C.P. No. 195 C.P. No. 195(16.514)

A.R.C. Technical Report(16,524)

A.R.C. Technical Report

MINISTRY OF SUPPLY

AERONAUTICAL RESEARCH COUNCIL

CURRENT PAPERS

A Cr i te r ion fo r the Pred ic t ionof the Recovery Characteristics

of Spinning Aircraft

BY

T. H. Kerr, B.Sc.

LONDON: HER MAJESTY’S STATIONERY OFFICE

1955

Price 2s. 6d. net

c. P. No. 195

U.D.C. No. 533.65 : 533.6..013.7

Technical Note No. Aero.2251

August, 1953

ROYAL AIRCRAPI! ESTABLISHMENT

A Criterion for the prediotion of the RecoveryGhsracteristics of Spinnug AircrxtY,

T. H. Kerr, B.Sc.

SUMMARY

In an attempt to establish a simple criterion for the prediction ofthe spin anl recovery chsraateristlcs of aircraft, It has been deduced thatthe two most important parameters are the unbalanced rolling moment coef'fi-dent about the wind axis in the spin and the ratio of pitching to rollingmoment of inertia. Using the results of full scale spinrung tests onthirty-three aircraft, it has been possible to establish empirux.1 rela-tionships between the estimated unbalanced rolling moment coef'fxient andthe inertia ratio which effectively dlvlde the aircraft into the threegroups which have satisfactory, borderline and unsatisfactory recoveryoharacterist~os.

A simple method 1s presented for estimating the unbalanced rollingmoment coeffxlent knowing only the shape of the aircraft. With thisinformation ard a knowledge of the mass distribution of the aircraft, theempirical relationships should give a good indxation of the spin recoverycharacteristus on new designs. Thu method 1s expected to be of partvxJ.arvalue to auoraf't designers in the early design stages since the methoddoes not depend on the result, of tunnel tests.

LIST I)F CONTENTS

Introduction

Previous Crlterla

Recent Evidence

3.1 Full Scale Tests3.2 Calculations based on Tunnel Rollzng balance Tests

Fundamentals of the Analysis and the Criterion

4.1 Fundamentals of tb.e Analysis4.2 The Criterion

Discussion

Extension to Swept Wing Aircraft

Conclusions

List of Symbols

References

?22%%

3

3

6

67

7

79

IO

11

11

12

13

LIST OF ILLUSTRATIONSFiwm

Body dsmplng and unshielded rudder volume ooeffiolent 1

Suggested criteria for recovery based on the vnng thickness/ 2chord rat10 t/c

Results of spinning tests on Meteor WH.320 3

The recovery standard of Meteor NB.320 for rudder only 4recoveries

The effect of rolling and. yawzng moment of uxx-tia parameters 5upon the yawing moment which must be supplied by the aircraftother than the wing for equilibrium 1n the spin

An estimation of the vnng rolling moment coefflcxents (uumd 6axis) in the spin

The orlteria for the predlctlon of recovery characterxtics 7of spinning aircraft

The extension of Flg.7 for swept wing aircraft a

-2-

1 Introduotion

The az.m of research into the spin and recovery characteristics ofaircraft must be to enable the accurate predlction of these chsracterlsticsin future designs. The method, present end past, is to build a scale modelto absolute dynsmlc similarity and complete R series of spinning tests inthe Vertical Tunnel. After making an allowance for the difference inReynolds number between the model and Pull scale alrcreft, the full scalecharacteristics of the spin are predicted directly from the model tests.These tests e.re not done until the design of the aircraft I.S slmcstsettled and it is, theEfore, very important that the designer shouldhave some guide, as to the probable characterlstscs of the nircraft inthe spin and durmg the recovery, m the early stages of the design.This is particularly important for eiementary end advanced trainers assatisfactory spin and a h~.gh standard of recovery 3.9 required of thesetypes.

a

%CO methods of producing such a prediotlon can be mentioned:-

(1) Calculations on the detailed design of each type based onthe aercdynsmic demvatlves obtaineu frcq rolling balance tests.To be effective, the data used in th1.s method must be obtained ator near flight Reynolds numbers an& this will not be possible untilthe rolling balance in the Bedford tunnel is in use.

