mounted, rear-facing,remotely controlled turret containing two20mm cannons. In addition, it was also pro-posed to equip the aircraft with two forward-facing 20mm cannons located in the wingsbetween each engine.

Propulsion would be provided by fouradvanced Metropolitan-Vickers (Metrovick)F.2/4 Beryl axial flow turbojet engines. Thisdesign was producing 3,500 Ib (15.5kN) of staticthrust by 1944, but the F.2/4 remained too unre-liable for use in any production aircraft. Lloydhoped that the ongoing problems with the F.2/4would be largely resolved by the time theAW.50 was ready for production and improvedthrust might be available. The engines wouldbe located within the wings on each side of thecockpit drawing air from four separate intakesalong the leading edge of the wing.

Split-flaps would be used, plus wingtip finsequipped with rudders. The AW.50 was to beequipped with a fully retractable tricycleundercarriage. Lloyd's initial plans were foran aircraft with a wingspan of 120ft (36.6m), awing area of 2,000ft' (185.8m') and an overalllength of 45ft (13/7m). The estimated maxi-mum speed was 470mph (756kph) at sealevel and 480mph (772kph) at 30,000ft(9, 144m), with a rate of climb slightly betterthan 3,000fVper min (914m/per min). Theanticipated service ceiling is unknown, butrange was expected to be 1,500 miles(2,424km). Gross weight was 49,7651b(22,573kg) with the bomb load accounting for12,0001b (5,443kg) of this figure. However,within months, the design underwent a sig-nificant revision, with the wingspan beingreduced to 112ft 6in (34.3m) and the wingbecoming thinner. As a consequence, thecentre section became recogni able as afuselage and the rear gun turret wasremoved, leading to nominal changes inweight and performance. A smaller twin-engine experimental version of this aircraftwas considered at some point during the

An initial Armstrong Whitworth design study foran advanced RAF flying wing jet bomber with thecompany reference AW.50. Bill Rose Collection

In 1942, the Directorate of Scientific Researchapproached Armstrong Whitworth Aircraft(AWA) to develop an aerofoil section for windtunnel laminar flow experiments. As a resultof this, the company's chief designer John'Jimmy' Lloyd (1888-1978) began work onproposals for a military flying wing aircraft.Lloyd produced an initial design study of anadvanced flying wing jet bomber for the RAF,which carried the company reference AW.50and he was authorised to continue with this

Armstrong Whitworth Flying Wings

The final and most advanced glider in the GALexperimental series was the GAL61 (TS515). It wascompleted in 1948 and shown at the SBAC Air Showat Radlell during the same year. It was thenprepared for testing at Lasham, but never flew.Bill Rose Collection

the glider entered a steep dive and becameinverted. Kronfeld told McGowan that he hadlost control of the glider and his observerimmediately bailed out. The altitude wasapproximately 1,000ft (300m) and his para-chute opened at about 100ft (30m) making ita close call. Regrettably, Kronfeld stayed inthe aircraft until it hit the ground.

A significantly more advanced GAL experi-mental glider was nearing completion at thistime, called the GAL61. It had been reas-signed the original designation TS515 thatwas intended for use with a fourth GAL56 pro-totype. GAL61 featured a raised cockpit in thecentre of the aircraft, with a prone position foran obselver within the wing. This design hadthe same 36.40 leading edge sweep as theGAL56 Maximum V and in overall appear-ance, it loosely resembled the powered

orthrop N-IM flying wing. GAL61 had awingspan of 51ft 11in (15.8m), an overalllength of 24ft 8in (7.5m) and a gross weight ofapproximately 5,1741b (2,346kg). A fullyretractable undercarriage was fitted as stan-dard equipment and the GAL.61 was a trueflying wing in all respects. There were no ver-tical stabilisers and the aircraft used elevonsand wingtip drag rudders for control. GAL61was completed by mid-1948 and displayedstatically at the 1948 SBAC Air Show atRadlett. It was then returned to Lasham fortesting, but approval to begin test flights wasnever granted and the project was finally can-celled in mid-1949. Although it was the mostinteresting of the four gliders, GAL61 neverflew and is thought to have been broken upfor scrap in 1950.

(13.8m), with a length 23ft 6in (7.16m) and aheight of 9ft (2.74m).

All three gliders were exhibited at the Sep-tember 1947 SBAC show held at Radlett Aero-drome and the second GAL56 'Medium U'was demonstrated by Robert Kronfeld after aHalifax towed it into the air. But the pro-gramme was dealt a severe blow on 12 Feb-ruary 1948 when Kronfeld was killed whiletesting GAL56/0 I (TS507). He had left Lash-man Airfield with his observer BarryMcGowan and they were towed behind aHalifax until reaching 16,000ft (4,876m)where separation took place. The purpose ofthe flight was to carry out stall tests, but hav-ing made a normal recovery from his first stall

"IIII

"

o

"

"lJ

"IIII

"

~__ I \

'.... ) '--- .../

nI II IU

(3m). Medium U was assigned the referenceTS51 00 and it flew for the first time on 27 Feb-ruary 1946 at Aldermaston and was thenmoved to Lasham. It was damaged in an acci-dent during 1949 and subsequently scrapped.The third GAL56/03 (TS513) was called 'Max-imum V' and featured a 36.40 sweep. It flewfor the first time on 30 May 1947 at Lasham.This version of the glider featured nose flapsand was designed to allow adjustment of thewing dihedral on the ground. It was equippedwith two sets of split flaps. The first hinged atthe 50 per cent chord line, with the secondrear set hinged at 70 per cent. Used one set ata time, this was selected on the ground. Thewingspan of 'Maximum V' was 45ft 4in

1

16 Secret Projects: Flying Wings and Tailless Aircraft British Tailless Aircraft 17

The AW.52G on the ground. Bill Rose Collection

Armstrong Whitworth AW.52G

Armstrong Whitworth AW.52 Jet (TG363)

