Airworthiness considerations for STOL aircraft

NASA

TECHNICAL NOTE

NASA TN D-5594

,q- O,

,A

Z I-..- ,.=: ts) Z

COPY

AIRWORTHINESS CONSIDERATIONS

FOR STOL AIRCRAFT

_v Holzhauser, by Robert C. lnnis, Curt A.

and Hervey C. Quigley

Ames Research Center

Moffett Field, Calif.

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION • WASHINGTON, D. C. • JANUARY 1970 2. Government Accession No.

1. Report No. 3. Recipient's Catatog No.

NASA TN D-5594 4. Title and Subtitle 5. Report Date AIRWORTHINt':SS CONSII)EP,.ATIONS FOR STOL AIRCItAFF ,Ianual _, 1970 6. Performing Organization Code 7. Author(s) R/_h_'rl C'. Innis, Curl A. HtAzhauser. and Hcrvey C. (_)Lli_Luy B. Performing Organization Report No.

A-3077 9. 10. Work Unit No.

Performing Organization Name and Address 12_-62-03-04- 0O- 2I NASA Anws Rc_arI'h ('en[¢'r Moll('ll Fi_'ld, C:dil.. 94035 ,11. Contract or Grant No.

13. Type of Report and Period Covered T( ( hll]l tll Nutc Sponsoring Agency Name and Address N'ati/)ll_ll A_'I'II]ILItlliL'N _lll([ _f)_i_ (_ Adl/lilli_tr_llillll W;tshil,_ton, I), C, 2054(i 14. Sponsoring Agency Code 15. Supplementary Notes 16. Abstract ('I'll( 1'i_ ill't' pl't_'_tqll_!d [t_l" SLtliS[LLt _Ol'y p{ I IOFIII_III'(! _lll(I hail(JUlia (ltlalit Jl'Yl *11 p_,_crcd-lilt STOL transp_a'l air_ rafl ul t iiilllllt,rc[_tl lil)(q';itii_ll. Th('st" Cl'i_Ul'i_l Wt'F(' [()rlIikl_J(_d ]'I'(111_ iIl_l_i']_l_ltioll _4_th(q'( it dLlI'irl_ NASA Ilil_hl If'sis lJI S']'OI_ ail'( l-all.

Tilt' 111_1111 ('IIII)il_ISiN iN ilia( (,(1 (H1 I11(' l_Llldill_-_li)l)l'oach lll(_d(' b("( _ltl_( _ this I'_il]lO l/a_ CaLlS_'d th(' _l'('_lt_'_-I (]i_fit'll_ly _l[l(I is [ht' IIlIJN[ dl'llltlndiIlg I)()rli(_ll I)l thE' fli_hl II wLI_ ('()li("l_d('d [h_t sTag _lil'( I'_l_t il_i]Jzi[l_ ])l)W( I" _(1 d{'vl_l()t) lJU (_lll _llu_y Llpt)I'(JLI(']] LIIId l_lllll \kith sll/_l[U,r Sl?t'i,d lllar_ills Itl_lll ('OllVIHltil)II_t[ _lii'('l'_lU NI) _i[l_J.( _ [_1( _)1 [h_il WI)ul(I l'_'_t( _ _]. d(!lllOnSll'_lI(,d t)('l'iOI'lllal/t (' 1o Iit']d l_:[l_t_l ( illlld Uo (h'v('h>l)t'd _)(,('/IuN(' d_l_l w(,]'(, ill_kll_i( i_ll_ [_j]" it_(,SSilJ_ ( ()ll_iS[('ll.('} r of I)£'I'[I)I'III_IIIUO.

_]('( [l_llli( _ll l'lll/tFf)l (QKII'_i( ltq'l_[il h _l't' lllOl't _ illll}Oi"_tlll ill OV(!l'_lll ]litl/0Ullg Ih:l[1 [ht'V _Ll't _ lI_l" ( OIl_,('[lli(ll/_t] _lil'('l':lll ])(_( _ILI,_C Ill( Stal)ility _HId Iltt_lll)lll _ ttro _l'lli,l'/l[ly [(Jw kt[ STOL sI)(_£'ds. }{i_]l It,v( ls of _li'l'l_(l_lalllil sl,lbiliIv _llld dalllplti_ art, ii_ll lJ_,l(,5.'_tl'il_, d(,sil'_'d b_'I'ausl' thw,} _lllrl_ lhc ;_il't t'_l_l Ill I_ _ Inl_l'(, _'asilv (listLlrbcd ill 51sty air.

Additi(m:ll v_,s, al'_ h i_ /_,s_ary [_r (h'I( rnlinin_ mr,r(, ac(uralcly t]w _,ifctt _[ _sls. sh('ars, and ( r,,ss;_i;_cIs _m i)t'rlorm;_nc_' and handling quMitws 18. Distribution Statement 17. Key Words (Suggested by Atahors_ STOL Df,r/(lrlllLilII c Crit(q'la Unclassili_'d - Unlimited STOL ltanctlin_ (2rilcria 19. 22. Price* Security Classif. (of this report) 20. Security Classif. (of this page) 21- No. of Pages ,3.00 Unt Iassined Um lassifi(,d 66 "For sale by the Clearinghouse for Federal Scientific and Technical Information Springfield, Virginia 2215] TABLE OF CONTENTS l SU_$IARY ...............................

i INTRODUCTION ............................

NOTATION ............................

DESCRIPTION OF VEHICLES .......................

DISCUSSION OF LOW SPEED ENVELOPE ..................

Illustration of STOL or Powered-Lift Envelope ...........

Lift-drag polars ........................

Vector diagram .........................

Normal acceleration capability .................

I0 Operati onal enve lope ......................

I0 Minimum speed, Vmi n .......................

Ii Climb and descent capability ..................

Restrictions Imposed on the Operational Envelope for Safety, . . . .

Propulsion system failure ....................

._linimum speed margin ......................

Available climb gradient in landing configuration ........

Maximum sink rate ........................

Flight-path adjustments .....................

Touchdown attitude .......................

Considerations of asymmetry .................

Summary of restrictions ...................

Conclusions ............................

DISCUSSION OF FIELD LENGTH FACTORS .................

Landing Field Length .......................

Consistency of performance .................

Demonstration of performance .................

Field length factor .......................

Obstacle clearance angle ....................

Take-Off Field I,ength .......................

Take-off gradient ........................

Take-off field length factor ..................

Terminal Area Operation ......................

Conclusions ............................

DISCUSSION OF HANDLING QUALITIES .................

Background ............................

Level of Criteria .........................

Criteria for Aircraft Response to Control Input ..........

Factors included in criteria ................

Lateral control criteria ..................

Directional control criteria ................

3O Longitudinal control criteria ..................

Flight-path control criteria ................

Substantiation of control criteria ................

Lateral control power .....................

Lateral control sensitivity ...................

Lateral control forces .....................

Lateral control cross-coupling .................

Page Directional control forces ........... • • • , • ° • ° Directional control cross-coupling ...............

Longitudinal control power ...................

Longitudinal control sensitivity, forces, linearity, and Flight-path control .......................

Aircraft Response to External Disturbances ............

Directional stability and damping substantiation ........

Dihedral effect .........................

Spiral stability . .......................

Lateral damping .........................

Conclusions ............................

CONCLUDING REblARKS ........................ ° REFERENCES .............................

TABLES .,,, ..... ° ........... ,.°°,o,°°,

AIRWORTHINESS CONSIDERATIONS FORSTOLAIRCRAFT

By Robert C. Innis, Curt A. Holzhauser,

and Hervey C. Quigley

AmesResearch Center

SUMMARY

Criteria are presented for satisfactory performance and handling

qualities of powered-lift STOLtransport aircraft in commercial operation.

These criteria were formulated from information gathered during NASAflight

tests of STOLaircraft. The main emphasis is placed on the landing-approach

modebecause this regime has caused the greatest difficulty and is the most

demandingportion of the flight.

It was concluded that STOLaircraft utilizing power to develop lift can

safely approach and land with smaller speed margins than conventional air-

craft. No single field length factor that would relate a demonstrated perfor-

manceto an operational field length could be developed because insufficient

data were available.

Mechanical control characteristics are more important in overall handling

than they are for conventional aircraft because the stability and damping are

generally low at STOLspeeds. High levels of aerodynamic stability and damp-

ing are not necessarily desired because they allow the aircraft to be more

easily disturbed in gusty air.

Additional research is necessary for determining more accurately the

effect of gusts, shears, and crosswinds on performance and handling

qualities.

INTRODUCTION

There is increased interest in short take-off and landing (STOL)aircraft

by airlines and government agencies for commercial air travel (refs. 1 and 2).

The low-speed characteristics of STOLaircraft should provide added conve-

nience to the air traveler because such craft can be flown into small air-

fields and restricted space thereby expanding air travel to a larger portion

of the population. In addition, STOLaircraft showpromise in alleviating

someof the congestion in and about our major air terminals by being able to

operate in currently unused portions of the airspace and airports.

At present, STOLcapability cannot be exploited in a commercial short

haul system. WhenSTOLperformance is achieved by reduced wing loading and

increased power loading, the aircraft has a low cruise speed, is quite dis-

turbed by gusty conditions, and contains considerable design and performance

compromisesto achieve safety. WhenSTOLperformance is obtained by including

a significant portion of the lift and control from the propulsion system

(hereafter referred to as powered lift), higher wing loadings can be utilized.

Such STOLaircraft are of interest to the major airlines because they have

good cruise characteristics, are less disturbed in gusty conditions, and have

improved passenger comfort. However, current airworthiness standards do not

permit exploiting the low-speed potential of powered-lift aircraft.

Before powered-lift STOLaircraft can be utilized commercially, several

questions must be answered. First and foremost is, what will be required in

terms of low-speed performance and safety margins? Second, how steep an

approach and climbout angle can these aircraft routinely fly, and how large

a runway will be needed for all surface and atmospheric conditions antici-

pated in daily operation? Third, what handling qualities must these aircraft

have to allow the pilot to fly small, steep patterns in a safe and easy

manner under instrument as well as visual flight conditions.

The study that follows was madeto provide guidelines and criteria for

answering the questions on performance margins and handling qualities

required for powered-lift STOLaircraft. Previously published NASA reports

on V/STOLand STOLaircraft were reviewed; these reports describe tests made

primarily to understand the capabilities and limitations of these aircraft

and to examine specific problem areas that occurred. The reported results

could not be directly used to answer the previous questions because the

results were either too limited or not addressed to aspects of commercial

STOLand V/STOLaircraft. Recently, the FAApublished "Tentative Airworthi-

ness Standards for Verticraft/Powered Lift Transport Category Aircraft"

(ref. 3). This publication provided someguidelines, but did not answer all

of the previous questions. Therefore data were extracted from previous NASA

reports, re-examined in light of more recent knowledge, and re-addressed to a

commercial environment so that guidelines and criteria could be developed.

Data were primarily extracted from tests with which the authors had close

familiarity so that the available information could be more consistently

examined. Greatest emphasis was placed on information obtained in the STOL

regime by STOLaircraft that had the most promising operational characteris-

tics. it was assumedthat the vehicles of interest will be flying in the 40

to 80 knot speed regime with descent and ascent angles of at least 6° , and

will require a high degree of maneuverability to operate in restricted

airspace.

Guidelines are developed for safe, low-speed operation with emphasis on

consistent performance over a wide variety of environmental and runway con-

ditions. Wherepossible, criteria for safe low-speed handling are presented

in a form that can be easily measuredand interpreted by the pilot. Data

that substantiate these criteria are also presented. In this report primary

emphasis is on characteristics in the landing modewhere it has been found

most difficult to achieve satisfactory performance, handling qualities, and

operational characteristics. Further, this is a critical area of the flight

envelope because of the numberof decisions and the judgment that must be

madein the brief period of time before touchdown.

The information and criteria presented should assist the designer in performing trade-off studies in the preliminary design phase. It should a]so give the potential operator and regulatory agencies a better idea of how the pilot may operate the STOL aircraft, and what he requires in terms of perfor- mance margins and handling qualities. I_qlile the information was developed primarily for STO1, aircraft, it should be equally applicable to VTOL aircraft operating in the STOI, mode. The information and criteria presented are by no means complete or intended to be conclusive. Areas requiring further research are noted, and it is expected that the criteria will be revised and expanded as more experience is gained and new unique vehicles are tested.