(ii) A sinple criterion based on the geometry of the aircraft andcompared. with the wealth of flight experience which is available.A criterion of this type is of a limited character and. at best canonly be treated as approximate for borderline cases but it ms$ proveto be a valuable guide to designers in the oases of the elementaryand advanced trainers.

Several attempts have been made in the past to proaum criteria ofthis type (ii) and these are discussed in the next section.

2 Previous Crrteria

The first attempt to produce a criterion was ms.~!Ie by Finn x.11 1937'.He recognised that the fad.zE of models to pass the model s?inning testrequirements was usually due to one or more of the follovz%ng:-

(:'-I a large distributicn of mass along the X-axis i.e. (C - A)large; thxi has subsequently been show to apply to model scaleonly snd the opposite is true full scale,(il) inefflcient body section 1.e. the body cross-section of a tspewhich produces a low dsmping due to rotation,

(ILL) deficiency In side area - again producing low body damping,

(iv) shielding of the rudder by the tallplane.

During spinning tests in the Vertxal tunnel most models spin withthe vnngs approximately horleontal and therefore the inertia dzf'ference(A - B) was not important.

As far as the inertia loading of the alrcrsft was concerned (C - A)was regarded as being the most xnportant parameter and therefore theinertia difference coefficient C - A

kfb3

was taken as sn independent

PS 5variable throughout. This parameter has a direct effect on h andthcrefoxo also indirectly affects the body damping.

-3-

The bcdy damping or the resistanoe offered by fuselages of similarcross-section to rotation in the spig may be expected. to vary as I: 2 Awhere x is the distance of an element of side srea A from the C.G. Inorder to compare the various designs the body damping ratio was written in

The areas of fin ad rudder shielded by the tailplane

v/and elevator, in a spin at 45O mcidenoe assuming a wake spread of x0, wereassumed to have zero demplng and were therefore ignored. The area urder thetailplane ad elevator was more effective due to the tailplane position aboveit and a factor of two was applied to this area.

Although it was realised that boa cross-section affected the bodydamping no allowance was made for this in the analysis.

The effectiveness of the rudder as a recovery control was assumed tobe proportixml to the unshielded rudder area ad was expressed in the form

Unshielded Rudder unshielded rudder area x distance from C.G.Volu~~3 Coefficient p

The two aercdynamic criteria developed i.e. the body damping ratio endthe unshielded rudder volume coeffloient, were plotted against the inertia

term O - AA\3 - A rough sepsratlon between passes ard fails was obtained

using firstly, the two coefficients separately and secondly, the coefficientsmultiplied together to form 'the damping power factor'. The second casegave better separation between the passes and failures ix pass the requiredrecovery stanaards.

This oriterion was modified by Tye ad Fagg2 in thexr extension of theenquiry, to include full scale spinning aircraft.

The ahanges they made were:-

(i) The bcdy damping ooeffioient was expressed in the formd

L2

IE lx2 ax

-8, s b2

where h is the depth of the bcdy (side view)

x is the distanoe measured along the bcdy ads from the C.G.

e 1s the weighting factor applied. to various parts of side area

9 maximum -ve value of x

&2 maximum+vevalueof x.

-4- ’

The factor E was given values

2 for fuselage and fm under tailplane1.5 for rudder under tallplane

-0.25 for shielded rudder and fm1.0 for remainmg side area.

(c) The uns~elded rudder volume coefficient 1s now defined as

where pqi is the unshielded rudder area.

~hls analysis indicated the minimum degree of hoe damping andunshielded rudder volume ooefflclent which the designer should aim toprovide but compliance with the criterion 13x3 not guarantee good spinningqualltles full scale as factors such as body section and inertia yawingmoment couples had been ignored.

A more complete analysis was offered by Pringle and Harper3 1nMarch 1952. Taken as a whole, this analysis represented a very largestep forward. over the work of its predecessors. An effort was made topressnt the basic causes of error in previous criteria. Moreover, itwas devoted entirely to the analysis of full scale recovery predictions.The cntenon suggested depended upon the equation for the eqlulibrium ofyamng moments (body axu) 1n the spin.

Unshielded rudder VOLWWcoefficxnt c bodg.&mplhg coeffiuent +inertia yawing moment + inferred "ring yawing moment coefflclant = 0.

Of these qua?titxes only the inferred wing yavvlng moment cannot becalculated and.-therefone for the_bordexlLne case tho uSe.rred wing SpaKingmoment was assumed to be equal to and of opposite sign to the sum of theother three quantities.