5120ft (36.58m)102ft (31m)2,61 1ft' (242.56m')2,250ft' (209m')80ft (24.38m)75ft (228m)32ft (9.75m) unconfirmed113,000 Ib (5 I,256kg)101,1501b (45,88Ikg)5x ROils-Royce AJ65 Avonturbojets, each rated at 6,500 Ib(2 .9k ) static thrust4x AJ65 engines, or po iblyRR RB 77 at 7,500 Ib static thru t640mph 1,030Kph575mph (925kph)50,OOOft (15,240m)Carrying a 10,0001b (4,535kg)bomb load, initially expected tobe 1,724 mile (2,776km)1aximum 20,000 Ib (9,071 kg)

bomb load. Either, one nuclearweapon, two ground penetratingbombs, 3x6,000 Ib (2,72lkg)bombs, or various smallerfree fall bombs

(revised)

CrewWingspan(revised)

Wing area(revised)

Length(revised)

HeightGross weight(revised)

Powerplant

Armament

Maximum speed(revi ed)

Service ceilingRange

Armstrong Whitworth AW.56 Flying WingMedium Bomber Project (OR.229)

the AW.52 prototypes was often difficult andtake-off and landing distances were unsati -factory. In May 1954, the second AW.52 wasbroken up for crap and it is thought that theglider met the same fate.

AW.56In late 1946, the UK Air Staff requested pro-posals for an advanced medium range jetbomber. Known as Operational Requirement(OR) 229, thi et out the RAF's future needsfor a fast, high-altitude aircraft, capable ofdelivering a 10,000 Ib (4,535kg) bomb to arelease point 1,500nm (I726 mile or2,776km) distant. Although the primalyobjective was the carriage and delivery of asingle atomic bomb, no British nuclearweapon had been produced at this time anda second requirement designated OR. I 00 1(later given the codenarne 'Blue Danube')was issued by the Air Staff to develop a pluto-nium bomb. Both the aircraft and the bombrepresented a significant technical challenge.The RAF was seeking a bomber that wascapable of sustaining a speed of 500kts(926kph) at an altitude of 50,000ft (15,240m)with a gross weight of 100,0001b (45,359kg).Bearing in mind the state of aerodynamicknowledge and ga turbine technology at thetime, this was a very tall order.

exhaust from the engines and there was aboundary layer control system connected tothe engine air intakes and pressure balancedcontrol surfaces. The aircraft also utilised afully retractable tricycle undercarriage with asteerable nose-wheel.

Like the AW.52G, the AW.52 was equippedwith vertical wingtip fins and rudders andcanisters were attached that contained anti-spin parachutes for emergency use. The trail-ing edge of the nying wing was occupied by asubstantial Fowler nap, which was contouredto allow for each j t exhaust pipe. After aperiod of tests and minor modifications, theaircraft was partly dismantled and trans-ported by road to the RAF ExperimentalEstablishment at Boscombe Down. Havingbeen reassembled, AW.52 spent the nextcouple of months undertaking ground testsand taxiing trials until a br ak in the weatherallowed the first night to be made on 13

ovember 1947.The trial was undertaken by Armstrong

Whitworth test pilot Eric Franklin whoreported that the aircraft handled as antici-pated. Following this ucce the AW.52 wasreturned to Bitteswell Airfield (no longer inexistence) for further testing. The secondAW.52 (TS368) was completed in the Whitleyworkshops during the summer of 1948 anddiffered slightly by having two less powerfulRolls-Royce Derwent 5 engines, each pro-ducing 3,5001b (15,56k ) of static thrust.Rolls-Royce engines appear to have beenchosen for reasons of reliability, although itremains unclear why Derwent turbojets werefitted to the second aircraft in preference tothe similar Nene units. TS368 fir t new on ISeptember 1948 and night-testing with bothaircraft continued until 30 May 1949.

On this date, John Lancaster was under-taking a series of trials in the vicinity of theBitteswell Aerodrome with TS363. After enter-ing a dive at about 320mph (510kph), Lan-caster encountered severe pitch oscillationand realising that the aircraft was about tobreak up, he ejected, becoming the firstBriti h pilot to use this system in an emer-gency. The second AW.52 was transferred tothe RAE Farnborough in October 1950 wheretesting continued until September 1953 whenArmstrong Whitworth left the programme.As a consequence, the RAE decided to con-clude operations.

Exactly how useful the Armstrong Whit-worth nying wing programme proved to beremains debatable. Lloyd hoped that thiswork would lead to a large nying wing airlinerpowered by six turbojets, but there were toomany technical obstacl s. Laminar nowproved to be largely unworkable, control of

AW.52G would normally glide at around 100-120mph (I60-190kph), with a maximum per-mitted speed of 250mph (400kph) and talloccurring at about 60-65mph (96-1 04kph).

Because there was so much interest inboundary layer control, an experimental sys-tem was devised to draw in air through a slotforward of the elevons. The idea was toreduce the possibility of wingtip stall at lowerspeeds and to achieve the suction, smallwind-powered pumps were fitted to each ofthe main wheel truts. Trials continuedthroughout 1945, with the aircraft usuallybeing nown by company test pilots CharlesTurner-Hughes or Eric George Franklin. Somerefinements were made to the glider as theproject progressed and it was then decided tobuild two somewhat larger experimental jet-powered nying wings. The project wouldreceive the company reference AW.52 andapproval und r the original MAP specificationE.9/44 by the Ministry of Supply.

The new aircraft would be approximatelythree quarters the size of Lloyd's initial designfor the AW.50. Its gross weight would be about34,0001b (15,422kg) and two Rolls-Royce tur-bojets would provide propulsion. In overallappearance the larger jet-powered AW.52was imilar to the glider, although the centreof the wing now contained a recognisablefuselage section that protruded forward. Thetwo turbojet engines were buried in the wingon each side of the cockpit section and the air-craft's internal structure was somewhat dif-ferent to the AW.52G, with metal being usedthroughout. The first AW.52 (TS363) was com-pleted in Armstrong Whitworth's assemblyshop at Whitley, Coventry, by early 1947. It hadalready been revealed to the public as a mail-carrying aircraft, although this appears to havebeen largely public relations spin.