NOTATION a incremental acceleration normal to flight path, ft/sec 2 n dV ft/sec 2 a X longitudinal acceleration along flight path, _-, longitudinal acceleration as measured by an accelerometer at the Ax center of gravity, g normal acceleration as measured by an accelerometer at the center of A z gravity, g BLC boundary-layer control mean aerodynamic chord, ft drag coefficient, including propulsive thrust CI) lift coefficient, including propulsive thrust lift coefficient in steady-state flight (21 g lateral control force, lb FI, P acceleration of gravity, ft/sec 2 g h height above runway, ft IFR Instrument Flight Rules moments of inertia, slug ft 2 Iyy

,xx}

Izz SL/Ixx I,p damping in roll, _p , 1/sec YY 1/see L r roll due to yaw rate, 3r ' _LII XX • 1/sec 2 dihedral effect• rolling acceleration for full lateral control surface deflection, L rad/sec2 Ixx' SL/Ixx rolling acceleration per inch lateral control deflection• LgLp rad/sec2/in.

aM/lyy i/sec Mq damping in pitch, _q , i/sec2/ft/sec speed stability, Mv _V _.,I/Iyy angle-of-attack stability, M(_ Sa ' i/sec2 _N/Izz , i/sec Np yaw due to roll rate, Sp SN/Izz , i/sec N r directional damping• Sr zz 1/sec2 directional stability• NB aB SN/I zz 1/sec N_ damping due to rate of sideslip, a(dB/dt)' N , rad/sec2 adverse yaw due to full lateral control, N 6 " Izz L roll angular velocity (right roll, positive)• rad/sec P PR pilot rating pitch angular velocity (nose up, positive), rad/sec q free-stream dynamic pressure, Ib/ft 2 q_ r yaw angular velocity (nose right, positive), rad/sec R rate of climb, ft/min C S wing area, ft 2 SHP shaft horsepower t time, sec ramp time for control input, sec t a time to reach 30 ° bank angle, sec t3o T transparency, average inboard propeller blade angle minus average outboard propeller blade angle, deg thrust T c ' thrust coefficient, q S T time to damp one-half amplitude, sec i/2 time to double amplitude, sec T2 V true airspeed, knots or ft/sec approach airspeed, knots Va calibrated airspeed (at low Mach number V C = Vv_), knots V C maximum flaps-extended speed, knots VFE V minimum airspeed in steady-state flight at reference power condition, min knots minimum control airspeed, knots VHC airspeed at which aircraft is rotated, knots V R crosswind component, knots VXW critical decision speed, knots VI optimum climb speed, knots V2 W gross weight, Ib angle of attack, deg indicated or uncorrected angle of attack, deg C_ U angle of sideslip, deg flight-path angle (above horizon, positive), deg aileron deflection, deg a flap deflection, deg f lateral control surface position, deg L lateral cop, troI position (positive movement producing positive °l,p moment; right, positive or clockwise positive), in., or deg longitudinal control position (positive direction producing positive _Mp moment; aftstick is positive), in.

rudder pedal position (positive direction producing positive moment; Np right pedal forward, positive), in.

rudder deflection (trailing edge left, positive), deg _F G pitch attitude (nose up, positive), deg "2 pitch angular acceleration when pitch rate is zero, rad/sec 2 u 0 damping ratio P air density', slugs/ft 3 @ density ratio T time to 67 percent of steady-state value, sec bank angle (right wing down, positive), deg bank angle after 1 sec, deg roll angular acceleration, rad/sec 2 roll angular acceleration when roll rate is zero, L6 , rad/sec 2 L heading angle, deg _ 0 yaw angular acceleration when yaw rate and sideslip angle are zero, rad/sec 2 W frequency, rad/sec directional frequency, rad/sec _d DESCRIPTION OF VEItICLES The first figure illustrates the variety, of aircraft that were tested and evaluated by NASA and that formed the basis of this report. The first NASA STOL flights were made in the Stroukoff YC-134A in 1959 (ref. 4). Since that time, the remaining STOL aircraft have been evaluated (refs. 5 to 15). The aircraft have encompassed gross weights from under 5,000 lb to over 150,000 lb and wing loadings from 23 to over 56 Ib/sq ft. Approach speeds have ranged from 40 to 90 knots. The geometric characteristics of these aircraft are LOCKHEED NC-130 B SHIN MEIWA UF-XS BREGUET 941 BOEING 367-80 CONVAIR MODEL 48 RYAN VZ-3RY DE HAVILLAND C-8A STROUKOFF YC-134A Figure 1.- STOL aircraft used for this report.

summarized in table I. The majority of these aircraft were powered by turbo- propeller engines, although two were powered by reciprocating engines. One jet aircraft, although perhaps not STOL in the sense of requiring only a short field length, is included because it had some degree of powered lift by virtue of boundary layer control and engine exhaust impinging on the flap, and its variable stability system had been used to help develop some of the handling qualities criteria that will be discussed later. Although these aircraft were used to formulate the criteria that will be presented, it should not be implied that they were all suitable for commercial STOL operation.

DISCUSSION OF LOW SPEED ENVELOPE Illustration of STOL or Powered-Lift Envelope The lift capabilities of STOL aircraft are as much a function of thrust as they are of angle of attack; thrust not only reduces the minimum airspeed but can also control the vertical acceleration. For these reasons, safe operating speeds cannot be related to a singular stall speed, but must be related to thrust or power utilized. A method of relating these parameters with an operational envelope is developed in the following sections.

Lift-drag polars.- Curves of lift and drag coefficients at different thrust coefficients for an illustrative STOL aircraft in a landing configura- tion (Breguet 941, ref. i0) are presented in figure 2. Since the curves include the thrust component, C D = 0 corresponds to level unaccelerated flight.

The following approximate relations are useful for later analysis: a n C L AC L .... 1 = -- (i) 2.0 g CLlg CLlg 6 / / ing #1 P _._ _/ (conslant power, a x CD 5 -- tl r -- z-t-- .._'_._._vorying airspeed] +- = (2) / I ,1 7"; g CLIg _] _ power, / ./_/ " ' CL [ I1\ constant _" 7.,4 The solid curve represents the lift

lit \,o,,,o, ,/A f-° and drag characteristics measured as

_ II [ condition / / angle of attack is slowly increased

1 -'1

and airspeed decreased at a constant , • o,:o _ £ power or throttle setting; the thrust - T °°:° - C coefficient increases at constant o ___.L._.__ I 1 I I I power as the airspeed is reduced CO a, deg because thrust is nearly constant as dynamic pressure is reduced Figure 2.- Lift-drag curves in landing configuration (ref. 10, _f = 98°).

(T c' = thrust/(I/2)pV2S). These characteristics represent steady-state flight conditions. The dotted curves are for constant thrust coefficients which correspond to values measured in accelerated flight when angle of attack is changed at constant airspeed. The change in lift and drag obtained by modu- lating power at constant angle of VTOL attack and airspeed is depicted by the dashed line starting at C L = 3.9 and = 0 °. The corresponding vector is /-- inclined over 75 ° and represents the / / Thrust-lo-weighl ./ angle through which the effective rohO / thrust (propeller slipstream) was vec- 1.0 tored by the large flaps deflected to _/ / 98°; consequently, large increases in / / lift can be obtained by increasing the / thrust level. In contrast, a conven- / tional aircraft has little deflection / / of the propulsive force and the lift / curves with power on are near the .5 / ;:( STOL :_:] !_;[,,,_!::qT:dt'.'_ < power-off curves.

Vector diagram.- Figure 3 com- pares the STOL thrust vector from /i,/'/ ]'_":'IC T ii "_ figure 2 with those for a VTOL and conventional aircraft. This diagram is useful in further illustrating the definition of STOL and explaining some of the implications of "powered Figure 3.- Resultant thrust vector in landing lift." The diagram refers primarily configuration,

to the landing portion of the flight envelope, which is the most critical

area in terms of attaining good descent capability with adequate control and

handling qualities. In the approach conventional aircraft primarily require

thrust to balance the drag, and the thrust level is only a fraction of the

gross weight. Modulation of thrust produces primarily the horizontal accelera-

tion while rotation produces normal acceleration. In contrast, the VTOL

requires thrust in excess of its weight, and modulation of thrust produces

primarily normal acceleration; horizontal acceleration is obtained by rotating

the thrust vector or aircraft attitude. The STOLaircraft lies between these

two and the thrust level required is a significant fraction, but still much

less than the weight. For the STOLexample, modulating thrust produces more

normal acceleration than horizontal acceleration, and rotation of the aircraft

produces less normal acceleration than does rotation of conventional

7- I •

airplanes. As a consequence, cor- 6 -- + /s' _-- _ "" Tc': 2 rections to the flight path during the approach are most expeditiously accomplished by modulation of the thrust, while attitude or angle of // J_/_-c_,_ attack is maintained relatively 4- _ '_'"_% Tc'= 0.7 cons tant.

CL .3 -- /I Power Vo/Vmin Condihon Normal acceleration / I -- O Approach I.O Unacceleroted stall Accelerated stal P capability.- It is important to (/._ _ "'**"To' : O G_ Max I.O 2 [] Approach 1.15 Approach examine the magnitude of normal / I _ Approach t.15 Accelerated stall acceleration available from • Max 1.15 Approach changes in power and angle of I-- _" Max 1.15 Accelerated stall attack to assess the maneuvering _, Approach I ,30 Approach

]

capability of STOL aircraft and 0 I I I I to determine a safe operating -8 0 8 16 24 e,deg speed. Figures 4 and 5 were prepared to illustrate the rela- Figure 4.- l, ift curves for accelerated and unaccelerated flight, landing configuration tionship between power, lift {ref. 10, :Sf = 98°).

coefficient, ratio of approach Angle of speed to minimum speed, and nor- attack Power mal acceleration. These figures 1,2-- f Vo_2 -I were based on the lift character- istics presented in figure 2.

\%,,7 "_//" The maximum power condition is -- _d, Increased for one of the four gas genera- tors inoperative with all four propellers operating by virtue of the interconnecting shafts used in the aircraft (Breguet 941, 4///'..../_..._..__ _221:Z ;;;s22;2d ref. i0).

1.0 t.I 1.2 1.3 1,4 1.5 The solid curve of figure 4 Va represents lift coefficients Vmin measured under steady-state Figure 5.- Calculated normal acceleration flight conditions at a constant capability for different approach approach-power setting. The speeds.

resulting maximum lift coefficient was 5.2. If the approach speed is chosen

at 15 percent above the minimumspeed (Va/Vmi n = 1.15), the lift coefficient

can be increased from 3.9 to only 4.5 by an accelerated stall at constant

power, llowever, if power is applied at the approach angle of attack, the lift

coefficient can be increased to 4.9; if power and angle of attack are

increased the lift coefficient is increased from 3.9 to 5.6. The correspond-

ing incremental normal-acceleration capabilities are summarizedin figure 5.

For reference purposes, the acceleration capability of a conventional aircraft

defined by (Va/Vmin)2 - 1 is also included.

The normal acceleration obtained with a STOL aircraft by changing angle of attack only, is less than that of a conventional aircraft at the sane ratio of approach to minimum speed, ltowever, this deficiency is compensated by the normal acceleration that can be provided by the propulsion system even near the minimum speed of the aircraft. Relations like those shown in figure 5 were also obtained with the other STOL aircraft tested; in each case, the portion of acceleration obtained from the propulsion system was dependent on the extent of thrust vectoring and the power available.

Although the previous characteristics were for wings level flight, the same principles apply for banked-turning flight, l@en maneuvering a STOL aircraft by banking, the pilot uses power to maintain the desired altitude or rate of descent at the approach angle of attack, and, therefore, the stall margin is not decreased in turning flight. For example, a 30 ° banked turn at Va/Vmi n = 1.15 would require an increase in lift coefficient from 3.9 to 4.5 which would be obtained by applying power and maintaining the approach angle of attack. In this condition, the angle of attack and stall margins are unchanged and an incremental normal acceleration of over 0.15 could still be obtained by rotation. In contrast, the airspeed of a conventional aircraft must be increased in turning flight if the stall margin is to remain constant.

Operational envelope.- Since the stall or minimum speed attainable varies considerably with power, it is necessary to examine the aircraft's capability in terms of parameters that interrelate lift, drag, and thrust (or power). This can be done by means of an operating envelope for steady state flight. Such an envelope was developed from the basic curves of figure 2 and is presented as figure 6. On the left, the envelope is given in terms of flight-path angle versus airspeed, which is useful in analyzing the aircraft's performance (obstacle clearance, landing distance, etc.), flow- ever, the pilot does not normally have a flight-path indicator, and an alter- nate presentation in terms of rate of climb versus airspeed is shown in the right-hand figure.

Minimum speed, Vmin'- The minimum speed line of figure 6 represents the lowest speed to which the aircraft is controllable in steady flight with each thrust level. The minimum speed in some aircraft may be coincident with a conventional stall, but in others it may be established by a control limit, or the onset of objectionable buffeting, undesirable pitching or rolling moments, or a rapid increase in sink rate. One particular aircraft, which l0 had its wing immersed in the slipstream of opposite rotation propellers, exhibited no noticeable behavior other than airspeed and sink rate slowly increasing, indicating that the aircraft had exceeded its maximum lift capabilities.

Power or thrust Mox_mum 800 -- 4OO \ \ \

\ \

\ \

\

\ \

Qch

\

\ \

\ \

-1200 Figure {i.- Operational envelope for landing configuration (ref. lO 6f = 98 ° ) steady-state flight.

(:limb and descent capability.- The upper and lower boundaries of the envelope in figure 6 are established, respectively, as the climb capability with maximum available thrust applied, and the maximum descent capability with flight idle thrust. The envelope is bounded on the right side by the struc- tural limit imposed on the configuration; in this case, the maximum flaps-extended speed, VFE.

Included on the figure are lines of constant power or thrust and lines of constant angle of attack. The local slope of the constant thrust lines indicate whether the aircraft is on the front or back side of the drag- velocity curve. Operating STOL aircraft on the back side of the drag- velocity curve has not posed the problem that has occurred with conventional aircraft where thrust cannot be used to rapidly develop normal acceleration.

ii

Restrictions Imposedon the Operational Envelope for Safety

The operational envelope developed in the earlier section represents the

aerodynamic capability of the aircraft if one does not consider powerplant

failures, safety margins, performance, or handling qualities. Safety margins

must be imposed on this operating envelope to establish safe approach, land-

ing, and take-off speeds. In the initial portion of the following discussion,

it will be assumedthat the propulsion system is interconnected to maintain

symmetry and that the handling qualities are acceptable. Later, the effects

of asymmetrywill be discussed, Propulsion system failure.- Reference 3 pointed out that safe operation with the most critical power-plant system inoperative will continue to be required for commercial transports utilizing powered lift. When a propulsion system failure causes no asymmetry (e.g., cross-ducted or cross-shafted) and no change in the minimum speed boundary, the only restriction to the envelope is a decrease in the available climb gradient. This effect is shown in figure 7 for the aircraft of reference I0. To utilize this envelope with such aircraft, the propeller control system must be highly reliable and be safeguarded so that malfunctions have relatively minor effects on controllability of the aircraft.