Another assumption was that the wing yawing moment wasproportIona to the thickness/chord ratio and over a vnde range of hwas assumed independent of 7~.

The diagram showing the plotted results (Fig.2) of inferred wingyawing moment against thickness/chord ratlo for thuty-one axrcraft showthat in general a separation, between aircraft which recover from thespin by normal control movements and those which fail, is reasonablydefined.

The method of calculation of the body damping coefficient,unshielded rudder volume coefficient and inertia yawxng moment coupleare repeated below.

The body damping ccefflcient is equal to

Eh-Sb2 J h x2 dx

The weighting-factor e tc3given values at an incidence of 45O inTable I below.

-5-

TABLE I

Effect of Bod.v Section on Damping in Roll

1%dy Cross-Section 1 E ata=l+5"t

Circular (pointed profile)RectangularEllipticalRound top, flat bottomRound top, flat bottom t strakesRound bottom, flat topRound bottom, flat top t strskes

Fini

FreeUnder tailplaneAbove tailplane

Rudder under tailplane

+o. 6+I.5t2.1+I.1t1.pt2.5t3.5"t1.5+3.0-0.4t 2

/ *Depending on width of strdces. This is for 0.014. 2:" where 8"is distance of C.G. to rudder post.

The inclusion of h in the bodz

damping coefficient in this formwas demonstrated by Irving and Batson . Their results showed that thebody damping coefficient was almost directly proportional to h over arange of h from 0 to 0.9.

No changes were made to the nnshiel&ea rudder volume coefficienthut-it iaworth nating that this has.been demonstrated to be independentof h at a given incidence again by Irving ana Batsod.

The inertia yawing couple z was also included. in the analysxs.This v,as estimated from the particular values of (A - B) for the airorsftat-d assume3 values of T-c PI 4, and CL in the spin. The methoa is shownin Ref.3 ad in this way, a guess at the inertia couple could be made.Calculations over a range of types shows that this may have a value of+J units.

3 Recent Evidence

3.1 Full Scale Tests

The results of some recent full scale tests on a Meteor S6 withfive different rnertia ratios of the pitching to rolling moment ofinertia are shoxn m Flgs.3 and 4..

The technique of the tests wes to apply 'outboard' or pro-spinengine at the stall and maintain the ongine thrust throughout the spinand recovery. The thrust of the pro-spin engine was measured during thespin and recovery at-d in the lower graphs (Fig.3) the time required torecover is plotted against the yawing moment appllea by the pro-spinengine. The recovery action taken by the pilot was either, full oppositerudder and the elevator moved aown until the spin stops (normal recoveryaction) or full opposite rudder with elevator remaining fully up until thespin stops (rudder only recovexxes). In both cases the ailerons weremaintsinea neutral.

-6-

These results show that the inertia loading of the aircraft made alarge difference to the standard of recovery. The mean slope of the curve(Fig.4) 1s negative and gives a greatly improved recovery stsndard as thepitching to rolling moment of lnertla was increases.

The recovery standard of the Meteor 8 for three inertia conditions.2 (Aircraft 30a, b and o where */A = 1.62, 2.23 and

?l2%2~2t$eY~f _end by this criterion the stsndarci of recovery shouldbe reduced an& not increased with */A as the full scale results xdioate.

3.2 Calculations based on Tunnel Rolling Balance Tests

A series of wina tunnel tests on a rectangular ClarkeYalng were madewith the NACA spinning balance by LBsmber and Zimmerman7 as part of aresearch on spinning aircraft. All SLX components of aerodynsmio forceand moments ore measured through a range of angles of attack, angles ofsidesllp, and values of E*b llke-ly-to-be obtained by spinninGaircraft,.

%-latter-pat of thoWepoz+xcor&.ins an-snalysls, using-this data, forestimating the spinning characteristics of an aircraft. The most inter-esting result, when these measurements are used XI celculatiuns with vvlngloadings and inertia ratlo common today, is shown In Flg.5.

The quantities plotted are the yawing moment (anti-spin) which mustbe supplies by parts of the aircraft other than the wing for equilibriumin the spin against the ratio of the pitching to rolling moment of inertiaparameter for incidences of 30 and 45O. This result again shows how ttx?equilibrium in the spin and therefore the recovery standard depends uponthe inertia loading of the 3ircraft. Other crossplotted results from thisreport showed that the wing rolling moments in the spin are dmost directly

proportional to !7bh v( >

. Calculations using strip theory on wings at

high Reynolds ndxrsconfirm this result up to h = 0.5 or 0.6.