Like the glider, this aircraft had an initialsweep of 17.5, followed by 34 to the wingtip,which provid d the distinctive boomerangshape. The AW.52 had a wingspan of 90ft(27Am), a length of 37ft 4in (I I Am) and aheight of 14ft 4in (4.4m). Empty, the aircraftweighed 19,6621b (8,919kg), with a grossweight of 34, 1541b (15,492kg) and the aircraftcarried a crew of two. The engines chosen forthe fir t aircraft were Rolls-Royce ene 2 tur-bojets, each rated at 5,000 Ib (22.2k ) of sta-tic thrust. The estimated maximum speed ofthe AW.52 was about 500mph (800kph), witha ceiling of about 35-40,000ft (10-12,000m)and a range of approximately 1,000 miles(1,600km). The aircraft featured a pres-surised cockpit and the pilot was equippedwith a Martin Baker Mk I ejector eat,although not the passenger. The AW.52 wafitted with a thermal de-icing system using

253ft lOin (16Alm)443.04ft' (41.16m')19ft 4in (5.89m)8ft 4in (2.54m)6,0001b (2,724kg) approx

one

100120 mpg (I 60 190kph) approx250mph (400kph)

2Initial 17.5, followed by 34 tothe wingtip (leading edge)90ft (27Am)1,314ft' (122m')37ft 4in (I lAm)14ft 4in (4.4m)19,6621b (8,919kg)34,1541b (15,492kg)2x Rolls-Royce I ene 2turbojets,each rated at 5,000 Ib (22.2k 1)static thrust500mph (800kph) approx3540,000ft (10-12,000m) approx1,000 miles (I ,600km) approx

Wing panWing areaLengthHeightEmpty weightGro sweightPowerplant

Maximum peedCeilingRange

CrewSweep

CrewWing panWing areaLengthHeightGross weightPowerplantGlide speedMax permitted speed

and rear eated observer, with onventionalnight controls, aside from a large handbrakestyle lever used to actuate additionalhydraulically-operated, horizontal controlsurfaces called 'correctors'. In addition to theother control surfaces, a substantial hydrauli-

ally operated one-piece slott d nap with alength of 15ft 4in (4.67m) is positioned alongthe entire trailing edge of the central wingsection. As a precaution during testing, anti-spin parachutes were housed in canisters oneach wingtip.

AW.52G utilised a fixed tricycle undercar-riage with brakes and a non-steering nose-wheel. The wingspan wa 53ft lOin (16Al m),with a wing area of 443.04ft2 (41.16m'). Over-all length was 19ft 4in (5.89m), the height was8ft 4in (2.54m). The gross weight of the aircraftwas approximately 6,0001b (2,724kg). The

The first of two AW.52 experimental jet aircraft(TG363) which were based on designs for thelarger AW.50 bomber. Two Rolls-Royce Neneturbojets provided propulsion, each rated at5,000 Ib (22.24kN) static thrust. Bill Rose olleclion

A staged company photograph of the AW.52G withcrew and observers. The unusual attachment to thetrailing edge of the port wing is a frame for the pitotcomb used to measure airflow. Bill Rose Collection

18 Secret Projects: Flying Wings and Tailless Aircraft British Tailless Aircraft 19

engineers into uncharted territory and it wasrecognised that a small demonstrator wasrequired to establish the viability of variousswept wing configurations.

During October 1945, the company's ChiefDesigner Ronald Bishop finally selected a smallexperimental tailless proposal known asDH.I 08, which was based on a Vampire with asingle swept tail fin. Propulsion was to be pro-vided by a single de Havilland Goblin 2 turbojetand because this prototype drew heavily on aproven design, the development time and costcould be minimised. In early ]946, the Air Min-istry approved full development of what wouldbe Britain's first swept wing aircraft. Two simi-lar, but slightly different prototypes would bebuilt to provide data for the proposed jet air-liner, plus future jet fighters like the DH.] 10 SeaVixen. One of the experimental aircraft wouldbe used to evaluate the performance of theswept wing at relatively low speeds while theother would have a transonic capability. Theseaircraft were now assigned the official (Exper-imental) references E. ]/45 and E.II/45.

Much of the DH.I08's design can be cred-ited to John Carver Meadows Frost (1915-1979) who is now remembered for his workon the Avrocar project in Canada. Both of theDH.I08 aircraft started life as fuselages forVampire F.l jet fighters that were takendirectly from the production line. As originallyplanned, the first aircraft was fitted with aGoblin 2 turbojet engine, producing 3,0001b(l3.4kN) static thrust. A new wing with a 430sweep and fixed leading edge slats were fit-

2x Bristol Orpheus turbojets,each rated at4,8591b (21.6kN)static thrust.lOx RollsRoyce RB I08 turbojets,each ratedat2,101lb (9.34kN)static thrustI,483mph (2,387kph)unknownunknown

I17ft 6in (5.36m)83'70ft7in (21.59m)16,3091b (7,398kg)

(VfOL operations)

DH.I08 Swallow

Armstrong Whitworth AW.171

CrewWingspanSweepLengthGross weightPowerplant(horizontal flight)

Maximum speedEnduranceRange

In 1943, the Brabazon Committee wasformed to consider requirements for civil airtransportation once the war had ended. Thisresulted in approaches being made to a num-ber of aircraft contractors for submissionsand one company who responded was deHavilland, which had studied the Commit-tee's Type 1V requirement for a jet-powered] OO-seat airliner. This resulted in proposalsfor a scaled up Vampire fighter, a tailless fly-ing wing using four Ghost turbojets and amore conventional concept that would even-tually become the DH.I06 Comet. However,such a project would take the designers and

(l2,192m) and progressively climbing to50,000ft (l5,240m), which would be attainedafter approximately 2 hours and 30 minutes.It was suggested that the Avon engines mightbe replaced by MetroVick F.9s, which wouldoffer improved thrust, lower fuel consump-tion and the ability to climb directly to 50,000ft(15,240m). However, the requirements for anew bomber were lowered after the TenderDesign Conference took place and this led toa significant revision of the AW.56 proposal,resulting in a slightly smaller aircraft. Theredesigned AW.56 would be fitted with thenew Rolls-Royce RB.77 engines, allowingthe use of four rather than five units. Thismeant that the tail engine was removed, lead-ing to better streamlining and a noticeableweight reduction. This improvement wouldprovide the AW.56 with performance similarto the proposed F.9 powered version.