Minimum speed margin.- The most important restriction to the operating envelope is the margin that must be Po_er or t_ruS t maintained from the minimum speed line. This margin is required for one ,_ "%13_,lmurr' , 4 engines or more of three reasons: First, to provide a range of airspeeds or angles of attack that would allow the pilot to maintain adequate control of the aircraft when wind shears are encoun- tered or when he inadvertently a]lows the approach reference parameter to deviate from the desired value; second, to provide protection from gusts which might momentarily increase the angle of attack or decrease the airspeed; and, third, to provide a maneuvering and flare capability.

Table II presents the margins required by the various aircraft tested and -12 -- indicates the reasons for selecting these margins. General]y, a 15- percent margin above the minimum air- 16 L I I , I 30 40 5'0 60 70 80 speed was selected when sufficient V, knc% normal acceleration was available for the landing flare. The effect of this Figure 7.- Loss of engine; symmetry maintained. margin on the envelope is shown in figure 8.

The 15-percent margin was suf- ficient to account for inadvertent speed excursions, wind shears, and gusts encountered during the tests, and also to permit maneuvering other than complete flaring of the aircraft.

It was not possible to evaluate each of those requirements separately; therefore, the margin is presented as a singular value. For the example aircraft used in figure 8, the 15- percent margin resulted in an approach speed of 60 knots at a 6 ° descent angle. It was calculated that this margin permitted any of the following: <2 _L "4'A' %_" (i) a vertical gust of l0 knots with- out buffeting, and larger magnitudes without exceeding maximum lift and I I I control limits; (2) an instantaneous -1630 70 80 40 50 60 7-1/2 knot horizontal airspeed reduc- ' / krOls tion with an altitude loss of less than 30 feet when power was applied Figure 8.- 'qinimum speed margin.

2 seconds after initial vertical acceleration; (3) an incremental normal acceleration of more than 0.2 g when power was applied throughout the angle-of-attack range including the stall; (4) an incremental normal accelera- tion of 0.15 g when angle of attack was increased rapidly; (5) a steady 30 ° banked turn with the capability of developing an additional 0.15 g by increas- ing angle of attack. Although the STOL flight tests included flying in some turbulence and winds, it is possible that more severe environmental conditions might be encountered in commercial operation and some additional margin might be required. Thus, it is concluded that a 15-percent margin is the smallest that could be tolerated by power-lift transports, and that this small margin would be acceptable because of the ability of STOL aircraft to increase both lift and flight path by power without changing airspeed or angle of attack.

Some aircraft require an addi- tional margin to flare at the steeper / / approach angles because power could #_l,croft / not be increased rapidly enough to O BR 941 A C-8A E develop the required normal accelera- / v > L2 O YC-134A tion. The added margin is indicated Y] NC- !30B rN 367 80 in figure 9 by the increased ratio of z:] cv- 48 approach speed to minimum speed as --- Ref J6 descent angle is increased. In this I I I0 I I -16 -12 0 4 8 figure the flight-derived data points _, deg are compared to the theoretically derived curve of reference 16 for which it was assumed that the pilot would perform the flare utilizing 85 Figure 9.- Ratio of approach speed to minimum percent of the maximum lift capability speed.

and touching down with an excess speed Ma_,:mum, margin of 15 percent. This assumption is conservative for the aircraft 8- 6 ° tested. It should be pointed out that not all of the aircraft required a 4 :_i_ii!_!iii!:iiii!:!i i!!i _iiiiiii!:i!ii!ii_ii:ii:ililiiiiiiilili:i:iliiiiiiilililililili complete flare prior to touchdown.

For example, the CV-48 required no flare at all; the BR-941 used only a "half flare" which reduced the descent velocity of 800 ft/min (at an approach angle of -8 ° ) to a contact velocity - 4 6,_ \ of 300 ft/min. For these aircraft, the speed margin did not have to be increased for the flare. This proce- T, deg 8 •-- i_ dure permits much greater touchdown accuracy and is less demanding of the pilot's judgment of the flare. It Added margin for / should be noted that these aircraft corr, plete flare were not flared by the addition of power because of the relatively long -16L I I _ I I 30 40 50 60 70 80 lags in the engine-propeller control V, knots systems.

Figure lql.- ,\dried margin for com}_lete Clare.

The envelope for an aircraft that requires complete flaring is compared in figure l0 with one that can utilize minimum speed margin because it has good energy-absorbing landing gear.

Available climb gradient in landing configuration.- It is necessary that the pilot have the option of dis- continuing the approach at any time before he initiates the landing flare.

Thus, tentative airworthiness stan- 7, deg dards (ref. 3) require a four-engine 8 aircraft to have a stead), climb gra- dient of not less than 1.8 percent or a rate of climb of not less than 200 2 ft/min in the landing configuration at the approach reference speed or flight-path reference criteria with the critical power-plant system inop- 16 I I 50 ,;_::' 5,::: C/,) 7 0 80 erative. Comparison of figures ii and ., KnOtS 8 shows that this requirement has little effect on the envelope for the configuration chosen. As the propul- I,i£ure ll.- I_,equircm{,nt for climb, i sion system is further vectored and a greater portion of thrust is used to develop lift in steeper approaches, it will be difficult to meet a climb gradient without changing the effective angle of the thrust vector. This is illustrated in figure 12. The normal approach envelope is shown on the left, the envelope for steeper approach in _411 er,q,r r;t L2 Sleeper approach ot --,4 Figure 12.- Change in operational envelope with flap or thrust vector-type control.

the middle, and for wave-off, on the right. A positive climb gradient could not be obtained in the steeper approach configuration (middle figure) with an engine inoperative. To obtain a good positive gradient, the envelope was shifted vertically by changing the configuration (with a thumb switch mounted on the throttle) to the normal approach and then to the wave-off configuration (ref. lO). The minimum speed at each thrust level was relatively unchanged, the safety margins were not reduced, and the adverse moment changes were small. Under these conditions such a thrust vectoring change was satisfac- tory. Similar vectoring can be obtained by wing tilt, or direct vec- toring of the thrusting source.

Maximum sink rate.- The next 7, deg restriction to the operating envelope is a limitation of sink rate to the maximum usable by the pilot while close to the ground. Experience with STOL and other aircraft has indicated that pilots are reluctant to exceed a rate of descent of i000 ft/min when below an altitude of about 200 feet.

Even in VFR conditions the time avail- able for making decisions becomes too I I short and the judgment required to -i6 8O 3C, 40 5,0 GO 70 execute the flare properly becomes knots excessive. Application of this restriction to the operating envelope is shown in figure 13.

Fi_,,,ur_ ]3.- I.]lllit rate of descunt.

Flight-path adjustments.- Another important factor in the selection of the maximum approach angle of a STOL 4 - aircraft is that the pilot be provided a margin of descent capability that [ will allow him to adjust the flight O I' path and touchdown point. This is particularly important for IFR y, deg

, : ililiiii !ii approaches because if the pilot over-

shoots the glide slope during capture or rides high on it while tracking, he must be able to get back to the glide -8 slope centerline without incurring a large speed increase. To perform this task it is necessary to change the -12 -- flight path by 2°; this margin is shown in figure 14.

-16 • I I ' I I Touchdown attitude.- The final 30 40 50 60 70 80 limitation (fig. 15) is not a restric- V, knots tion at all, but indicates that the landing distance may be increased dis- Figure 14.- Flight-l_ath margins. proportionately if the aircraft is not in a satisfactory touchdown attitude upon completion of the landing flare.

Per for mclnce penalty This condition can arise when the approach speed is increased to obtain additional normal acceleration to flare the aircraft or to improve the

............................. i)

handling characteristics in gusty environments. If the touchdown atti- tude is incorrect, it is necessary to ! J ":_: increase angle of attack and attitude ),, deg slowly as airspeed decreases until the touchdown can be satisfactorily made.

This problem is not unique to STOL aircraft, but is more pronounced because of the relatively large changes in angle of attack that are associated with a given change in velocity (i.e., a 5-knot change in -12 -- airspeed results in a S ° to i0 ° angle of attack change for balanced flight).

63L0 I I J I I -I 40 50 60 70 80 Considerations of asymmetry.- V, knots Figure 16, obtained from reference 5, shows the effect of a propulsion l:igure 15.- Penalty due to improper touchdown attitude. system failure that results in thrust asymmetry. The left-hand sketch shows the effect of an engine failure on the climb gradient and minimum control speed, and the right-hand sketch shows the result of imposing the previous Maximum power ._`.:::::::::::::::::::::::::::::::::i:_:_:_:_:i:!:!:_:_:i:i:_:_:?_:i:i:;_:?i:i:i:_i_ii!!]!_[i!_i!_i_iii!!i_: 4 . 4 engme Performance ...... _i_iii_i_iii1iiiii_i_iiiiiiiiiii!i_iiiiiiiii!i_iiii_ii_!iiiiii;!i_!_i_i_i_i_i_i_i_i_ii_i_!_i_i_i_iiiii;!i_!_ii_i;iii_ii _ j s eog,ne pena.y ;_ ::_]'!i Moderal e i _ --Large _2 L I I I ! I I • I I I I I I 40 50 60 70 80 90 iO0 40 50 60 70 80 90 I00 V, knots V, knols Figure 16.- Engine failure creating thrust asymmetry.

safety margins. It is seen that a propulsion system failure that causes asym- metry severely limits the STOL operation. Operating at the higher speeds required for safety also results in a significant performance penalty because of the improper attitude on touchdown (discussed in an earlier section). A further consideration is the deterioration of handling qualities when asym- metry occurs. It has been found that the minimum control speed for STOL air- craft can be limited by the lateral control as well as the directional control.

Summary of restrictions.- When all of the previous restrictions are imposed upon the operating envelope, they leave a rather small area from which a safe approach speed can be selected that will still produce reason- able performance. It is important that this situation not be misinterpreted to indicate a lack of flexibility. The small area remaining represents the steady-state situation from which transient excursions can be safely made into the shaded areas, and changing the configuration (as sketched in fig. 17) provides the wide latitude for adapting the aircraft to meet the needs dictated by the operational environment.

Cruise Landing Wave off Chmb 8_ Take-off [d ;,";,';,'::, :; t ___J 20 L I _ • I I • I I I • _ I I 40 60 80 40 60 80 40 60 80 I00 I00 ! 20 140 160 V, knots V, knofs V, knots V, _no% Figure 17.- Changing envelope to suit operational environment.

It is apparent that a single airspeed will not suffice as an adequate reference for all approach angles; however, the operational envelope can be examined to determine the best flight-path reference criteria. For the example STOL, lines of constant angle of attack are alined roughly parallel to the region of desired operation. The pilot chose an angle of attack near 0 ° to provide an adequate margin during the approach. This reference angle of attack had the added advantage of being independent of gross weight and configuration for the example aircraft.

Conclusions STOL aircraft can utilize the propulsion system to develop a significant portion of the lift in the approach.

For the aircraft tested a representative level of thrust-to-weight in the approach was 0.2. In this mode, modulation of power produced more normal acceleration than horizontal acceleration.

A single airspeed could not be used as a reference for all flight-path angles of the aircraft examined because of the strong influence of power on the minimum speed. A method is presented whereby the low-speed flight envelope can be analyzed to determine a suitable speed, angle of attack, or other flight-path reference to establish a safe operation.

A minimum speed margin of 15 percent was used as protection from gusts and to assure maneuvering and flare capability. This small margin could only be used when large aerodynamic changes were not caused by an engine failure, when power produced an incremental normal acceleration of 0.2 g, and when modest flight-path angles were used in the approach. As flight-path angle was increased in the approach, the margin had to be increased to permit suf- ficient normal acceleration if a fully flared landing was required. A posi- tive climb gradient was desired in the landing configuration with one engine inoperative; however, a configuration change can be permitted if there is no loss in lift and there are no adverse moments. When the aircraft is less than 200 feet above the runway, the sink rate should be less than 1000 ft/min.

A flight-path margin of 2 ° is required to steepen the flight path beyond the normal approach value. Increased airspeed margins for increased normal acceleration can create performance penalties by causing improper touchdown attitude.

When large aerodynamic changes are caused by an engine failure compa- rable margins must be imposed on the characteristics after engine failure, and these restraints can limit STOL operation. The minimum control speed for STOL aircraft can be dictated by lateral control rather than directional control.

DISCUSSION OF FIEI_DI,t:NCI'IIFACTORS

Landing Field Length

Consistency' of performance.- In the previous section, margins were discussed that would provide safe approach and landing speeds for STOL air- craft, llowever, there are additional operational aspects that must be con- sidered. For example, what can be done to assure that routine operation is possible under all runway and atmospheric conditions when a landing strip length, say 1500 feet, has been established? Further, is there some way that the demonstrated landing performance can give answers closer to the opera- tional performance? Finally, how do handling qualities affect tile ability to operate into STOL strips?