Much time and effort in the past has been devoted to showing thatthe pro-spin wing moments depend upon the thickness/chord ratio of thewxng and rolling balance tests have shown this to be true. In view ofthese previous results, a graph of the suggested wing rolling moments isshown in pig.6 based on the Rssumption that the vnng rolling moments area linear funotlon of thickness chord ratio.

4 The Cn terion

4.1 Fundamentals of the Anal.vsis

On exanming the equilxbrium of an aircraft in the spin one canconsider either the eqLulibrium between the znertia and aerodynamic forcesabout the wznd axes or about axrcraft body axes. If aircraft body axesare chosen then there 1s equilibrium between the inertia and aeroaynamlccouples about each axis and any small change about one axis will affect

, the equilibrxnn about the other two.

If aircrsft vvlnd axes an chosen then since the centrifugal forceson all parts of the aircraft act radially from the axis of the spin, therecan be no centrifugal couple about that axxs and therefore the eqlulibriumof the spin is entxrely aeroaynamlc 1~1 nature. Thus, for equilibrmm inthe spin the pro-spm moment dus to the wings must equal the antI-spinmoments due to other parts of the aircraft in the spin.

-7-

---:. CPB + &.?g + ep, = 0 (wind axis)

If the rudder is central and the aircraft is in equilxbrium then

--&pg +ep, = 0.

This does not represent the complete picture of the equilibrium inthe spin as each of the inertia couples about body axis must be balancedby corresponding aerodynamic couples. These can exert a marked lnfluencaover the equilibrium of the spin and the recovery.

The xnfluence of the pitching moments of inertia 1s greatest indecidiny the rate of rotation about the spin 3x1s for equilibrium at agiven incidence. Thhls 1s reflected in the equation for deriving

for the aircraft.

'I'he rolling and yawing inertia couples Influence the angle of tiltof the vrlngs and thus the sdesllp in the spin. T2xir influence on thespin and recovery characterrstlcs of an alrcraft ore shcwn In Figs.3, 4and 5 and the xmpcrtsnt parameter appears to be the ratio of ths yawingand rolling inertia couples which can be expressed in the form.

A - Bb 3

PS Tj0 =A-BepA-B = 1-BC - B C - B A A

as A+B fi C.

Therefore the twu most important parameters am:-

and (ii) (j - F) -

For the recovery from the spin when the rudder 1s deflected againstthe spin if

then theoretically we have the case of a borderline aircrsft which justdoes not recover from the spin. If these do not equal zero, then we havean unbalanced rolling moment coefficient (URMC) about the wind sxls and ifthis is anti-spin or pcsltive then the aircraft should reccver from thespin. Ths results of calculations of this type are plotted (Figs.7 and 8)for approximately 25 aircraft. Bcundarles can then be drawn in separatingthe aircraft whxh were pass, borderline, or fail in their full scalespinning tests.

-8-

Thus to assess a new aircrsft the calculation of the two importantparameters should be male and the reccvery characterlstlcs assessed bycomparison with the empirical boundaries which have been drawn on Flgs.7and 8.

In a calculation of this type It is Important to keep it as simpleas possible and only cne spinning incidence is used to assess thecharacteristics of the spin. This is 45' at the plane of symmetry as

(i) this is a mean spinning incdence of an average azrcraft,

(il) the rate of rctaticn about the axis of the spin will be aminitium and is the worst case when body damping is considered.

4*2 The Criterlcn

The criterion given below is based broadly upon previous criteriavnth mcdlfzcaticnsitothe methods of assessing the results.

The method of appllcaticn should be as fcllcws:-

(4 to estimate h from the general layout snd loading of the aircraft

where JR is the aspect ratio

b, ‘= CC,-‘Aj.8PS cy\

(b) To calculate Che anti-spm rclllng moment ccefficlent (wind axis)due to the body at a mean incidence of 45'.

Values of E can be obtained from Table I and Flg.1 and assuminga wake spread ever the tailplane of 30'. It is usually impcsslble toallocate one value of E for a complete fuselage annd it is, therefore,essential to allocate, a particular value of E for each section of thefuselage as the calculation 1s made.