Another alteration to the revised aircraftwas the use of a raised cockpit canopy for thepilot. Offset to the port side, this would haveimproved visibility considerably. AWA antici-pated full completion of this second designwithin two years and testing of the first proto-type about five months later. But the designwas finally rejected by the MoS in favour ofthe Avro 698 and Handley Page H.P.80 whichbecame the Vulcan and Victor.

Underside view of OH.I08 TG283 in flight.Bill Rose Collection

AW.l71The AW.171 is often listed as a supersonic fly-ing wing by aviation sources and thereforedeserves to be briefly mentioned. ArmstrongWhitworth produced several designs duringthe mid-1950s to meet a MoS specification fora one-man supersonic research aircraft witha vertical take-off and landing (VTOL) capa-bility. The first was the highly swept AW.17]which would be powered by two BristolOrpheus turbojets for horizontal flight and 10centrally located Rolls-Royce RB.I 08 enginesmounted in an upright position for lift. Thiswas followed by the AW.I72 which utilised aslightly different wing profile and propulsionsystem. It appears that Rolls-Royce's leadingengine designer, Alan Arnold Griffith, wasinvolved with this project and he produced anumber of broadly similar designs which alsoutilised centrally located lift engines.Although the supersonic VTOL demon tratorwas cancelled at an early stage, there wassome lingering interest in the idea of a largerVTOL bomber or transport aircraft based onthe AW.171, which continued until 1957when a wide number of advanced researchprojects were officially cancelled.

effective laminar flow. At this time, experi-ments to determine the viability of laminarflow had been sponsored by ArmstrongWhitworth and were being conducted atRAE Farnborough with a modified Hurricanefighter (23687). Initial results had been disap-pointing, but research continued, with amodified Gloster Meteor Mklll jet fighterand AWA's flying wing programme. TheArmstrong Whitworth laminar flow wing forthe AW.56 was to have a span of 120ft(36.58m) and an area of 2,611 ft' (242.56m').The overall length of the aircraft was initially80ft (24.38m) with an unconfirmed height of32ft (9.75m) and a gross weight of 113,000 Ib(51,256kg). To meet the performancerequirement of B.35/46, five Rolls-RoyceAJ65 Avon turbojets would be used forpropulsion, each rated at 6,5001b (28.9kN)static thrust. Four engines would be installedin the wing roots with air intakes on each sideof the cockpit and another engine would belocated in the tail, with its own dorsal air inletlocated mid-way along the fuselage. Thegross weight was calculated at 113,0001b(51 ,256kg), which suggested that the aircraftwould have a maximum speed of 640mph(I ,030kph) and a ceiling in excess of 50,000ft(l5,240m).

The AW.56's fuselage section would bebuilt using fairly conventional materials andconstruction methods, although the two-sparwing would be covered with an unusualmulti-layer skin similar to that developed forthe experimental AW.52 flying wing.

A fully retractable tricycle undercarriagewas planned for the bomber and a flight con-trol system developed for the AW.52 wouldbe used, comprising of two tandem controlsurfaces called a 'corrector' and a 'controller'.The 'corrector' acted as a longitudinal trimcontrol and the 'controller' was similar infunction to an elevon. A crew of five wouldman the aircraft, housed in a pressurisednose section, with access to the bomb bay toallow in-flight arming of the nuclear weapon.No ejector seats would be installed, but it wasproposed that the entire nose section wouldbe jettisoned in an emergency and could belowered to the ground by four substantialparachutes. As an alternative to a nuclearweapon, the AW.56 could carry a variety ofconventional stores with a maximum weightof 20,0001b (9,071 kg). Another interestingdesign feature was the bomb bay doorswhich opened transversely with sections slid-ing fore and aft. This was considered an effec-tive way of minimising impact on theaircraft's speed and trim.

A long-range mission would involve the air-craft ascending to an altitude of 40,000ft

The revised AW.56 design using fourof the new Rolls-Royce R677 jetengines and undergoing changes tothe cockpit and various aerodynamicfeatures. Bill Rose Collection

The initial design of the AW.56 jetbomber produced to meet Ministry ofSupply Specification 6.35/46 in early1947. Bill Rose Collection

vv

rating a host of advanced features with thecompany reference AW.56.

From the outset, Armstrong Whitworthbelieved that a tailless configuration wouldgenerate fewer aerodynamic problems attransonic speeds. They also proposed a seriesof wing slots that would allow boundary layercontrol at all speeds, leading to theoretically

47~7f1~n:E]~I>'..,j

Arm tr ng Whitworth r sponded toR.229 with an initial ubmission and not sur-

prisingly th de igners drew heavily on theirflying wing and laminar flow research. Afterth Ministry of Supply issued the more spe-ilk pecification B.35/46 in early 1947, Arm-

strong Whitworth submitted detailed plansfor a swept-wing tailless jet bomber, incorpo-

20 Secret Projects: Flying Wings and Tailless Aircraft British Tailless Aircraft 21

I39ft (II. 8m)25ft lOin (7.9m)

,800 Ib (3,991 kg)I xde Havilland Goblin II turbojetrated at3, I00 Ib (13.8kN)static thrust280mph (450kph)

/A730 miles (I, I74km)

I39ft (11.89m)327. 6ft (30.47m)I33kglm' (27Ib/ft')26ft lOin (8.17m)14ft (4.27m),940 Ib (4,064kg)

I xde Havilland Goblin 4turbojetrated at 3,73 Ib (16.67k ')static thrust0.42677mph (I ,090kph)40,000ft (12, 192m)

CrewWingspanLengthGross weightPowerplant

De Havilland DH.I08 (TG283)

Maximum speedCeilingRange

De Havilland DH.I08 (VW120)CrewWing panWing areaWing loadingLengthHeightGross weightPowerplant

ThrusVweightMaximum peedCeiling

post-war era meant actually going out and try-ing an idea to see if it worked. Both the US andRussians lost experimental aircraft during thisperiod, simply as a result of pushing the enve-lope that little bit further.