At the present time, insufficient flight data are available to verify tile consistency in STOL landing performance over a range of approach angles and speeds, runway conditions, and atmospheric conditions. The effect of handling qualities on landing performance has not been studied systematically.

It has been observed that, with a skilled pilot under favorable conditions and without constraint of a touchdown area, adverse handling qualities will not greatly affect the minimum landing distances, but will merely increase pilot workload. On the other }land, adverse handling qualities can significantly affect landing performance if tile aircraft is disturbed by gusts, the touch- down area is confined, or the pilot is less skilled. Consequently, less importance should be placed on minimum landing distances, and more importance should be placed on measuring the landing distance with a realistic task.

This is discussed in a later section.

Figure 18 shows grossly the Touchdown c,lbpplng orec_ area landing performance that can be obtained with a STOL aircraft over a wide variety of conditions. This fig- ure contains measurements of 60 land- ings of the Breguet 941 over a 50-foot

I

simulated obstacle at the threshold.

The measurements were made for differ- 2o0 ent approach angles and various cross- ,., 4r)S , 800 4)mtance from lnres'_old, ft winds, and included pilot training.

The aircraft touched down 350 to 500 feet from the threshold and stopped I igLlr_ lg.- \ariation in lamling distance over 550 to 850 feet from the threshold.

a _,.'idc variety of conditions.

When it is realized that the range of stopping distances is only about three airplane lengths, it is seen that good STOL landing performance can be obtained under various operating conditions.

To examine the individual effects of different runway and braking conditions, landing techniques and pilot judgment, the landing performance over a 35-foot obstacle was calculated for an aircraft like the Breguet 941.

Figure 19 shows the effect of different runway and braking conditions. The calculations were based on a 40,000-pound propeller-driven STOI, aircraft approaching on a 7-1/2 ° glide slope i at 60 knots. It was assumed that 4 PrOps rn reverse + brakes one of four engines was inoperative (propulsion system is interconnected 4 Proog in reverse, no brakes for symmetry and control), and there was a 1-second delay between touch- - -_ 2 Props in reverse t + brakes down and deceleration. The solid I ] 2 Props thrusting bars are for a dry-prepared surface Br0keson_y while the dotted bars are for a slip- --" pery surface such as ice. The top bar is with all propellers reversed 400 800 _200 FeLO0 and antiskid brakes applied. The Total d_stance from 55ft to slop, ft calculated distance over a 35-foot obstacle to a stopping point on a dry Figure 19.- Effect of different braking conditions surface was 700 feet. This compares on landing performance.

to the 550- to 850-foot distance over a 50-foot obstacle measured in flight for similar conditions. This performance was obtained with an approach speed consistent with all of the restrictions placed on the operating envelope presented earlier. Thus, good STOL performance can be obtained with a high degree of safety. For this case, the average deceleration during the ground roll was about 0.4 g which was not uncomfortable to the passengers because the deceleration increased smoothly. An icy surface is estimated to increase the distance by only 150 feet, about two airplane lengths. The second bar is for all propellers reversed and no brakes. The third bar is for the condition of one propeller not reversing; the opposite propeller is positioned at the same blade angle by safety features incorporated with the interconnect. The last bar is for brakes only, with the propellers producing near zero thrust. For this case, where propellers are not reversed, a slick surface like ice is quite detrimental. Figure 19 shows that the landing performance does not change appreciably over a wide range of failure and runway conditions provided symmetrical reversing is utilized.

Figure 20 shows the effect of '_,kqots y, deg different piloting techniques or eo -7_/21 judgement. The top bar is for the basic case shown in the previous

|

70 7//2_Part,alflures figure. The next bar shows what happens when the pilot is 10 knots _ fast, or there is a lO-knot tail wind. The increase is only 150 feet, less than two airplane lengths. The .L: :;-_.}" L? :¢.| otentlal -7 _/2 FuB fla_ next bar reflects the small increase in air distance if the airplane is 7O -7t_ r_,flo,e,flown on a 6 ° glide angle instead of _ver_ attilude a 7-1/2 ° glide angle. All of these 1 I i 1 I distances are for partially flared 400 BOO _200 landings to reduce the vertical Total d_stance from 55 fl to stop, ft velocity at touchdown to about 5 ft/sec. The next two bars repre- Figure 20.- Effect of judgement and techniques on landing performance.

sent the total landing distance for 2O fully flared landings. There is a small increase in air distance; however, this is not the main concern. A fully flared landin_ must be initiated at a greater height above the runway, and it is difficult to judge precisely where the aircraft will touchdown. If the aircraft is rotated too soon or too much, the floating distance will be a large portion of the ground roll. It was noted in the previous section that if an added speed margin were applied, the contact attitude could be quite unfavorable and it would be necessary to decelerate in the air; this effect on performance is shown in the last bar.

Demonstration of landing performance.- The landing performance for STOL aircraft should be demonstrated under conditions close to an operational environment and factors pertinent to that craft should be used for determin- ing the operational field length. The flight path should be constrained to a designated obstacle clearance angle as well as a designated landing area, __ _ / Lateral offset and a task should be included to expose adverse handling characteris- tics. This is in contrast to the _"---'_ i ', I current procedure of FAR 25 and 121 _--_.'_ _. __ Re_no_e hood (refs. 17 and 18) of permitting a "maximum effort" landing demonstration anywhere on a dry runway and then dividing this distance by 0.6 to cover operational environments. A different Vert call K,,___ Touchdown offset _ / method is recommended because one fac- //S,op tor cannot cover the effects of gust, wind shear, runway condition, and Landing d_stance it: landing technique for all STOL air- craft. One possible method of demon- Figure 21.- l)emonstrntion of landimz performance.

strating the performance for STOL aircraft is shown on figure 21.

The top part of this figure represents a plan view of the approach and landing area, and the bottom an elevation of the same area. The solid lines represent the center line of a visual landing aid set at the glide angle desired for certification and intercepting the runway near the intended touchdown point. The short dashed lines represent an ILS type system set at the desired flight-path angle but offset from the visual aid. The pilot would start an ILS approach, and at an altitude of about 200 feet, he would remove his hood, correct the offset, stabilize, and continue to a full stop. Another possible method might be to offset the ILS at an angle, say 20 ° , rather than a lateral distance. In either case, a specified number of landings could then be used to determine the demonstrated landing performance. Phototheodo- lite coverage would document the performance and disqualify landings made when the aircraft went below the designated obstruction angle.

Field length factor.- A singular factor such as 1/0.6 should not be applied to the demonstrated performance to obtain an operational field length.

Instead, factors should account for the method of operation (e.g., partial versus fully flared landing), sensitivity to disturbance (e.g., aspect ratio and wing loading), runway surface, and whether or not reverse thrust is used.

At the present time, insufficient systematic data have been obtained to

formulate field length factors. If the critical case is with one engine

inoperative and only two propellers in reverse, according to figures 18 and

19, the average demonstrated landing would probably be about 900 feet with a

partial flare. An appropriate field length considering cross winds, tail

winds, and wet runway, would be about 1200 feet if satisfactory handling

qualities were provided. The corresponding factor would then be about I/0.7S.

A visual guidance system should also be used at the airport runways so

that the pilot has a flight-path reference at least to the threshold. Such

a reference is an important assistance in steep descents to avoid large

last-minute corrections and to safely utilize the maximum performance of

the aircraft.

Obstacle clearance angle.- Since STOL aircraft will operate in restricted airspace, appropriate obstacle clearance margins are of importance. There are at present insufficient data to determine safe margins. _en exanining the aircraft descent capability, not only must the planned approach flight path and obstacle clearance margin be considered but an additional descent margin of 2 ° is necessary for correcting tracking errors or tail wind (discussed earlier).

Take-Off Field Length Take-off gradient.- Performing STOL take-offs with a steep gradient profile requires a large thrust-weight ratio, and a high acceleration occurs even at lift-off speed. This is shown in figure 22 by a take-off time history from reference i0.

The time from brake release to rota- _oo i: i_! !

tion speed, VR, computed critical decision eo _v,, vR_ .............. "_"-_ Vc, knots 40 : [ it :i],#4' i ::':: : : speed V I, and optim_n climb speed, V2, is 60 ,,-, , _ ___'!f_i_.T :sk_°_s's_c short, and since the propulsion system was 20 _ ::s _!__ Li interconnected, recognition of an engine failure was difficult. The primary effect 20 I! ! _ !!iT ::-}::_ I !I t']___i: } of an engine failure on this aircraft was

oeg ,o o ii/;ii!t ii i liiiiI

a reduced climb gradient, The minimum climb gradient will have to include an appropriate obstacle clear- ance gradient which will be related to the 0 4 8 12 16 20 24 28 32 area in which the aircraft is intended to _, sec be operated. For a usefui STOL aircraft, Figure 22.- STOL take off.

it seems reasonable to expect the safe take-off climb gradient to be of the same magnitude as the landing gradient; therefore, it would be expected that gradients of at least 6 ° will be required when an engine is inoperative.

Take off field length factor.- For a steep gradient STOI, aircraft a large thrust-weight ratio will be installed, and the runway conditions are not as critical as they have been with some subsonic jet transports; however, the rejected take-off and one-engine-out climb performance will have to be examined as in current regulations (refs. 17 and 18). For the example STOL craft used in the previous sketch the take-off distance to 35 feet with an engine stopped at VR is 900 feet. The corresponding start-stop distance was computed to be 700 feet. In comparison, the take-off distance to 35 feet with ali engines operative is 800 feet.

Terminal Area Operation Since STOL aircraft will be required to operate in restricted airspace some knowledge of the minimum pattern size that can be safely flown is needed.

Examples are provided in reference i0. Under VFR conditions, it was found that 90 ° turns into the final approach could be performed as low as 300 feet at 65 knots and the maximum operational bank angle was about 30 ° . This corresponded to a radius of less than 1000 feet.

The take-offs and climbouts were simple to perform under VFR or IFR conditions; the procedures and handling characteristics were similar to those for a conventional turboprop transport.

The importance of providing good handling qualities for IFR operation was noted in references 7 and 10. When these were provided, steep approaches at STOL speeds could be safely made in IFR conditions to altitudes of 200 feet with unsophisticated guidance or display systems.

Conclusions Less emphasis should be placed on demonstrating maximum performance.

More importance should be placed on demonstrating consistent performance with a task that simulates environmental conditions that may be encountered in routine commercial operation and that exposes adverse handling characteristics.

Insufficient systematic information is available for relating field length factors to operational considerations and handling qualities.

DISCUSSION OF HANDLING QUALITIES Background Several reports have been published on handling qualities criteria for V/STOL aircraft (e.g., refs. 19-21); however, these have been primarily oriented toward military missions and requirements, and their acceptance has been limited because of the inability to verify the criteria by flight exper- ience with representative V/STOL aircraft. The present report is directed

toward providing the regulatory agencies with criteria for safe operation

of STOLaircraft in a commercial environment.

The criteria presented herein are based on pilot opinion and quantitative

data accumulated from flight investigations of STOLaircraft. It was assumed

that these aircraft will operate under instrument conditions and will be

required to maneuver in confined areas where turbulence and wind shears are

likely to be encountered. Criteria are not presented for each facet of han-

dling qualities because there was insufficient data. The major emphasis and

greatest quantity of data is on the lateral-directional handling in the

landing-approach regime because this area has caused the greatest difficulty

and is the most demandingportion of the flight. Limited information is given

on longitudinal handling as well as on the wave-off and take-off regimes.

The criteria are presented separately from the data used in their

formulation. The section containing the data presents considerable discussion

to substantiate the criteria and to showhow they were formulated. Criteria

related to aircraft response to control input will be presented first because

it is most important that the aircraft have a good control system and good

control characteristics in order to operate satisfactorily in the severe

environment anticipated with the low levels of stability and dampingnormally

present at STOLspeeds. A succeeding section presents criteria related to

aircraft response to external disturbances.

Since handling qualities are judged by pilots' opinions, criteria were

formulated so that they could be easily recognized and appreciated by the

pilot, readily evaluated for compliance, and they would include the effect

of factors that influence the response or behavior of the aircraft. The task

of providing meaningful criteria is compromisedby manyproblems which include

the difficulty in measuring the parameter in question, the interaction of

several variables upon pilot opinion, the inability to vary the pertinent

parameters systematically, and the fact that the pilot seldom has the opportu-

nity to evaluate the aircraft behavior under all environmental conditions. It

must be recognized that the handling qualities criteria given are limited in

scope and will require continued revision as new aircraft concepts with more

advanced control and stabilization systems are developed and tested.

Level of Criteria

A question arises as to which level of pilot acceptability the criteria

should reflect for commercial transport operation. Two levels have been

indicated in the criteria: one level should be exceededto obtain satisfac-

tory handling in IFR as well as in VFRflight; the other represents the

lowest level for a particular parameter that a pilot can tolerate without

significantly compromising the task or mission. In this report the pilot's

opinion of a specific parameter or aircraft behavior was evaluated according

to the Cooper Rating Scale first presented in 1957 (ref. 22). This scale,

shownbelow, is a shorthand method of indicating pilot opinion; however it

had several shortcomings when these handling quality evaluations were related

to commercial transports. These shortcomings included the introduction of

stability augmentation and undefined failure modes, the concept of normal and

Primary Can be Operating Adjective mission _Numerical Description conditions rating rating accomplished landed Yes Yes Excellent, includes optimum Yes Yes Normal Good, pleasant to fly Satisfactory Yes Yes Satisfactory, but with some mildly operation unpleasant characteristics Yes Yes Acceptable, but with unpleasant characteristics Emergency Yes Doubtful Unacceptable for normal operation Unsatisfactory operation Doubtful Yes Acceptable for emergency condition only I Doubtful No Unacceptable even for emergency condition l Unacceptable No No Unacceptable - dangerous No No No Unacceptable - uncontrollable operation lO Motions possibly violent enough to No No Catastrophzc prevent pilot escape IFailure of a stability augmenter.

emergency operation, and the separation of the landing task from the primary mission. Additional discussions are contained in reference 23, and a revised scale is presented. The revised scale could not be used in this report because the majority of data were drawn from previously published reports.