-'PB = A I chx2&x

Sb2

(c) . To calcula.te the snti-spin moment due to the movement of the rudderfrom neutsa~ $$$Uy antl-spin.

G:= !g (Fig.1 )

(4 To estimate the wing rolling moment coefficient frcm Fig.6.

-Y-

$2 ,,;;;g the unbalanced (ant-G-spin) rolling moment coefficient Prom

umo = er: + &Fn + e,

(f) mot this result against, 1 - s and assess whether the aircraft( >

is likely to have a satisfactory spin recovery characteristic by referenceto the empirical bound~es which have been drawn on Figs.7 and 8.

5 Discussion

A critical examination of Fig.7 will show both the weaknesses andadvantages of this criterion over previous attempts.

(i) Although s good deal of effort has been applied, no evidencehas heen found to show that the Hurricane (Airoraft 4) was a fail3and a fairer classification is thought to be borderline. It wasclassified in this way rn Fig.7.

(ii) The criterion does not ap ear to apply to aircraft of theVampire-Venom type (Aircraft 31 . Allocation of values of E inPthe formulae for body damping .&p< do not appear to represent atrue case even when the side area is multiplied by 2 to representthe damping over the twin booms.

(iii) There is a considerable amount of scatter for aircraft whichhave positive values of 1 - f .

( )It is important to note that the criteria takes account of

rudder power alone as a recovery control but in flight the normalrecovery action is to use opposite rudder and to move the elevatordown. As the value of zB deoreases the elevator becomes inoreas-ingly important as a recovery control as csn be seen if Fig.3 isexamined. Ihe Value of the elevator as a recovery control isextremely difficult to assess as it depends upon

(1) the effectiveness of the rudder in reducing the rate ofrotation in the spin

and (2) the effeot on the rudder of applying down elevator duringthe reoovery. This can be extremely important if, whenapplying doml elevator during a recovery the shielded area ofthe fin and rudder is increased. !Chis could have a veryserious adverse effect.

Thus the scatter shown when 1 - f( >

is positive probablydepends upon the effect of applying dov.n elevator during therecovery from the spin.

A word of warning might be offered to designers here, as itis dangerous to rely on elevator power for the recovery from thespin. At best it leads to long recovery times with high rates ofroll during the recovery which cause stressing difficulties in thewings and possible disorientation of the pilot during therecoveries.

-lO-

(iv) The criterion shows an improving recovery standard a3 thepitching moment ?f inertia of the aircraft is increased. Thisagrees with recent full scale experience (Aircraft 3Oa, b, o) andthe calculations based an tunnel tests.

This is the most important improvement over previous criteriawhich have always shown a decrease in recovery standard as thepitching moment of lnertla has been increase&.

tVI The criterion includes an estimate of the wing rolling momentwind axis) and when more data becomes available it can be applied

with only slight modification to the criterion for both straightand swept vvlng aircraft. A summary of present data for swept wingeircraft is discussed in the next section.

6 Extension to Swept Wing Aircraft

The calculation of unshielded rudder area and body dampingcoef'ficien': KU be modified in that sllowsnce must be made for thesweepback of the tailplane. It is suggested that the present AP.970recommendations be continued and 30 and 60° lines for estimating theshiel&d fin and rudder area be drawn from a point one thzrd of the spanfrom the fuselage side.

Very little is known of the rolling moments of swept wings atspinning incdences but recent full scale experience shows that at lowvalues of h (less than 0.2) the spin 1s little different from that Ofsimile straight vrsng aircrsft. As a result of this it is suggested thatthe same estimated wing rolling moment coefficients be used althoughcomparison of the CL-~ curves indicate that the results given by thisedterinn ar?likely to be a little pessimistic. Fig.9 shows the resultsfor three swept wing sk%rLLft plo LLd on il graph which ~ncldes theboundnry for passes and fails on straight wing aircraft.

7 Conclusions

(1) The limitations of a criterion of this type cannot be overemphasised particularly in borderline cases but It is hoped thatit Will prove a valusble guide to designers, particularly in thecases of elementary and advanced trainers.

(ii) The results (Figs.7 and 8) do separate arcraft into threeclasses, pass, borderline and fail for the recovery from the spin.

