December, Intro4uc Ory. (Reasons For J Guiry.)--In

835SU IARY01 R E P O R T S G. 5 A, B, C . D DOF ACCIDENTS TO "X " AEROPLANES.P r e s e n t e d b y t h e Accm , zs C o m I E .Re er 8 and Memoranda, No.629.December, 1918.SUMMAZ '.--(a) Intro4uc ory. (Reasonsfor J guiry.)--In the Spring of1918 the Accidents Department recorded a considerable number of accidents t o " X " aeroplanes which had in common one or other of the two followingfeatures :-(i) The machine was seen to dive at a gradually increasing angleand finally stTike the ground nearly vertically, although onseveral occasions evidence seemed to show that the pilot wasmaking every e ort to right it.(if) In the course of an unsuccessful loop the aeroplane was aptto remain in an inverted position, despite any effort of the pilot.(b) Range of he inv6stigaJion.--The Committee's work can be dividedinto three parts :--(i) A critical analysis of the circumstances in which the accidentstook place. (See report G. 5 A.)(if) After consultation with the designers, suggestions of modifications which on general grounds could be considered likely toincrease the safety of this type of machine. (See reports Nos. G. 5 B and G. 5 C.) These were carried out on two machinesand were found to produce the expected improvement.Off) Finally, an exhaustive experinwntal investigation at the R.A.E.which gave exact numerical values for the most satisfactoryarrangement, both as to rigging, shape of wing, and tail planesetting. (See report G. 5 D.) .(c) Co lusio ts.--The simplest method of improving the longitudinalr mtrol of " X " aeroplanes is to set the taft plane at its largest possible egative angle and to fit an elastic to pull the stick forward with such Qrce as to keep the aeroplane in trim at a suitable speed. The fitting ofsmall chord elevators would increase the improvement sti further.I f a more elaborate set of alterations are to be made the arrangement aso n X10 (type E) appears to be the best.In each case, however, the aeroq3lane will still be unstable longitudinally. If it impossible to move the [ower plane backwards owing to the necessity of redesigning the fuselage,an improvement would be obtained by increasing the area of the fixed.tail plane. (See also p. 847).Rm'O T(PART I) O T H E ACCID .NTS C O M M I T T E E O N A C C I D E N T STO " X " AEROPLANES. (G. 5 A.)I r o d u c t l o n . - - 1 . I n t h e course of t h e i n v e s t i g a t i o n of the ccidents t o " X " a e r o p l a n e s referred t o t h e Accidents Committee,m u c h general i n f o r m a t i o n has b e e n g a t h e r e d a b o u t t h e machine.B5200Ax

836There is considerable evidence to show that the machine" hangs"at the top of a loop its control is, as in all types, lost with comparative ease. The inverted aeroplane immediately loses speedand the controls become ineffective.2. The results of tests on an inverted model wing are givenin Appendix I. These show that the wing resistance of theinverted machine is extremely high, the maximum li /drag ratiois as low as 3, and that the stalling speed is far higher than inthe correct position. Lastly, the maximum lilt coe cient maybe as low as half the normal.3. With regard to the effectiveness of the controls, theoreticalconclusions drawn from a number of full-scale tests made at theR.A.I . on the ef ciency of tail planes and elevators are given inAppendix II. The occurrences foreseen therein have been verifiedin flight by i Iajor A., the 0 / C 53rd Training Squadron, R.I .C.,Captain B., Captain C., /[r. D. and Captain E. (see Appendix HI).4. There appears to be general agreement that control whichwas lost with the machine invertecl is recovered when the engineis cut off, or Cuts itself off, but it .should be noted that the immediately following attitude of the machine is a position of nosediving beyond the vertical. It is probable that when diving overthe vertical, as long as the engine is kept on, this aeroplane cannotbe righted by the action of the pilot on the elev.ator.' 5. These conclusions are supported by :-(a) Theoretical calculations based on the available erodynamical data. (Appendices I and II.)(b) The evidence submitted by two experienced instructors.(Appendix I I I ( ) and (5).)(c) The experimental flights made at the request of theCommittee by three pilots (Appendix I I I ( ), (d), (e)and ( ).)(d) The comparative study of six fatal accidents. (Appendix IV.)6. The machines to which these accidents occurred will bedenoted by X1, X2, X3, X4.X5 not referred toX6 not re erred tothe Department.the Department.I n all these cases the aeroplanes were seen in a nose diveeither vertical or over the vertical, and although there appearedto be ample time to straighten out, the machine finally struck theground vertically ; in all cases it would appear that the enginewas on. A brief report on each of the four accidents referre tothe Committee will be found in Appendix IV.7. I t possible that in some instances the flyer being inexperienced may have committed some error of judgment, but t h e