It was determined that the level for satisfactory handling corresponds to a rating of 3-1/2 for both the original Cooper scale and the revised scale; the second lower level of criteria corresponds to a rating of 5 to 5-i/2 on the original Cooper scale and a 6-1/2 on the new scale.

If the aircraft's handling qualities exceed the satisfactory level for each criterion, the aircraft should be capable of performing its mission under a wide range of environmental conditions. This is not true if the aircraft has several parameters that fall into the lower tolerable level of criteria. Then the pilot workload might become high enough to endanger the safety of the aircraft even though each parameter could be tolerated by itself. A method has not been found to sum or weigh properly these ratings of individual characteristics to arrive at an overall rating for complete mission phases such as "approach" or "landing"; a separate judgment and rating is required by the pilot.

Criteria for Aircraft Response to Control Input Factors included in criteria.- The following comments pertain to all axes; in succeeding sections detailed discussion will be made for each axis.

Control power: The need for control can generally be divided into three separate requirements; trim, maneuverability, and stabilization (largely due

to environmental disturbances). In somecases, these requirements may be

additive; however, in most of the STOLaircraft evaluated, they appear to

be dominated in each of the axes by only one of the requirements. It is

recognized that these requirements may be different for aircraft configura-

tions which are not similar to those evaluated, and that they may be changed

by stability augmentation systems. The criteria are not related to gross

weight because it has been assumedthat the aircraft of interest are in the

30,000 to I00,000 ib range where the control power requirements are fairly

constant.

Control sensitivity: In addition to adequate control power, it has been

found that control sensitivity plays an important role in the pilot's ability

to control the aircraft. The maximum control deflections presented in the

criteria are not meant to provide optimum control sensitivity of the system

but rather to define a lower limit, assuming that the aircraft response to

control deflection is linear.

Linearity: The requirement for linear aircraft response to control

deflection is not well defined. Figure 23 shows three types of system charac-

teristics that were present on the STOLaircraft tested.

c_ _Lp Figurc 23,- Illustration of different control systems.

When an abrupt increase in roll response occurs as the control deflec- tion is increased, as in the right-hand sketch, there is a marked tendency for the pilot to overcontrol the aircraft laterally; this tendency has not been noted in those aircraft for which the nonlinearity was in the opposite direction, as in the middle sketch. Although the pilots have not commented on undesirable nonlinearities in the other axes, similar results would be expected.

Mechanical characteristics of control systems: The primary flight control systems of STOL-aircraft should be designed so that the pilot can easily operate them with one hand while he adjusts thrust or power to control flight path with the other. Such items as breakout forces, friction, force gradients, sensitivity, control harmony, linearity, lags, and inertia all influence the pilot's opinion of the aircraft's handling qualities. In some cases poor mechanical characteristics have completely masked the aircraft's inherent stability and the pilot's impression of controllability.

The maximum control forces specified in the criteria are consistent with the concept of one-hand operation. They combine the individual effects of breakout, friction, inertia, and gradient forces; simplified criteria are not intended to imply that each of these items is not important in its own right.

The mechanical characteristics of the control system of each of the aircraft tested and some comments are presented in table III.

Cross coupling: Cross coupling is usually covered in the discussion of stability; however, it will be included in the response section because it is generally apparent to the pilot as a result of his control inputs. The intent of the criteria is to minimize undesirable disturbances about and along axes other than the one the pilot is controlling.

Apparent damping: The criteria for damping covered in the response section includes the effect of control system lags which the pilot finds difficult to discern and separate from the aerodynamic damping of the vehicle.

The resulting apparent damping affects the pilot's ability to control the attitude of the aircraft precisely.

Lateral control criteria.- The following table presents the proposed criteria for aircraft response to lateral control inputs at the STOL refer- ence speed or angle of attack.

Parameter to l,evel for l_evel for Item be measured satisfactory operation safe operation Time to 30 ° bank angle No more than 2.4 sec No more than 2.9 sec Roll acceleration blore than 0.4 rad/sec 2 Hore than 0.3 rad/sec 2 1. Control within 1/2 see power Maximum control No more than 60 ° wheel No more than 90 ° wheel deflection deflection or 5-in. deflection or 7-in.

stick deflection stick deflection Maximum force to 20 lb 40 it) 2. Force achieve item 1 Roll acceleration per Should not increase Insufficient data 3. IAnearity unit stick deflection (_5 _s/A _ ) max 0.3 0.6 4. Cross &t) Not noticeable Not objectionable coupling an/g Less than -0.1 Less than -0.2 5. Apparent Number of control No more than 2 No tendency for pilot roll reversals to induced oscillation dmnping stabilize The maneuvers to test compliance with all criteria should be initiated from trimmed, wings-level, nonturning flight, and should be performed in both directions. Compliance with the criteria of items i through 3 should be demonstrated by performing abrupt rudder-fixed, lateral control steps of increasing magnitude to the limit of control authority or until the required response is achieved. The control power criteria were developed to provide some measure of the ability to maintain the desired bank angle in turbulent air. Control in turbulence has been the most critical requirement for lateral control at the approach and take-off speeds. The time should be measured from the initiation of the control action by the pilot.

An engine failure on a STOL aircraft without an interconnected propulsion system can cause an asymmetric rolling moment that may be more critical than the yawing moment, particularly if a large part of the lift is developed by the propulsion system (refs. 4, 5, and 12). To maneuver adequately during an approach with an engine inoperative, sufficient lateral control must remain to satisfy the criteria for safe operation with the remaining engines at the power level required for the selected approach flight path angle.

Less maneuvering is required during the wave-off and take off, and the lateral and directional control with an engine inoperative can be reduced to the level implied in reference 3.

Compliance with items 4 and 5 should be demonstrated by performing abrupt, rudder-fixed turn entries to bank angles of at least 20 ° . STOL air- craft which have low directional stability can develop high sideslip angles during maneuvering which hinder the pilot's ability to make precise heading changes and accurately control sideslip during crosswind landings. This problem of cross coupling (turn coordination or "adverse yaw") is best cor- related by the ratio of peak sideslip excursion to the bank angle (_B/A_) developed during rapid turn entries.

Lift losses caused by spoilers used for lateral control have created minor coupling problems. Such losses have been related to the incremental normal acceleration, measured at the center of gravity, with maximum control deflection.

Apparent roll damping cannot be easily defined by classical means because it includes the effects of both the aerodynamic characteristics of the airframe and the mechanical characteristics of the control system. It becomes troublesome to the pilot when he cannot easily arrest and stabilize an established roll rate. In the more extreme case, it may be manifested by a continuous roll oscillation which is sustained by the pilot's control activity. Because of insufficient knowledge of the interaction of these characteristics, the criterion is presented in a qualitative rather than in a quantitative term.

Directional control criteria.- The following table presents the proposed criteria for aircraft response to directional control inputs. Data to sub- stantiate these criteria are presented in a later section. Compliance with the criteria should be demonstrated by steady-state sideslips performed at a constant heading and by abrupt rudder pedal steps with lateral control fixed at the trim position. Tests should be performed in both directions.

The most critical requirement for directional control has been the ability to trim the aircraft, and this has been primarily manifested in com- pensating for a crosswind component in the approach and landing. Figure 24 shows that large crab, or sideslip angles, are required for moderate cross- winds at STOL speeds. Since this has serious design implications, and since the crosswind that may be encountered is unknown, the control power criteria for a satisfactory level of steady-state sideslip angle is not given in a quantitative form. The criteria also specify an angular response to rudder pedal deflection. This is important to enable the pilot to rapidly decrab

Parameter to be Level for satis-

Level for safe

Item

measured

factory operation

operation

Steady-state

No less than 15 °

s n'(v)

sideslip angle

i. Control Time for 15°

2.2 sec 3.1 sec

power change in head-

ing

Maximum pedal At least

At least ±2-1/2 in.

deflection

±2-I/2 in.

Force to achieve

Less than 150 ib Greater than 50 ib,

2. Force item 1

but less than i00 ib

Variation of side-

Linear to specified sideslip angle. At

3. Linearity

slip angle with

larger values, increased pedal deflec-

pedal deflection

tion for increased sideslip

Effective

Positive, but less Positive, but

dihedral

than 50 percent less than 75 per-

4. Cross

maximum lateral cent maximum

coupling lateral control

control power Not noticeable

Responseabout

Not objectionable

longitudinal axis

6O the aircraft prior to touchdown and to quickly reduce unwanted sideslip angles that occur during maneuvering.

The time for a 15 ° heading change was 40 k chosen as an indication of the direc- tional response; this time should be _, deg Cross wind measured from initiation of a pilot 30_ \ compo_enl, input starting at trim sideslip angle.

knols 2o-- _ Another critical requirement is __ 20 _ that the directional control counter- _0 act an asymmetric moment caused by a i i propulsion-system failure. For the 040 I I 6O ,_ _2o STOL aircraft without interconnected V, knots propulsion systems, large asymmetric moments occurred with a powerplant Figure 24,- l:ffect of cross wind at low airspeeds.

failure; it is proposed that sufficient control be available after failure to satisfy the level for safe operation specified in the previous table.

The most prevalent cross-coupling effect resulting from directional control input is the lateral response or dihedral effect; the concern is that sufficient lateral control power be available at the maximum sideslip angle specified to insure adequate control of bank angle.

Longitudinal control criteria.- The following table presents the pro- posed criteria for satisfactory aircraft response to longitudinal control inputs. All the aircraft tested had adequate response and damping; therefore, it was not possible to define a lower acceptable level. Compliance with these criteria should be demonstrated by performing abrupt longitudinal steps and abrupt attitude steps of at least I0 °.

Level for Parameter to be Level for satisfactory Item measured operation safe operation Less than 1.2 sec Time for i0 ° attitude change Pitch acceleration More than 0.5 i. Control within 1 sec rad/sec 2 power No more than ±5-in.

Maximum control Insufficient column deflection deflection information to provide 40 Ib Maximum force to criteria 2. Force achieve item 1 Should not increase Pitching acceleration 3. Linearity per unit stick displacement No more than one Number of control 4. Apparent reversals to damplng stabilize The longitudinal control power requirements may be dominated by either maneuvering or trimming depending upon the aircraft configuration and the nature of the approach and landing technique that is used. The maneuvers that required the greater longitudinal control were rotation at take-off and flaring at landing. The level of control power specified in the criteria should be satisfactory for any longitudinal maneuvering that might be consid- ered for a commercial transport. When abrupt attitude changes may be objec- tionable, other methods of developing normal acceleration should be considered; criteria for other methods are discussed in the section on flight-path control.

The longitudinal control system must also be capable of trimming the aircraft throughout the flight envelope. It must be possible to attain the 3O

minimumspeed as defined in the section "MinimumSpeed Margins" under all

conditions of weight and loading for which the airplane will be operated.

Sufficient nose-down pitching response must be available at the minimumspeed

under all power conditions to effect a satisfactory recovery. An additional

requirement is needed to insure that the pitch attitude can be controlled

adequately prior to touchdown. For the latter requirement it is proposed that

the longitudinal control be sufficient to trim the aircraft at the desired

attitude in ground effect at a speed corresponding to the approach reference

criterion minus 5 knots.

The longitudinal control requirements for sensitivity, force, linearity,

and apparent damping are comparable to those for lateral control. Although

it is noted that harmony between lateral and longitudinal control should

exist, no related criteria are presented.

Flight-path control criteria.- One of the characteristics of powered- lift aircraft is that less normal acceleration is available from longitudinal control, and the pilot must use additional methods of developing normal acceleration for controlling flight path. Some consideration must be given to the response characteristics of these other methods of control so that the pilot can adequately control the aircraft's flight path, particularly during the approach and landing. The following criteria are divided into three modes of flight-path control: Mode A. For flare and touchdown control when an incremental accelera- tion of less than 0.15 g can be developed by longitudinal control.

Mode B. For flight-path tracking when an incremental acceleration of more than 0.15 g but less than 0.30 g can be developed by longitudinal control.

Mode C. For gross flight-path changes including wave off, regardless of the normal acceleration developed by longitudinal control.

In order to determine whether the criteria for Mode A or Mode B apply, abrupt longitudinal control steps should be performed with the aircraft trimmed at conditions for the flight-path angle selected for the performance demonstration. Mode C applies to all aircraft. Compliance with the criteria listed in the table should be demonstrated by performing steps with the flight-path control. The aircraft attitude should be maintained constant with the longitudinal control; the initial conditions are with the aircraft trimmed at the flight-path angle selected for the performance demonstration.

The acceleration should be measured at or near the aircraft center of gravity.