As a footnote, the DH.I 08 trials would inspirethe British film director David Lean to make afictional movie called The Sound Ban-ier,which was released in 1952. Starring RalphRichardson and Ann Todd, it wa very suc-cessful at the tirne, but is now largely forgotten.The film featured a 'n w' j t fighter called thePrometheus that was used in an attempt tobreak the sound barrier. Prometheus was actu-ally a ene-powered Supermarine Swift Type535 prototype (WI 19).

descent at 30,000ft (9,000m). Although therewas no serious instability, Derry later reportedthat the controls became heavy and unrespon-sive, despite the power boost system. But oncehe dropped down below Mach 0.96, everythingreturned to norrnal and Derry resurned hisnight to the de Havilland fa ility at Hatfield.

Trials with the third DH.l08 continued intothe following year, when the aircraft appearedat the Farnborough Air Show and took thirdplace in the SBAC's Challenge Trophy AirRace. It was then officially passed to the Min-istry of Supply which continued with testing atRAE Farnborough. Unfortunately, the thirdDH.I08 was lost on 15 February 1950 with itspilot Squadron Leader Stuart Muller-Rowlandbeing killed. He had been undertaking dives athigh transonic speeds above Bucking-harnshire when the aircraft broke up. Wreck-age carne down in the Little Brickhill area,with a number of parts found near Bow Brick-hill Church and some components as farremoved as Husborne Crawley. The local fireservice was mobilised, along with the police,who were instructed to cordon off the area asome aspects of the aircraft remained classi-

fied. Witnesses reported hearing an explosion,although this may have been a onic boom.The cause of the accident wa later deter-mined to have been an oxygen system failure,resulting in the pilot losing consciousness.

The first DH.l 08 was still Oying, but on I May1950, it crashed at Hartley Wintney, Hants,during stalling trials killing the pilot SquadronLeader George Genders. The three DH.108Swallow aircraft completed a total of 480nights, which provided invaluable data,despite all being lost in fatal accidents. Manyobservers have claimed the aircraft werejinxed, cursed or the design was fatallyOawed, but other believe these are notentirely fair comment . At that time, theDH.I 08 represented the cutting edge of aero-nautical research and these aircraft were reg-ularly Oown at their performance limits. Muchof the research undertaken in the irnmediate

De Havilland's third DH.108, based on aproduction Vampire F.5 fighter and configured forhigh-speed flight, was given the official referenceVW120. Piloted by John Derry, it set a world airspeed record of 605.23mph (974kph) on a 62-mile(IOOkm) closed circuit during April 1948.Bill Rose ollection

As a consequence, the main spar fractured,the wings sheared off and the fuselage cameapart. Wr ckage showered down into themuddy waters of Egypt Bay near Gravesend,Kent, and de Havilland was killed. Recoveryof debris proved especially difficult and Geof-frey de Havilland's body could not be located,although it finally washed ashore at Whit-stable on 7a tober 1946. This tragic accidentwould hold up high-speed night testing foranother year, during which time it wasdecided that the problems that had cau edthe accident could be effectively dealt with.In July 1947, a decision was taken to replacethe second DH.108 with a similar air raft,albeit with a number of modifications andimprovements.

The third DH.I 08 was based on a produc-tion Vampire F.5 fighter and it received theofficial serial number VW120. It was slightlymore streamlined than TG306 with a pointednose, giving it an overall length of 26ft 9~in(8.165m). A sleeker, more aerodynamiccockpit canopy was fitted, the pilot's ejectorseat was lowered, which reduced visibilityand the cockpit remained unpressurised.With engine performance steadily (althoughslowly) improving, this aircraft was fitted withthe latest Goblin 4 turbojet, producing 3,750 Ib(I6.68kN) of static thrust. As a final majorupgrade, the third Swallow was fitted withpower-assisted elevons to address the prob-lem of pitch oscillations.

Replacing Geoffrey de Havilland as thecompany's hief test pilot was wartime night-fighter ace John Cunningham (1917-2002)who made his first night in the new Swallowon 24 July 1947. The following year on 12 April1948, this aircraft piloted by de Havilland testpilot John Derry (I 921-1952), set a new worldair speed record of 605.23mph (974kph) on a62 mile (I OOkm) closed circuit, later earninghim the prestigious Segrave Memorial Tro-phy_ Flying this aircraft, Derry would alsobecome the first British pilot to exceed thespeed of sound, which took place on themorning of 6 September 1948 above Windsor.

For some weeks the Swallow had beentested at gradually increasing speed and onthis occasion Derry decided that conditionswere good enough to push the aircraft a littlefurther. Rapidly descending from about40,000ft (I2,000m), the aircraft nudged pastMach I just before Derry pulled out of th

bridge was being used for classified RAFexperimental work and its lengthy, well-maintained runway made this an especiallysuitable location. Somewhat surprisingly, thebase was closed in March 1948 and passed tothe USAF everal years later.

The ecrecy surrounding the DH.I 08 con-tinued until the beginning of June 1946 whenParliamentary Secretary of the Ministry ofSupply Arthur Woodburn visited de Havillandat Hatfi Id and disclosed the aircraft's exis-tence (which had now returned to the facil-ity). While minor modifications were beingmade to the aircraft, it was already apparentthat a tailless design based on prevailing tech-nology was less than ideal for an airliner.Meanwhile, work continued on constructionof the second prototype designed for high-speed night, which was fitted with a slightlymore powerful Goblin 3 engine rated at3,3501b (I4.9k ) static thrust. The aircraftwas marginally shorter than the first DH.I 08with an overall length of 25ft lOin (7.874m) asopposed to 24ft 6in (7.467m) and featured a45 leading edge wing sweep. Automaticleading edge slots were fitted that could belocked by the pilot, a more streamlined cock-pit canopy was used and there were improve-ments to the night controls.