(iii) The principsl advantage of .Jhe new criterion 1s that itincludes the important parameter K and indicates that the recoverystandards will improve as g increases. This is in agreement withthe results of 801ne American rolling balance tests and recentfull scale experience.

(iv) Although the elevator, as a recovery control, is not incldedin the criterion, Its importance has been emphaslsed in the diagramsand calculatxns.

(v) As soon as the results from the Bedford tunnel rollingbalance tests are availsble, the methods of extracting the wingrolling moments, body dsmplng and rudder powr should be revisedtaking adv3xtage of these results

- 11 -

LIST OF SYMBOLS

aistance of element of sde area from C.G.

area of element of side area = h dx

wing area

wing span

unshielded. rudder area

depth of fuselage

maximum negative value of x

maximum positive value of x

distance OP centeroid of the tinshlelded rudder area from C.G.

weighting factor applied to various parts of side area

x

A

s

b

J+ch

&I

c2

c

E

Rb.

c - Ah=gy 0, =b3

-7-TgPs 2

n

V

fix-"pwG

A

B

C

P

rate of rotation about spin

true rate of descent ft/sec

bo4y damping coefficient

axis

1wing rolling moment coefficient

unshielaea rudder volume coefficient when therudder is deflected against the spin

rolling manent of inertia

pitching moment of inertia lb.ft2

yawing moment of inertiadensity at altitude 00nsiderea slgs.ft2

i

wind axes

/

i

-12-

No.

1

2

3

4

5

6

7

Axthor Title, eta.

E. hnn Analyszs of routine tests of monoplanes in theRoyal Aircraft Establishment Free Spinning Tunnel.R. &M. 1810. July 1937.

G. E. Pringle andD.J. Harper

H.B. IrvingAS. Batson

A.V. Stephens

H.B. Irving andA.S. Batson

Kerr

Barber and

Splnrung Criteria for monoplanes.Report AD 3141, 194.0.

The spinnug of model aircraft and the predictionof full scale spin ard recovery chsracterlstics.R. &M. 2906. March 1952.

Spinnxng experiments on a single seat fighter withdeepened body and raised tailplane.Part I - Model Ekperlments.R. AM.. 1421. December 1931.

part11 - Nl scale spinning tests.R. &M. 1421. December 1931.

Spinning experiments and calculations on a modelof the Fairy IIIF Seaplane with special referenoeto the effects of floats, tailplane modificationsand. floating ailerons ad interceptors.R. &M. 1356. June 1930.

Full scale spin and recovery characteristics of aconventional straight wing aircraft with fivedifferent ratios of the rolling to pitching nrnzntsof inertia.(To be publish&)

Spinrung characteristics of wings.Rectangular blade-Y monoplane wxng.NACA Report 519.

- 13 -

FIG. I.

+%B O D Y DAMPINCj -COEFFICIENT

x-h = yj-p

/

EhsZds

-e,

UNSHIELDED RUDDER

VOLUME COEFFICIENT

AR -. S H A D E D A R E A

.t : D/ST: B E T W E E N CENTROID O F S H A D E D

AREA AND C.q. O F A I R C R A F T

FIGI. BODY DAMPING AND UNSHIELDED RUDDER

VOLUME COEFFICIENTS.

6 0u3 W

0

co

a,-A

50 I I AIRCRAFT ADDED SINCEPUBLlCATlON OF REPORT

025 No. AERONo. AEROAzg .2456 1

BORDERLINE

FAIL

PA55

940 P3Oaq.62

A?Bc l-12 1n’ p 3

INFERREDA30 b

METEOR B I

WING 30YAWlNq

MOMENT 29COEFFICIENT 3 2 b

x 10%.

r,

02320

0Z:b

A27

2 82 8 0909

2222813813 00 I8I8

IO I

L

31

00. IO 0 II

l 5

2 aI 1:5 0. I4 0 I5 0 I6 (A.

‘I OXFORD23 SKUA3 MARTIN BAKER F5/434 HURRICANE5 JOCKEY6 SPITFIRE7 NIGHTHAWK

. 8 MASTER

Q3 WELLESLEY10 P O P M 10

9 II MACjISTERI?. MEMTORI3 HARVARDI4 DEFIANT t4 PRENTICEIS MOTH MINOR 23 METEOR416 TYPHOON 26 CORNELL

010 I7 BRISTOL 133 27 ATTACKER18 PROCTOR 20 WYVERNI8 PB I’ M 20/2 29 METEOR 420 P &P M 28 30 METEOR 821 MOSQUlTO 31 VENOM22 VICKERS F7/4l 3 2 SEAHAWK

.23 SEAFIRE 3 3 PROVOSTI 0.10

UNSPINNABLE ‘II

PASS

M E A N WIN? THICKNES5/CHORD R A T I O $

FIG. 2. SUGGESTED CRITERIA FOR RECOVERY BASED ON WING.THICKNESS/ CHORD RATIO.