Level for

Parameter to Level for safe

Item Mode

be measured satisfactory

operation operation

A

Incremental nor- ±0.I g Insufficient data

mal acceleration

B

Incremental nor- ±0.I g Insufficient data

mal acceleration

I. Control

Steady-state 6° 200 ft/min

power C

climb angle

All

Incremental 2° greater Insufficient data

descent angle than selected

approach angle

A

Aircraft Achieve IA in Insufficient data

response less than

0.5 sec

Aircraft Achieve IB in Insufficient data

2. Response B

time

response less than

1.5 sec

C

Aircraft Achieve IC in Achieve IC in less than

response less than 4.0 sec

2.0 sec

3. Cross All

Pitching moment Not noticeable I Not objectionable

coupling

Substantiation of Control Criteria

Lateral control power.- Pilots have been more critical of the control of STOL aircraft about the lateral axis than about the other axes. Precise con- trol is required because small bank angles generate large yaw rates at low speeds which quickly produce heading excursions. The ability to maintain the desired bank angle in turbulent air has been the most critical requirement for lateral control of STOL aircraft at take-off and landing speeds, at least for moderate sized aircraft evaluated with all engines operating. This was concluded because less than 40 percent of the available control was used during extensive maneuvering when a satisfactory level of control was present.

Little lateral trim was needed for crosswind landings, and little or no trim was required for thrust asymmetry because all powerplants were operative during the evaluation.

Figure 25 summarizes the lateral control power measured and evaluated with the various STOL aircraft at approach and take-off speeds, The results are given in terms of maximum angular-acceleration capability of the aircraft; that is, the acceleration produced by a step input with zero rate of roll.

The values were measured in flight primarily by aileron reversals as dis- cussed in reference 24. The angular acceleration presented in these figures cannot be related directly to the aircraft response nor to the pilot's impression of controllability; items such as the mechanical characteristics of the control system, pilot's recognition of needed corrective measures, aircraft damping, cross coupling, and Aircraft V, kno_s

8r

sensitivity to gust disturbances all O BR 941 60 I.O 1.6 [] UF XS 55 1.3 play a part in lateral controllability.

A C 8A 7O .8 The interaction and relative impor- z_ C 8A (Ailerons only) 70 .8 tance of these factors is not accu- YC 1&4A 80 .7 rately known; however, these factors _' YC 154A(Aderonsonly) 80 .7

i

V NC 130B 70 .9 were considered in fairing the data of _" NC-150B 85 .8 figure 25 and will be discussed to ['N 567 80 85 1.0 _, 367 80 II 5 .4 some extent in the following para- IO-- _ -- /1 CV-48 55 5.3 graphs. The faired data represent the Q vz-5 60 <1 relation of pilot rating and accelera-

\

o3_ tion for a common damping (_ = 1.0) and a good control system. The con- _o' r°d/sec2 3 6 trol power criteria presented earlier 5 N.3 were not related to gross weight \PR3_ because it has been assumed that the aircraft of interest are in the 30,000 to i00,000 lb range where the control power requirement is fairly o &4_ 7_ constant.

I J IIII 1 03 I I 111 5 I0 30 50 IO0 300 W, lb x I05 The angular acceleration and force characteristics of the different Figure 25.- Lateral acceleration. Pilot rating STOL aircraft are given in figure 26 next to symbol.

along with a summary of comments on these characteristics. Figure 26(a) contains data from the BR 941 tests (ref. 9) in which the control power was changed by different combinations of spoilers, ailerons, and differential propeller pitch and was evaluated at approach and take-off speeds. Satisfac- tory ratings were given for those configurations providing at least 0.4 rad/sec 2. Although the control sensitivity (control power per inch deflec- tion) was different for each configuration, the sensitivity was satisfactory for all cases tested with the possible exception of that rated 7_i/2, and should not affect the control power rating. Subsequent tests with this air- craft in IFR operation and moderate turbulence (ref. i0) showed that 0.4 rad/sec 2 was satisfactory under the more adverse test conditions.

The ratings of the NC-130B (fig. 26(b)) reflected low sensitivity. These ratings were based on an evaluation of IFR approaches in gusty weather with all engines operative. At 70 knots almost full lateral control was required to balance an inoperative engine because the propellers were not intercon- nected and considerable powered lift existed. Insufficient control remained to maneuver the aircraft during approach and landing; therefore landings were not performed with an inoperative engine.

The UF-XS (fig. 26(c)) had satisfactory characteristics in the approach condition with all engines operative. The evaluation of control was limited, and engine-out tests were not permitted.

Even though large lateral-control power was available on the YC-134A with spoilers and ailerons (fig. 26(d)), precise control of the aircraft was dif- ficult because of the rapid increase in response at 30 ° wheel position, the o Q_ E_ o _ o "__ ._ -- o o _.c_ ,_ _D_ o" •_ _ _ _,,.-- o _ _ _._ E- _ , .__o _o -° _- _ 8 0 O _..

o- _ o o c. N=_ N o o o _ T °- 8 2 ._-_ =- _ _ D D I _ a_ mm ii o _.n _ E$ o-- % _ :.,E: ,s, E "- 2 o _ o ° '_= _'_,.

0 .{.2_ _ <[ -" c,.i i..4 E ,j c o oJ (,D c: o o o ii s d c LL o n .:> o I c c.)

O3 U 6O c __1 O0 CO _.0 '_" r'4 0 co £0 _ oJ o f,D :-e.-_ o region frequently used in control of the aircraft. The large increase in force produced when the spoilers were engaged at about i0 ° wheel position and the rapid increase in response at 30 ° wheel position combined to produce unsatisfactory characteristics that masked the control-power ratings of this aircraft.

The CV-48 had satisfactory control power and sensitivity_ however, the lateral damping being a little low created a tendency to overshoot a specified bank angle.

The 367-80 (fig. 26(f)) had more control power than could be utilized in any maneuvers performed. The roll acceleration at large control deflec- tions was sufficiently high that for this large airplane there was some con- cern of possible structural damage. Initial low-speed tests of the 367-80 (ref. 15) were made with an aerodynamic tab control; the control was rated unsatisfactory by the pilot (PR-4-1/2) because of the high force gradient and nonlinearity. Installation of a powered control, with the forces shown in tile figure provided a satisfactory control system. The control character- istics o£ the 367-80 at higher speeds are discussed in reference 25.

Aniliation of force The lateral control characteristics of the C-8A were surprising. For the normal configuration of spoilers plus 8R 9,4. I 8<p, in, ailerons, the available control power

J

was almost 50 percent greater than what was considered satisfactory on the BR 941 -4 (tested at nearly the same speed and weight); yet the pilots rated the lateral control of the C-8A unsatisfactory (PR = 5). The turn-entry coordination of the C-8A was acceptable, the control friction was moderately high and the 0 !i: force gradient was low. The pilots com- mented that the sensitivity was lower than desired and expressed some dissatis- faction about the nonlinear relation of rad/sec control power and wheel deflection. The C-8A was quite easily disturbed in tur- iiillii .......

0 ::q bulence, and occasionally the pilot required full lateral control to recover 50 : : : : :_ ::= ::. 7- =7 :--_:- [:= :.r or compensate for gusts; hence, the pilot felt the need for additional con- trol power. Because of these interesting deg 30 !i: characteristics, additional comparisons were made with the BR 941; the time his- tories of rapid full control input for I0 2O the two aircraft are compared in fig- ure 27. It is seen that the response 0 I 2 t, sec to pilot control is quite similar and angular acceleration can be obtained Figure 27.- Comparison of C-8A and BR-941 rapidly for both aircraft. The C-8A has lateral response.

a higher angular acceleration capability, but this is compensatedby a higher

damping in roll so that there is a similar response. Several factors may con-

tribute to the poor lateral-control rating of the C-8A. These are: high-

aerodynamic damping, low-control sensitivity at small-control deflections, increased sensitivity at higher deflections, large control wheel deflections, moderately high control system friction, and poor centering characteristLcs.

The calculated relation between Slep response and angular acceleration is t_d, deq given in figure 28 for several param- O,sec Ioi PmlUS ,o ijiii!:,:i:!,::,:i: _i_ eters. The top part of figure 28

i

relates the parameter of bank angle

H

after 1 second, ¢i, to control power ¢0' for a step and ramp input. The parameter, ¢I, has been used in V/STOI.

specifications (refs. 19 and 21), but i:{ ::1 it is difficult to obtain represen- tative measurements and to correlate ::7:T ...... iii data because of the strong influence of control input shape and lags. The i!!!!!!! [ii! ii_ remainder of figure 28 relates the time to 30 ° bank angle (t30) and con- trol power. The parameter t30 was proposed as an indicator of control _,-.t ....

P' i 11 power in references 20 and 26 for conventional aircraft. Bank angles 50' sec 2.0 _ ......

of about 30 ° were the maximum nor- mally used during low-speed maneuver- ing of STOL aircraft. Data are _ 0.3 sec ramp I tL: _i_ !!ii ['.i'.!['._ i:_} insufficient for determining whether 0 2 4 6 8 i0 _o, rod/see:;' this parameter correlates; the values of t30 specified in the criteria Figure 28.- Bank angle and time to bank.

were computed for the angular accel- eration listed in the criteria, a lateral time constant of 1 sec, a transport lag of 0.i sec (time between force application and control-surface movement), and a ramp input 0.3 sec long. The criterion of time to 30 ° bank angle can be easily measured and evaluated, and it includes the damping and control system characteristics. The criteria include a desired level of angular acceleration, even though this is redundant in some cases and more difficult to measure, to assure sufficient ability to counter gusts quickly.

The calculations relating response and angular acceleration are based on the following equations: 1. For step input ..-7-= 57.3 t + z2 1 1 , deg/rad/sec 2 ¢ It (e -t/T )] ¢o

2. For ramp input

- 57.3 _t z2 a

deg/rad/sec 2

I t z _3_-t/_

_ e(ta/'_)- (t/_)] 1'

2 t a

These equations are for a single degree of freedom. The transport lag must

be included after tile response is computedby these equations. Tests with

STOLtransports showedthat a rapid lateral control input was approximated

by a O.1 sec transport lag and a ramp input 0.3 sec long; these time constants

were used for tile calculation of t30 and ¢I given in table IV.

Lateral-control sensitivity.- Sensitivity was not varied systematically nor independently of control power on the STOL aircraft tested. In no case was there adverse comment about too high sensitivity near zero-control deflec- tion; however, there were adverse comments about too low sensitivity and increased sensitivity with control deflection. Additional information on control sensitivity varied in a systematic manner is given in figure 29 for tile 367-80 tested at i]5 knots. Care must be exercised in using these data because of tile higher test speeds and limited turbulence encountered. Fig- ure 29(a) shows the importance of sensitivity; but also that a broad range can be utilized to obtain satisfactory handling. The lowest sensitivity for satisfactory handling at if5 knots was 0.04 rad/sec2/in, control deflection.

On ex_lination, table IV indicates that this level of sensitivity was unsatis- factory for all STOL aircraft except the UF-XS. This aircraft, a seaplane, was not required to maneuver exten- sively and was equipped with attitude stabilization. Based on the other @#I, deg STOL aircraft a sensitivity of at least O.l rad/secE/in, is suggested for satisfactory handling at STOL speeds (table IV).

Lp,deg The Breguet 941 had a nonlinear 0 2 4 6 8 I0 12 Equivalent _LD, in variation o£ angular acceleration with control deflection that gave higher (o) Senslhvity; 1.05 _< r R <_142 sec, little or no control sensitivity at small lateral turbulence, fhght and simulalor deflections than at large deflections O Simulotor no 'urrbulence (fig. 26(a)). This relation was satis- 6 [] Simulalor - with turbulence factory. The YC-134A and C-8A aircraft A In fhght-little turbulence had low sensitivity at small deflec- :!: i :..:: :..i tions and greatly increased values at midlateral control deflection rod/sec2 :!i !

(figs. 26(d) and (g)); these character- .2 istics were unsatisfactory because they caused overcontrolling.

0 2 4 6 8 I0 12 The STOL transports with wheel- Equivolent 8Lp, in type controls had maximum wheel deflec- (b) Control power; "r R = I.I tions that were too large. Reference Figure 29.- Lateral control 367-80 (ref. 25), 20 specified that the wheel deflections optimum force gradient; V = lib knots.

for conventional transports be limited

to a maximum of ±60° . For STOLmaneuvering the wheel motion must be compat-

ible with one-hand operation. To compare the characteristics of aircraft

having sticks and wheels, it was assumedthat linear motion at the rim of the

wheel was the pertinent factor. Thus a 60° wheel deflection corresponds to a

stick deflection of about 7 inches, for an average wheel radius of 7 inches.

The only STOLwith a stick deflection this large was the VZ-3RY; for this air-

craft, full lateral control could not be used because the stick contacted the

pilot's knee.

Lateral control forces.- No systematic study was made at STOL speeds to relate control-system friction, force gradient, aerodynamic stability and control, and pilot opinion; however, based on the information presented in figure 26 and table IIl, some general comments can be made. First, the forces must be sufficiently low that one-hand control can be maintained easily over the entire control range; second, the friction must be low enough to permit good centering of the control; third, there must be harmony between the axes to avoid inadvertent control application.

The control-force criteria are stated merely as a maximum force to achieve the control-power requirement because there is insufficient informa- tion to prescribe levels of breakout force, friction, free play, lags, gradi- ent, etc. The level of forces specified in the criteria are based on the tests of the STOL aircraft. Data from the 367-80 at 115 knots were included because it was tested with different force gradients. The fact that the cri- teria presented are insufficient is clearly demonstrated by a comparison of the force characteristics and pilot opinions of the 367-80 and C_8A aircraft (figs. 26(f) and (g)). The maximum forces for these two aircraft were similar and below the level specified for satisfactory operation; however, the C-8A control system was unsatisfactory because there were 6-1b of friction, 6-1b breakout, and a low gradient which caused poor centering and produced a spiral-type divergence. On the other hand, the 367-80 had a satisfactory system with the same maximum-force level, but it had 2 Ib of friction, an 7-1b breakout force, and a 1.3 ib/in, gradient.