This aircraft received the serial numberTG306 and was built to explore night at muchhigher speeds and perhaps supersede theexisting air speed record of 616mph (991 kph).

TG306 first flew during June 1946 andapp ared at the Society of British AircraftConstructors (SBAC) September 1946 showat Radl tt (which moved to Farnborough in1948). As well as being a static exhibit, it wasOown by Geoffrey de Havilland during theshow's final event and clearly impressed thevisitors. A couple of weeks later, on 27 Sep-tember 1946, Geoffrey de Havilland took offfrom Hatfield in the second DH.I08. He hadnow logged 13 hours in the aircraft andintended to push it towards its limits duringpreparations for an attempt on the air peedrecord, which had now been unofficially bro-ken in this aircraft. The emergency anti-spinparachute containers had been removed andthe aircraft's metallic surface had been pol-ished to a bright finish. However, things wentdisastrously wrong above the Thames Estuarywhen de Havilland encountered pitch oscilla-tions at Mach 0.9.

The clean, compact de Havilland DH.l 08 taillessexperimental aircraft, which was based on theVampire jet fighter and drew heavily on wartimeGerman aeronautical research. This photographshows the second prototype, TG306.Bill Rose Collection

"10

r- ed IMH 3

The third and most advanced DH.t08,equipped with a Goblin 4 turbojet,producing 3,750 lb (I6.68kN) of static thrustand power-assisted elevons to assist withcontrolling pitch oscillation problems.Chris Gibson

enced this design. Having been completed inwhat would now be regarded as an astonish-ingly short period of time, the aircraft wasassigned the official serial number TG283 andthe DH.I 08 made its first night on IS May 1946,with test pilot Geoffrey Roald de Havilland JraBE (I 91 0-1946) at the control . (Son of theoriginal Geoffrey de Havilland.) The test nightwas relatively uneventful and de Havillandreported the DH.I08 handled well, despitebeing restricted to a maximum speed of280mph (450kph).

While the DH.I 08 was not a classified pro-ject, some of the aircraft's technology wasconsidered sensitive and details of night trialswere treated with secrecy. There were run-way repair problems at Hatfield, so it was aneasy decision to continue with trials at RAFWoodbridge in Suffolk. At that time, Wood-

t d. Utilising quite a lot of wood in its con-struction, the new wing had a span of 39ft(I1.88m) and an area of 328ft' (30.4m'),which was 15% greater than the Vampire.

ontrol of roll and pitch was undertaken byelevons located outboard of the trailing edgesplit Oaps and containers were fitted to thewingtips that carried small anti-spin para-

hutes for emergency use.A single vertical fin now replaced the twin

tail boom fitted to the Vampire and thisresulted in a sleeker more elegant looking air-craft which was given the name wallow,although it was never formally adopted. As ar suit, this elegant-looking aircraft appearedto be something of a Vampire crossed with awartime Messerschmitt Me 163B rocket inter-ceptor, perhaps reOecting the fact thatGerman aeronautical development had inOu-

22 Secret Projects: Flying Wings and Tailless Aircraft British Tailless Aircraft 23

General arrangement of the Rolls-RoycevrOL flying wing designed in the late1950s by Geoffrey Wilde and John Coplin.Bill Rose CollectionJUK Patent Office

They settled on a basic layout that wouldutilise a large number of vertically positionedturbojets in the centre of the aircraft to pro-vide lift. An illustration shows the use of 30 jetengines, but the complexity of such a systemwould probably have made this design unat-tractive for a production aircraft. Louvredinlet and exhaust shutters would be used tocover the engine in level flight and four jetengines in ducts mounted on the trailing edgeof the aircraft would be used in forward flight.Wilde also envisaged a boundary layer, dragreduction system with air being drawnthrough conduits from the wing and fed to theintakes of the rear-mounted engines.

Proposed dimensions of this VTOL designare unknown, although the cluster of enginesat the centre of gravity would result in a sub-stantial wing thickness/chord ratio of asmuch as 20 per cent. Combined with a rela-tively low structural weight, this would allowa considerable payload capacity for cargo orpassengers within the forward wing on eitherside of the central cockpit area. As a light mil-itary transport, it might have been especiallyuseful for inserting and extracting SpecialForces during unusual operations.

General performance figures are unknown,but it seerns likely that speed, range and pay-load capacity would have been superior to thepresent day Bell-Boeing V-22 Osprey. Nothingcame of this paperwork study, although Rolls-Royce applied for a patent in 1959, namingWilde and Coplin as the inventors. This waspublished in 1963 and it received the refer-ence GB920875. There was no further notableinterest in flying wings at Rolls-Royce.

1-8 -3f

258Ft (l7.6m)750Ft' (69.67m2)55ft (167m)2xRollsRoyce AJ65 Avonturbojets, each provided 6,5001b(28.9k ) stalic thrust. Specificafterburner performanceunknown.650700mph (I ,OOO,!, 100kph)e timate50,OOOFt + (l5,240m +) estimate4x30mm Aden cannons

Rolls-Royce vrOL Flying Wing

25

In 1956, Geoffrey Light Wilde was appointedas the Head of the Advanced Projects DesignOffice at Rolls-Royce in Derby. An outstand-ing engineer, Wilde is now best rememberedfor his initial work on the RB211 engine, butback in 1956, his newly formed departmentbegan detailed studies of many differentideas for civil and military engine applica-tions, including a number of VTOL aircraftdesigns. Wilde was particularly interested indeveloping a fixed wing VTOL aircraft andspecifically, a flying wing, which would use acluster of small engines for lift with separatepropulsion for horizontal flight. He began todevelop this particular idea in associationwith one of his designers called John Coplinand it seems very likely that A.A. Griffith, whowas the company's leading VTOL proponent,also took part in this study.