837impulse to elevate would be so natural that one must assume tha in almost every case during the period of dive, the flyer w a sexerting a large, if not his maximum, pull on the control columnto right the machine.8 . B a s e d on three lines of independent evidence (a) modelexperiments and aerodynamieal caiculations, (b) full-scale experiments made by several independent pilots, (c) critical analysis ofthe series of accidents, the Accidents Committee has come to theconclusion that as at present rigged and constructed the machinecan get into a position in which it is impossible to right it byany effort on the elevator control so long as the engine is keptfull on.9. The effort on the elevator required to keep the machineon a steady path increases as the path becomes steeper and thespeed greater. With the engine on there appears to be a definitecritical angle oi descent (possibly 70 to 80 ) beyond which thoelevator control becomes inadequate, and once past this angle, inspite of any effort the pilot may make, the steepness and speedof the dive will increase until finally the machine is over thevertical.10. Two semi-stable attitudes are then possible :-(a) A nearly vertical path with the axis of the machinevertical.(b) A glide at between 300 and 40 with the aeroplane onits back.The machine will pass easily from (a) to (b), but as long a the engine is on, only with an e/ ort from (b) to (a).11. Assuming the pilot to be making every effort to rish the machine, but having left the throttle open and the spark o the former of the two paths alone is stable.12. With the machine inverted the petrol supply stops andeventually the engine fails, and by failing restores control to thepilot. The machine then swings back, and tends to right itself,but before the aeroplane can pass the critical angle, the speed hasagain risen and the engine started, locking the machine oncemore between position (a) and (b).13. It would appear that in practice either of the two positionsmay initiate trouble.I/ a dive is allowed to increase in steepness, position (a) isfirst reached. If the aeroplane fails to complete a loop it firstlocks on position (b).14. All the evidence available with regard to the six accidentsdetailed m Appendix IV harmonises with one or other of thesealternatives. It is probable that the aeroplane comes out ofthese positions without much di eulty if the engine is switchedoff. It depends on the effectiveness of the l teral control whetherit can be got out b y rolling.

83815. T h e C o m m i t t e e h a v e records tha# t h e designers of t h i sm a c h i n e h a v e for s o m e t i m e b e e n a w a r e t h a t t h e a e r o p l a n e ism o r e easily controlled w h e n a n a d j u s t m e n t of rigging dJffereni f r o m t h e one used for s t a n d a r d t e a c h i n g is a d o p t e d , a n d it a p p e a r st h a t t h e m a n u f a c t u r e r s h a v e since N o v e m b e r , 1917, b e e n usinga specially rigged " X " a e r o p l a n e as a h a c k machine. A t a b o u tthis d a t e s o m e " X " aeroplanes c a m e t o b e t e s t e d b y t h e F a r n b o r o u g h A . I . D . Staff a n d similar i n f o r m a t i o n was p u t on recordas t h e result of t h e s e i n d e p e n d e n t tests.(Signed)M. O ' G o A ,Lt.-Co .,Clmirman Accidenta Committee.12th March, 1918.INDEXTO APPENDICES.Appendix I . . . .Wing coefficient at Negative Angles.Appendix I I . . . .lZ.A.E. Report and Technical Dept. Report.Appendix I I I (a) Major A.'s Report.(5) O/C's Report (53rd Training Squadron, lZ.F.C.). Co) Captain B.'s Report.Captain C.'s Report.(,) Mr. D.'s Report.(D Captain E, of Tralnlug Division.Appendix IV.- . Note on Accidents to X1, X2, X3, X4.APPENDIXI.Expe mcntal data on wings at eg ive angus of incide ,e.mThe appendedcurves (Fig. 1) by the N.P.L. to the Advisory Committee for Aeronauticsgive the forces and moments on an inverted wing. The experiments werecarried out on the wing section R.A.F. 14, which differs considerably fromthat used in the " X " aeroplane, but the following characteristics areprobably common to all wing sections when inverted.1. The lift coefficient is less than half that of the wing in the normalattitude and at the corresponding angle of incidence.2. The maximum ratio of lift to drag is less than one quarter.The stalling speed of the machine when upside down will; therefore,be 40 per cent. lower than when in normal flight, and control will be lostif the aeroplane is flown horizontally for any distance when inverted.APPENDIXII.Results of heoreticalinvestigation.--Lieut. Wood, of the R.A.E., attendedat the Accidents Committee Meeting of the 26th February, 1918. Hestated that the " X "aeroplane, Irom calculations made at the l .A.l -.,would stall at about 35 miles an hour and trim at its normal speed of levelflight, but is unstable. When at a high speed in a vertical or over verticalclive with the engine on, a force on the control column in excess of theeffort which the pilot can make or the controls withstand would be required to right the machine. Its elevator control has been found byexperiment to fail at 75 lbs. pull by the buckling of the control column.