Lateral-control cross coupling.- For STOL aircraft with low-directional stability, high sideslip angles develop during maneuvering and the pilot can- not make precise heading changes, cannot accurately control sideslip during touchdown in crosswind landing, and in some cases is concerned about stalling the vertical fin. In addition, when the aircraft is disturbed at low air- speeds, small bank angles develop large yaw rates, and it is difficult for the pilot to maintain the desired heading. These problems of turn-entry coordina- tion or cross coupling (or perhaps lack of cross coupling) will be illustrated first in figure 30 by the time history of a step-bank maneuver performed with the NC-130B at 70 knots (ref. 7). It can be seen that although the desired bank angle was obtained in 2-1/2 sec, 7 sec elapsed before the heading changed in the correct direction. During this time a large sideslip excursion occurred. The difficulty a pilot has in coordinating such a turn is shown in figure 31 by the different amounts and phasings of the rudder required to com- pensate for adverse yaw, yaw rate damping, and roll rate.

The degree of turn-entry coordination has been related to the ratio of peak sideslip excursion to the peak bank angle, AB/A_. The correlation of A6/Z_¢ with pilot's opinion of turn _oil,t, ,,,,, coordination is presented in figure 32 41-\ | for different aircraft and for a range of lateral-directional characteristics studied on the simulator. The A6/A¢ is measured during a rapid bank-angle change with the rudder fixed as illus- 0 I [ I l I trated in figure 30. This maneuver is -4 ] l ] % I T I similar to that performed for a rapid heading change or a recovery from an upset. Figure 32 shows that when the value of A6/&¢ was above 0.3, turn entry became a problem and the pilot gave an unsatisfactory rating (PR worse '- -20 than 3-i/2). Nhen AB/A¢ was above 12L- -24 I 0.6, the aircraft handling became unacceptable for normal operation.

I i I I I I For the NC-130B turn entry pre- °/_ , IX, i I , , , sented in figure 30, the AB/A¢ was _ - 04 0.8 and the pilot rating was 6-1/2 in 08 I I VFR and 7 to 8 in IFR. The parameter AB/A_ is generally not dependent on the magnitude of bank angle nor rapid- HI ity of control input. This parameter can be easily visualized and evaluated -32 I I I b b I i 0 2 4 6 8 IO 12 14 16 by the pilot when given a calibrated Time, sec sideslip indicator.

Figure 30.- Time history oF the response of the £ NC-i30B to a step bank maneuver; V = 70 knots (ref. 7).

f:.!., %; ,j} :f'_ ,.;_.$S: "t _ PR -IOO I I I I I

o

p, deglsec ]_I -5 I I I A_rcrQtl V, KFIOIS 5 0 BR 941 60 r, deq /sec 0 k i_-_- l -5 4 t I I I I I I Z_ C- 8A 70 .-- i r n v r Alero od e 0 _ se yow , NC 130B Basic 70 NC-130B Augmented 70 k 367-80 Bos,c 85

..,... 2

5 Yow rote damping 367 80 Augmented 85 L'_ VZ 3RY (VFR) 60 8 g .... L_. --. RoI , ....

" Simulator 60-80 ,o LX/ :L, ....

I I I I I _ 150 2 4 6 8 I0 I ] I I I / Tpme, sec 0 .2 .4 .6 .8 LO Figure 31.- Time history showing rudder require- ments for a coordinated (6 = O) turn maneuver; Figure 32.- Retation of turn entry coordination V = 70 knots, NC-130B (ref. 7).

and pilot opinion in IFR.

4O

Cross-coupling parameters such as Np and N_ have been varied both in

flight and on the simulator (refs. 7 and ll) to ascertain their effects on

turn-entry coordination and to compare the AB/A_ with pilot opinion. The

improvements in handling produced by positive N and N: augmentation on the

NC-130Band 367-80 aircraft are shown in figure _2. Th_se improvements cor-

relate with simulator results and show that AB/A_ is useful in assessing

turn-entry coordination.

The coupling of the lateral control with longitudinal motion has been a smaller problem than turn-entry coordination, Some aircraft that used spoilers for lateral control have had minor lift-loss problems. The midspan spoilers of the C-8A aircraft produced an incremental normal acceleration of about -0.15 g at the center of gravity with full lateral control deflec- tion; this acceleration was marginally acceptable. The BR-941 used outboard spoilers and less than -0.1 g was incurred with full lateral control, this acceleration was not troublesome and 3O was considered satisfactory. None of the straight-winged STOL aircraft expe- 20 rienced any significant pitching moment when lateral control was applied. On 4_, deg the other hand, the pitching accelera- tion developed by lateral control on the early configuration of the swept- I wing 367-80 flying at 85 knots was quite objectionable. The outboard .4 spoiler panels, a major contributor to the pitching moment, were subsequently .2 disconnected, and the pitching moment p, rod/sec was essentially eliminated. The result- I ing loss in laterai control power was of little consequence because there was more control than necessary. Quantita- I I I I I -.2 tive values of acceptable pitching motion are not available.

r, rod/se£ Apparent roll damping._ The roll damping of all STOL aircraft was satis- factory or at least acceptable (pilot I I I I I rating of 4 or better). Specific I0 levels of roll-time constant are pre- sented in a later section on stability and damping. Apparent roll damping is included in the aircraft response to I control criteria because the pilot has -5 difficulty isolating control-system lags when evaluating damping during I I I I I I -I0 lateral-control maneuvers.

0 2 4 6 |, sec An illustration of a time history of a step bank for a STOL with low roll Figure 33.- I2xamplo of step bank _vith low apparent damping; V = 60 knots, rudder fixed.

damping is presented in figure 33. To

maneuverquickly the aircraft is rapidly banked from level flight to 30 ° using

a large control deflection. In order to stop the roll rate at the desired bank angle the pilot applied opposite control of a magnitude as large as the input. More than two of these reversals were needed to stabilize near 30 ° bank. This control activity was considered unsatisfactory (PR = 4), see reference 10.

Directional control power.- The significance of low airspeed during crosswind approaches is illustrated in figure 34. At low airspeeds the crab or sideslip angle required to track the runway centerline is considerably greater than the pilot is normally used to. For some of the STOL aircraft evaluated during moderate crosswinds, it was easier to use the sideslip or wings-down method because the amount of bank angle required to balance the aircraft was small, and it was not necessary to decrab the airplane through a large heading angle just prior to touchdown. The maximum sideslip capability of the aircraft evaluated is included in the figure, and it can be seen that none of the aircraft could be considered for operation with crosswind compo- nents in excess of 25 knots. Since it is possible that these sidelip angles might be limited by proximity to stalling the vertical tail, it can be real- ized that the crosswind requirement for STOL aircraft is an important design consideration. It will be difficult to design STOL aircraft to operate safely in crosswinds that are over 40 percent of the approach speed. There- fore, the steady-state directional control criteria are given quantitatively only for the minimum level of safe operation.

60 Cross w4nd component, knots 60 50 _ I 0 50 _- Mo_ sideshp flown 0 BR 94J

K

\

£) YC i34A 40 X 40 \ V NC 130B ,, deg ,8, deg _ 567-80

so

2o _ __ 2o- _ _ _ _ 01 L L I I OL I I I I 50 5,3 20-- 2,) -- Crosswind component, __, deg ,.,5,deg knots _ _ C' -- ,o,2o, so ,o ---m--, -_- _- L i J J 01 J 1 J I OI 4O 60 8C, I O0 120 40 60 80 I O0 120 ',/, _noPs V, knots (o) Crab OpprOSCb, 5:0 (b) Wing dawn approach, _,, O Figure 34.- Effect of airspeed on cross-wind approaches.

To rapidly decrab the aircraft, .6-- r [ ! , Ir, -- ] | Aircrofl V, Knots or to quickly reduce unwanted side- slip angles that occur during maneu- [] UF -xs 55 t 0 BR 941 60 V NO _30B 7O !

vering, angular response to rudder _" NC K30B 85 i I 567-80 85 pedal deflection is also desired.

.4 -_ _s. 367- 80 1i5 i Figure 35 gives the available control cv 48 60 '_o, r°d/sec2 z_5 power measured for the different STOL aircraft and the pilot ratings based on the ability to maneuver the air- craft with all of the powerplants iI operating. It was concluded that the time to change heading 15 ° was a reasonable task. The times listed in the criteria were computed for a 3O 5O I00 300 0.3-sec ramp control input with angu- W, lb x 10 3 lar accelerations of 0.16 and Figure 35.- Directional response, Pilot rating 0.08 rad/sec 2 and a damping time next to symbol.

constant of 4 sec using the same equations as for the lateral response described earlier. For STOL aircraft without interconnected propulsion systems large asymmetric yawing moments occurred with an engine failure; in some cases, the asymmetric rolling moment of powered-lift aircraft was more critical (refs. 4, 5, and 12).

Directional control forces.- There were only a few comments on the directional force characteristics of the STOL aircraft. The UF-XS was the only STOL aircraft for which the pilot noted that the force gradient was too low, and this information was used to develop a minimum force level for the criteria.

Directional control cross coupling.- The most prevalent cross coupling from directional control input is the lateral response or dihedral effect.

It is generally agreed that dihedral effect in moderate amounts is a desirable feature. Figure 36 presents the steady-state sideslip characteristics of Control hm# C0nlro, I,m,t Con_s,! hq'l* :; .>., :-;,,. ;.>. ,. :; ;:;;: >;:. ,,,,;, R',gh_ I 0 s2.¢._'////x///,,.,,,/;,,. ,N. _,../,/..;:;_.,_'///./i//x//,;

_ ]

_ (%,/%.,.J ,%/% _Lp/_L_,_, ' .5 - "_ SL_Lp_, r_P/8'tP_°" 5r_ l'_i Left I.O 2_Z/._#_/._//_.z///f/4/_////F&-/.///_l/////A¢/_#f I©- ,,oe_ ol t_ I I i0 L_ J_ I I -20 -tO 0 10 20 -20 .-10 0 I0 20 20 -tO 0 I0 _'O ,_, deg B, aeq _,deg In} BR 941. V< : 62 knots (b) NC-_30B, Vc : 70 knots (c) 567-80, Vc - 9C' knolS Figure 36.- Steady-state sideslip characteristics.

several STOL aircraft. These airplanes required little lateral control in steady sideslip, and this lack of dihedral effect was not considered a defi- ciency. References iI and 14 have indicated that negative dihedral effect is highly undesirable and can produce a spiral instability. On the other hand, too much positive dihedral effect creates undesirable lateral-directional oscillations, and this will be discussed in a later section. For the purpose of this directional control section, however, the concern is that sufficient lateral control power be available at the maximum sideslip angle specified to insure adequate control of bank angle.

Longitudinal control Rower.- The available nose_up longitudinal A,rcCaf t V, knoIs 0 BR 941 60 control power and rating are pre- [] UF-XS 55 sented in figure 37 for the dif- 0 YC-134A 70 ferent STOL aircraft in the NC ]30B 70 20 '_" NC - _30B B5 approach, landing, and take-off [!_ 367-80 85 speed range for one center of 367 -80 I_5 gravity. For the CV-48 and BR 941 1.6 -- ,I Cv 48 55 a vZ 3 60 full nose-up control was used at 8o, rad/sec the lift-off speed for the best 1.2 take-off performance. Figure 22 showed that a i0 ° attitude change .8 was quickly made, and the pilot reported that the maneuver was .4 simple to perform (ref. I0). For the BR-941 half of the nose-up control was used to flare the 3 5 ! 0 90 50 I O0 300 aircraft from the steeper approach W, Ib x 103 angles (refs. 9 and lO); this attitude change was required to develop sufficient normal accel- Figure 37.- Pitching acceleration. Pilot rating next to symbol. eration by lift increases, and also to produce the proper fuse- lage attitude at touchdown. For the 367-80 the trim required in ground effect at 85 knots reduced the angular acceleration from the value shown to an inadequate value which was rated 4.

For this larger aircraft large negative normal accelerations occurred at rearward "passenger locations" when large control steps were made at altitude.

From these tests it would be inferred that in order to maintain passenger comfort, the angular acceleration would have to be restricted on the larger aircraft and greater emphasis would be placed on developing normal accelera- tion by other means, such as power or direct-lift control.

Little longitudinal control was needed during the approach because flight path was controlled primarily by modulating engine power, and moderate angle- of-attack excursions produced by atmospheric disturbances could be corrected by small longitudinal control inputs.

The longitudinal control system must also be capable of trimming the

aircraft throughout the flight envelope. Figure 38 gives the variation of

elevator angle with angle of attack for two STOLaircraft. Because of the

low static stability and high control effectiveness, little control is required to trim over T. E40 the angle of attack and power _/////.//////////////._ range in the approach and landing down _/_///.,/,'//////////////'_7/.,_ configuration. (The trim changes - T Power | , Idle that occur with power application 20-- 1--

///.85 r °x

will be discussed in the next 0 1 _ 1 I section.) All aircraft had suf- .30_ 8e, deg _ .25 ficient nose_down pitching response at the minimum speed to effect a satisfactory recovery.

y///////////,, v//////////_y////////////_ No criteria for this latter 4O requirement can be given other y/,,'//Z///'_ W//',/////'//////////,x///× than to state that the longitu- h I I T E. 6OoL ' , up I 0 IO 20 I0 O _O 20 dinal control must be capable of au, deg a, deg developing a nose_down pitching SHP (o) BR 941; 8f = 98 °, T: 12, _ = 500/eqgine [b) NC-13OB, 8,[ : 70 °, cg: .25_ velocity at the minimum speed under all power conditions.