ANCjLEOF

T I L T

IOR A D I U S -

\ -_\-, 1-c _ _

- - -- - -

0 5 IO I5APPLIED PRO- SPIN YAWlNq MOMENT IO?

1 F l G . 3 .

I.62 NORMAL INERTIASI 97 - - 0 + 25 %223-.- 0 + 50 ‘A1.255---- A + 25 ‘/.I.12 - - - A + 37i %

3

J25

i c,’O NFIG.3. RESULTS OF SPINNING TESTS

METEOR WH 320.

FIG. 4.

15THRESHOLD OF

N ON- RECOVERY

IO 3 C,’

IO

F I G . 4 . R E C O V E R Y S T A N D A R D O F

M E T E O R 8 W H 3 2 0 F O R R U D D E R

ONLY R ECOV ERIES.

F I G . 5 .

YAWINGYAWING

COEFFICIENTCOEFFICIENT

FIG. 5. E F F E C T O F R O L L I N G AND

YAWING MOMENT OF INERTIA

PARAMETERS UPON THE YAWING MOMENTW H I C H MUST B E S U P P L I E D B Y T H E

A I R C R A F T O T H E R T H A N T H E WING.

F O R E Q U I L I B R I U M I N T H E S P I N .

FIG, 6.

.8 tT

IO % 12% I4 % 10%

6

0 - 5 -10 -IS -20 -25

ESTIMATED WINE, ROLLlNq MOMENT

COEFFIC IENT X IO3 (WING AXES) &

FIG. 6. AN ESTIMATION OF THE WING

R O L L I N G . M O M E N T C O E F F I C I E N T S

( W I N D AXES) I N THE S P I N .

I

23456783IO1112I314IS161718IS20212223242s262728253031

. .

OXFOROSKUAMARTIN - BAKER F 5 143

HURRICANEJOCKEYSPITFIRENlCqHTHAWKMASTER

WELLESLEYP &PMIBMAqtSTiRMENTORHARVARDDEFIANTMOTH MINORTYPHOONBRISTO i I33PROCTORP8.P M20/2P&P M26MOSQUITOVICKERS F7,/41SEAFIREPRENTICEMETEOR- 4CORNELLATTACKERWYVERNM E T E O R 4METEOR 8VENOM

.I .

P A S S

u UNSPINNABLE

@ P A S S

0 BORDERLINE

. F A I L0 A I R C R A F T O F

VAMPIRE -VENOM

TYPE FOR WHICHTHIS CRITERION FAIL!

030 b

032 b

-l-T-IrUNBALANCED

u3 ANTI - SPIN29 40 251 ROLLlNq M O M E N T

COEFFICIENT x IO 3(WIND AXES)!I-!-a-30 o. . 3 0 c

30

53 I I I

1201320.

026 I O4 06 I

32 SEAHAWK33 PROVOST3 4 -F I A T 4 4 9

FIG. 7. THE CRITERION FOR THE PREDICTION OF zi

RECOVERY CHARACTERISTICS OF SPINNING AIRCRAFT. :h

FIG. S

PA55

00

-2 0

F I G . 8 . E X T E N S I O N O F F I G . 6. F O R ’

SWEPT WING AIRCRAFT.

k.P. No. 195(16,524)

A.R.C. Technical Report

Crown CopyrIght Reserved

York House. fingnuay, IIONDON, w.c 2 : 423 Oxford Street, LONDON, w.1P 0 Box 569, LO?KXX’, s B I

13a Castle street, EDlN~“RG”, 2 I St Andrew’s Crescent. CARD~FF39 King street, MANCrnR. 2 Tower Lane. mu.wx, 1

2 Edniund Street, B~MWCMAM, 3 80 Cbchester Street. BELFASTor from any Bookseller

1955

Price 2s. 6d. net

S.O. Code No. 23-9007-95

C.P. No. 195