CeilingArmament

Hawker P.I077

CrewWing panWing areaLengthPowerplant

Maximum speed

I55Ft (l6.8m)745ft' (69.2m')43Ft 6in (l3.3m)27,260 Ib (l2,365kg)2xRolls-Royce AJ65 Avonturbojets, each providing 6,500 Ib(28.9k ) static thru t632mph (1,0 17kph) at sea level50,000ft+ (15,240m+) estimate3x30mm Aden cannons

Hawker Siddeley PJD.l44One of the more unusual fighter designs toemerge from Hawker Siddeley's design officeat Kingston during 1949 was the tailless two-man P.I077 jet fighter, possibly intended tomeet Specification FA/48. It is almost as if thedesigners had taken a look at the WestlandPJD.144 proposal and then set about upgrad-ing it. The wingspan was 58ft (17.67m), thelength was 55ft (16. 76m) and was to be pow-ered by two stacked Rolls-Royce Avon turbo-jets equipped with afterburners. Air intakesfor the engines would be positioned aroundthe fuselage about mid-wing and the P.I 077would carry 1,000 Imp gallons (4,546 litres) offuel. Specific perforrnance figures areunknown, but this design was probablyintended to be capable of high tran onicpeed at altitude with a ceiling in the region of

50,000ft (I5,240m). The aircraft carried acrew of two and would be equipped with airintercept radar. The undercarriage consistedof a bicycle configuration with fullyretractable twin nose wheels and four largermain wheels. It seems almost certain thatoutrigger stabilising wheels would have beennecessary. A radar dish would be fitted withinthe nose of the aircraft and two large verticalfins and rudders were positioned in the cen-tre of each wing. Proposed armament werefour 30mm cannons in a ventral pack. eed-less to say, this interesting design never pro-gressed beyond the drawing board.

considered their options for an aircraft with atail and fitting one would negate any dragreduction advantages of the tailless design.The selection process continued during thefollowing year with the de Havilland DH.II 0finally being chosen as the best availablede ign. This led to a contract being issued inOctober 1948 for the construction of a DH.II 0prototype to meet Specification NAO/46 andfor evaluation as a new RAF night fighterunder Specification FA/48.

Maximum speedCeilingArmament

CrewWingspanWing areaLengthGross weightPowerplant

Westland PJD.l44

British Tailless Aircraft

a split central duct feeding both engines. ThePJD.129/144 concept has been comparedwith the Junkers EF.128 which was a Germanproposal produced for the 1945 EmergencyFighter Competition. To what extent thedesigners at Westland were influenced bycaptured research material is hard to say asmuch of the documentation for the EF.128project fell into Soviet hands. But it appearsthat the Westland fuselage had more in com-mon with aircraft like the de Havilland Vam-pire and the similarities to the German designare rather superficial.

The main differences between the PJD.129and the PJD.144 were alterations to the wingsand fins. The leading edge wing sweep wasreduced from 45 to 38, made possible bychanges to the wing's outer chord and the finsize was slightly increased. PJD.142 wasbroadly similar to this aircraft, but lacked thevertical wingtip fins and utilised a singleupright tailfin. PJD.143 was the most conven-tional of these studies with an upright fin andhorizontal tail surfaces. It was also decided toalter the engine layout and have the turbojetsside-by-side. Initially, Westland designedthese concepts around two unspecified tur-bojet engines in the Rolls-Royce ene classwith a static thrust of about 5,000 Ib (22.24kN),although the RAE decided that the newly pro-totyped and more advanced Rolls-Royceaxial-flow Avon (then known as the Axial Jet- AJ.65) was the preferred choice. Estimatedmaximum speed at altitude was about630mph (I ,0 13kph) and general dimensionsand weights complied with the AO/46 spec-ification. Armament appears to have been thesame type for all variants and in this ca e,three 30mm cannons were proposed.

A fully retractable tricycle undercarriagewas proposed for all models and the aircraftwould have been equipped with arrestorlanding gear. A radar dish for airborne inter-ception was housed in the aircraft's nose andthis system would have been operated by asecond crew member who was seateddirectly behind the pilot and facing rearward.On 15 December 1947, a Design Study Con-ference was chaired by the Director-Generalof Technical Development (Air) for the MoS,Stewart Scott-Hall, to assess all the submis-sions for NAO/46. The Westland proposalswere an early casualty and it was decidedthat performance projections for this aircraftwere unsatisfactory and there were generalconcerns about the tailless design. Delegatesfelt that in addition to lacking adequate per-formance, the Westland proposal might suf-fer from tip tailing and the anticipatedapproach speed was too high. The Confer-ence also decided that Westland had not fully

Secret Projects: Flying Wings and Tailless Aircraft

Hawker-Siddetey's PlO77taillesstwo-man jet design produced bythe Kingston Design Office in1949. Bill Rose ollection

24

The tailless Westland PJDI44 carrier-based warplane designed in responseto Specification N.40/46 during the late1940s. Bill Rose Collection

issued to a number of aircraft contractors andone of the companies that responded wasWestland. It undertook a detailed study,which resulted in an initial proposal desig-nated PJD.129, followed by three differingvariations. These were PJD.142, PJD.143 andPJD.144, with the designs completed in late1947.

The version preferred by Westland was thePJD.144, which was similar in overall appear-ance to the initial PJD.129 proposal. It was acompletely tailless design with verticalwingtip fins and rudders and this proposalutilised two turbojets in a stacked configura-tion along similar lines to the later EnglishElectric Lightning. Air intakes for the engineswere positioned on each side of cockpit with

In the immediate post-war years, the RoyalNavy b gan to consider its future needs for amore advanced carrier-borne twin jet fight rwith the ability to operate at night. It wouldneed to be capable of achieving a maximumspeed of 500kts (927kph), have a gross weightof no rnore than 30,000 Ib (13,608kg). A maxi-murn wingspan of 55ft (16.8m) was recom-mended and folded, the span could be nogreater than 18ft (5.5m). The rnaximum per-missible length was 43ft (I 3.1 m). As istedtake-off was anticipated and the armamentwould cornprise two or ideally four 30mmcannons. This Operational Requirement(0R.246) led to Specification NAO/46 being

Westland PJD.144

Sir Bames Wallis Kt, CBE, FRS, RDI, FRAeS(1887-1979) is best remembered as the sci-entist who devised the bouncing bomb thatwa used by the RAF in May 1943 to attack theM6hne, Sorpe and Eder dams in the Ruhr.This mission was called Operation 'Chastise'(more widely known as the DambustersRaid) and it eventually earned Barnes Wallisinternational fame, making him one of therelatively few British aircraft designer to beportrayed in a feature film (The Dambuslers,1954). The bouncing bomb project wasessentially a one-off undertaking and when it

Vickers Swallow

The single-seat Heston JC-9 which wasbuilt in sections and transported toWeybridge for use during the Swallowprogramme. However, it was neverreassembled and remained in a hangaruntil the project ended and it was finallyscrapped. Bill Rose Collection

At this time, the design that would be used forthe fir t two full-sized prototypes (VX770 andVX777) was essentially complete, althoughthe RAE had insisted on some significantchanges to the wing structure in late 1948,which would delay the start of construction atChadderton and Woodford until early 1951.Roly Falk undertook the maiden flight of thefir t Type 698 prototype (VX770) on 30 August1952. The test went well, aside from prob-lems with trailing undercarriage doors whichrequired careful inspection by a Vampirechase plane and a company 707. It was finallydetermined that this didn't interfere with theaircraft's ability to land and Falk put VX770down on the runway with no problems.

Production aircraft were to be power d withBristol Olympus engines, but these were notavailable at that time, so VX770 was fitted withfour lower-powered Rolls-Royce RA.3 Avons.In October 1952, it was decided that the newbomber should be given a proper name and aCommonwealth capital city was favoured.Ottawa was the preferred choice, but this ideawas soon dropped and Vulcan wa elected.

resulted in the aircraft crashing into the groundmoments after it left the runway. Both testpilot and Chadwick were killed. This repre-sented a significant setback to the programmeand almost led to cancellation of the contract.Within a matter of weeks, (Sir) William Farrenhad been appoint d as the new TechnicalDirector and soon managed to calm officialworries. This ensured continuation of the pro-ject, with Ministry confirmation being receivedon 27 November 1947.

Longitudinal stability remained a problemand the idea of utilising a horizontal tailfinsimilar to the Gloster Javelin was briefly con-sidered. This was eventually dropped, but asingle upright tailfin was retained. There werealso issues with the staggered engine layoutand this was finally dropped in favour of a lesscomplicated side-by-side arrangement. Workon the jet bomber continued and Avro wasawarded a contract to proceed to the hard-ware stage on 1 January 1948. The companywould now build a full-sized mock-up of thebomber, two scale-sized proof-of-conceptdemonstrators and two full sized prototypes.

The demonstrator initially comprised ahalf-sized aircraft called the Avro 710, whichwould be powered by two Rolls-Royce Avonturbojets, and a one-third sized single Rolls-Royce Delwent-engine version with the refer-ence Avro 707. Work began on the prototypesin 1948, but it was realised that building theAvro 710 would be as complex and expensivea a full sized aircraft, so construction of thisaircraft was stopped. However, it was decidedthat the number of Avro 707s should beincreased and five would eventually be built.

The first Avro 707 (VX784) flew on 4 Sep-tember 1949, but crashed on 30 September1949, killing the pilot Fit Lt Eric Esler afterentering a low-speed stall which is b lievedto have been caused by faulty air brakes.

proposed an additional crew area within theleading edge of the port wing. There wouldbe two separate bomb bays at each side ofthe engine compartment, each havingenough space to contain a single large bomb.The fully retractabl undercarriage layout is alitlle less clear from the one availabl drawingof this early design, but it appears to comprisetwo rear-mounted main wheels and two sep-arate nose wheels. The whole package wassomething of a design tour-de-force, combin-ing many new ideas in a largely unprovenaerodynamic shape.

When Roy Chadwick finally retumed to theoffice he was not entirely pleased with thedirection the project had taken and it seemsthere were disagreements about what he per-ceived to have become a science fiction con-cept. However, Roy Chadwick finally camearound to the idea of flying wing design andsoon began to suggest modifications andimprovements. The appearance of the newType 698 began to evolve towards a triangu-lar delta-winged shape, although some of theexisting features were retained. The forwardengine inlet was replaced with two intakes atthe leading edge and the rather strange dualcockpit layout was replaced with a singleenclosure. Chadwick kept the two bombbays, but now decided that wingtip stabiliserswere necessary. The undercarriage under-went considerable revision with the favouredchoice being a bicycle arrangement supple-mented by outrigger wheels.

There were al 0 attempts to design variouscrew capsule escape systems, but this wasconsidered too complicated and the ideawas finally abandoned. With concems aboutweight, Chadwick hoped to reduce the num-ber of turbojets to two, assuming it becamepossible to build an engine with 20,OOOIb(89k ) static thrust. This twin-engine deltawing bomber continued to evolve and a fuse-lage began to form in the centre of the wingand protrude forward beyond the leadingedge, making it easier to produce a pres-urised crew compartment and allow extra

space for essential equipment. But it wasclear that two engines remained an unrealis-tic option and the Avro team eventuallya epted that four turbojets were necessary.

Modifications continued and the Type 698now featured upright wingtip fins with ruddersand four Bristol BE. 10 (later Olympus) turbo-jets arranged with two, staggered above theothers. Air intakes for upper and lower engineson each side of the aircraft were located oneach ide of the forward fuselage ection. On23 Augu t 1947, Roy Chadwick was being car-ried as an ob erver on the test flight of an AvroTudor II. Incorrect servicing of the ailerons

unusual crescent-shaped wing with decreas-ing sweep and decreasing chord, which wasintended to provide a constant limiting Machnumber across the entire wing at a highc