Care must be taken when attitude Figure 38.- I!xamples of long_tudinal control stabilization systems are required for trim.

included in the control so that the pilot will be given proper information to recognize the amount of control remaining at high angles of attack when low stability is augmented. Some of the STOL aircraft were also tested over their allowable center-of-gravity range, and no significant trim problems occurred. In the case of the BR 941 at the forward center of gravity the rotation rate was reduced at nose wheel lift-off speed; however, the take-off performance was not noticeably affected.

To assure acceptable control near the ground to properly adjust touch- down attitude, to avoid porpoising, and to compensate for ground effect, it is proposed that the longitudinal control be sufficient to trim the aircraft to the landing attitude in ground effect at a speed corresponding to the approach reference criteria minus 5 knots.

The response and damping of all the STOL aircraft were rated 4-1/2 or better, and therefore it was not possible to specify criteria for a lower, acceptable level of control power.

Longitudinal control sensitivity, forces, linearity, and apparent damping.- The requirements for these characteristics are comparable to those for lateral control. Although it is noted that harmony between lateral and longitudinal control should exist, no related criteria are presented.

Flight-path control.- There are three general flight areas in which the throttle can be used for flight-path control: one, tracking of the flight path during the approach and preliminary portion of the landing; two, control

of sink rate at touchdown; and, three_ making gross changes to flight path

such as for wave off and turning flight. Criteria are presented separately

for each of these areas.

In the tests of reference I0, the engine response to modest throttle

changes corresponded to a lag of about 0.5 sec plus a first-order time con-

stant of about 0.7 sec, and there was little lag between normal acceleration

and power changes. It was noted in these tests that an incremental normal

acceleration of more than ±0.I g could be obtained by throttle application.

This response was acceptable for tracking the flight path during the approach

down to about 50 ft provided that little pitching acceleration was produced

by power. The pilot felt that larger engine lags and time constants would have reduced the ability to track the ILS glide slope. This response was too long to arrest the sink rate at touchdown, however. None of the STOL trans- ports were flared by increasing power because the engine response was too slow to develop the desired normal acceleration for flaring, and the aircraft also had to be rotated for proper ground attitude. "Yherefore the normal acceleration required for flaring was developed by rapidly increasing the aircraft attitude which increased the angle of attack. Reference 27 noted that a time constant of less than 0.5 sec and a thrust-weight ratio of 1.09 (approximately an incremental normal acceleration of 0.09 g) was needed for satisfactory control of touchdown for V/STOL vehicles. It is felt that these values are also desired for STOL operation if power is used to flare. The response for gross changes to the flight path are the least stringent in terms of engine response characteristics. In the tests of reference 9, 2 sec were required to achieve wave-off power. This delay was satisfactory pro- vided the pitching moment produced by power was sma11. Without a throttle- elevator interconnect, the pitchdown acceleration of this deflected- slipstream configuration negated the incremental normal acceleration even though the corresponding trim required was a small part of the available longitudinal control power (see ref. 9). References ii and 13 gave simulator results of studies that include the effects of cross coupling between pitching J_oment and power.

No data on desirable throttle characteristics were obtained for these STOL aircraft.

Aircraft Response to External Disturbances The purpose of this section is to specify levels of stability and damping that will limit the excursions of the aircraft when disturbed from trimmed conditions and that will limit the time and effort required by the pilot to correct these disturbances. Some of the aspects of stability and damping, such as cross-coupling and apparent damping, were included in the section entitled "Criteria for Aircraft Response to Control Input"; however, even when these effects are minimized, the response of the aircraft to external disturbances must be considered.

Lateral-directional stability and damping criteria.- The following table presents the proposed criteria for aircraft response to disturbances at STOL speeds.

Level for Parameter to Level for Item satisfactory be measured safe operation operation i. Directional Period of Less than Insuffic3ent oscillation stability 12 sec data 2. Directional Time to Less than Must be positive damping 8 sec half amplitude 3. Dihedral No criteria, effect insufficient information Time to double 4. Spiral Not less than Not less than stability 20 sec amplitude 5 sec 5. Lateral Roll time Less than Less than 4 sec constant 2 sec damping Directional-stability and damping substantiation. The directional- stability and damping characteristics of several STOL aircraft are given in figure 39. The left-hand figure shows that there is no correlation between pilot opinion and directional frequency. For these aircraft the behavior was dominated by the damping and cross coupling associated with low stability (low directional frequency). When adequate damping and satisfactory turn coordination were provided, as on the augmented 367_80 and NC-130B and in simulator tests, the lowest directional frequencies tested were acceptable.

These tests indicated, however, that lower directional frequencies might not be acceptable because the static directional stability would be too low.

The right-hand portion of figure 39 relates pilot opinion and the Dutch-roll damping parameter, _md, for the STOL aircraft with a directional frequency range of 0.5 to 1.2 rad/sec. For damping ratios less 0.3, _d is approximately inversely proportional to time to half or double the amplitude of the Dutch-roll oscillation. In general, the ratings improve as the damp- ing is increased; however, the turn coordination also influences the ratings to a considerable degree, as indicated by the less favorable ratings shown in the right-hand figure for A_/A_ of 0.6 compared to those for _B/A_ of 0.3. Another factor is the method of providing damping at the low frequen- cies. For example, when the directional damping of the NC-130B was augmented by a signal proportional to yaw rate, the damping of the directional oscilla- tion was improved; however, the pilot rating was not changed because a large sideslip angle was incurred in steady turns which the pilot considered unsatisfactory (ref. 7). When damping was augmented with sideslip rate damp- ing, a significant improvement in damping as well as turn-entry coordination occurred. The atmospheric conditions had a strong bearing on the test results; when the sideslip rate damping was based on wind information (i.e., differentiation of a sideslip vane), the aircraft was unsatisfactorily dis- turbed in turbulent conditions whereas when the damper was based on a flight- path sensor the results obtained were satisfactory. Satisfactory ratings for the 367-80 at 85 knots were obtained only when both sideslip rate damping and satisfactory turn entry characteristics were provided with an augmentation system.

A_rcroft V, Knots A/7/A_ 0 BR 941 60 0.1 0.4 [] UF-XS 55 .25 _, C- 8A 70 .17 .45 0 YC-134A 80 .14 _7 NC-130B 70 • 12 .8 _I' NC-130B 70 .4 • NC-130B 70 .4 .3 t'-. 367-80 85 -. 12 .75 • 567-80 85 very vary zJ CV-48 60 .3 __: ,' Simulator 60-80 vary very Flogs - IFR rating Solid - Augmented _-_-+-.+-_ =-777 : p i i PR , -.m ..... t.........

L/_ _:_"_' :f'_,'_ _ A/91A < 031---; ]p sec_S_i .2 .4 .6 .8 1.0 1.2 _d,rad/sec Figure 39.- Pilot opinion of directional frequency and damping for several STOL aircraft.

The effect of poor directional characteristics is more pronounced during IFR approaches than VFR because the pilot requires more precise control of heading. For the basic NC-130B the IFR task became impossible and was rated 7-8 as compared to 6-1/2 for VFR; in contrast, the Breguet 941 was rated 4 for both VFR and IFR.

It is concluded that STOL aircraft will be unsafe if the directional oscillation is undamped or divergent, and that this oscillation must be damped to I/2 amplitude in less than 8 seconds to be satisfactory (PR = 3-1/2). How- ever, this criterion is not sufficient by itself; the aircraft must also comply with other criteria such as those presented for satisfactory cross coupling before safe and/or satisfactory directional stability and damping characteristics can be assured.

Dihedral effect.- The dihedral effect on the different straight-wing STOL aircraft did not cause problems at STOL speeds, but on the swept-wing 367-80 it produced a divergent Dutch-roll oscillation which caused the low directional damping of the basic airplane.

Figure 40 shows the variation of Wd 8 -- Simulaled BR 94P % pilot rating with the parameter -L6, _d = .74 o for two values of Np and three 7-- values of directional stability, _d, Np = -.05 obtained with the 367-80 (ref. 14); o also included are results from the PR ,_- simulation of the Breguet 941 .24 (ref. ll) where Np was small. When Np is low or negative (as is gener- 3-- __1.05 .24 ally the case for unaugmented air- __ I I I craft), near zero L B is preferred 2 L q _8 -.4 0 .4 .8 1.2 to keep the sideslip angle small - L/_, I/sec 2 while the aircraft is being maneu- vered; when optimum Np is provided, the pilot is more tolerant of L_ Figure 40.- Effect of dihedral on pilot rating.

because the No coordinates the turn (refs. ii and _4). In turbulence at STOL airspeeds, reduced dihedral effect was preferred because it reduced the rolling disturbances produced by sideslip angles from the gusts. On the other hand, reduced dihedral effect, can produce spiral instability when roll due to yaw rate, L r, is present (ref. ll).

Spiral stability.- The effect of spiral Aircraft V, knots L B _w d stability on pilot opinion is shown in fig- 0 BR 941 60 -5 07 ure 41 where simulator results as well as some [] UF-XS 55 -03 25 [_. 367-80 85 -14 09 flight results are given for the 367-80 and -- Sire 80 -14 09 Breguet 941. The spiral stability is shown in [] Sire 60 0 to-3 I terms of the reciprocal of time to half ampli- tude (stable) or of reciprocal of time to double amplitude (unstable or divergent). For the tests with the 367-80, a slightly stable PR condition (TI/2 = 20 sec) was considered opti- mum; increased stability was objectionable because of the necessity of holding lateral control in a steady turn. Satisfactory han- dling in STOL approaches can be attained with L ........ spiral instability, provided the bank angle ,3 .2 ,I 0 .I ,2 .3 .4 does not double in less than 20 sec. If the I I I I bank angle doubles in less than 5 sec, these characteristics may be unsafe, particularly in Figure 41.- Variation of pilot rating IFR operation.

with spiral stability.

In addition to the previous requirement another factor, inability to trim, must be considered. This characteristic is difficult to separate from spiral stability. The spiral stability could be evaluated on the aircraft used for figure 41 because they all had lateral control systems with good mechanical characteristics. On the other hand, aircraft with poor wheel cen- tering, such as the C-8A, could not be trimmed laterally; consequently, the "aerodynamic" spiral mode was masked. When the C-8A was laterally disturbed and the control was returned to the position for trimmed wing-level flight, the bank angle doubled in I0 sec, and this spiral instability would have been acceptable. However, when the control was released rather than returned to

the correct position, a large rolling momentremained. This momentproduced

an average 20° increase in bank angle in 5 sec, and was unacceptable for IFR

operation. At the present time, no criterion has been developed that provides

for such inability to trim in the spiral mode; however, it is believed desir-

able to limit the bank angle or roll rate that would occur after the control

is pulsed and released.

Aircraft Lateral damping.- Figure 42 compares the :L}' BR 941 ratings and time constants for the different [_ LJF-XS STOL aircraft with information from reference 26.

_'_ C 8A 367-80 The roll time constants for the STOL aircraft 367-80 Augmenled tested ranged from 0.6 to 3.3 sec (table IV).

,_ CV-48 The CV-48 with the 3.3-sec time constant was [] Ref. 26 rated as having too low damping (PR = 4). The remaining aircraft had time constants less than 1.3 sec, and these were satisfactory. Addi- tional information on roll damping was presented PR in references 26 and 28. Reference 28 suggested that the roll time constant for transport air- craft should not exceed 2 to 3 sec for satisfac- tory rating; whereas, reference 26 suggested that 1.3 sec be the maximum. Based on the cur- I rent STOL information, the criterion of a maxi- 0 _ 2 3 mum roll time constant of 2 sec is suggested for rR , sec satisfactory handling of STOL aircraft at low }:i_[ll't" 2_2,- koll time constLtnt.

speeds.

Not only should there be a criterion for maximum roll time constant, but there should probably be a minimum value to prevent excessive disturbance by gusts. An unaugmented aircraft with a small roll time constant, TR, has high aerodynamic roll damping, -Lp. This aerodynamic damping is produced by high section lift curve slopes whlch in turn increase the sensitivity of the air- craft to gusts. Consequently, low damping is desired to avoid being disturbed in turbulent air. This presents a conflicting requirement because the pilot desires good damping to lateral control inputs. An example was the C-8A which had a low roll time constant; the pilot rated the damping good, as noted in figure 42, but stated that tile aircraft was quite disturbed by gusts. On the other hand, the augmented 367-80 with the same low time constants, but half provided artificially, was less disturbed by gusts and had good damping to roll control. Unfortunately, there is no easy method of evaluating an air- craft in a controlled gust environment in order to develop appropriate crite- ria. Consequently, considerable operational experience is required to evalu- ate gust sensitivity. The data in figure 42 were primarily from the pilot's evaluation of damping of aircraft motion to his lateral control input; there- fore the ratings are not necessarily a measure of the aircraft's sensitivity to gusts.

Longitudinal stability and damping.- Insufficient information is avail- able to formulate desired levels of longitudinal stability or a criteria to evaluate the stability; however, the level of static stability should not be so low that the resulting short period motion is aperiodically divergent.

5O

The static and dynamic longitudinal stability levels for the different

STOLaircraft have been quite low. Nevertheless, the longitudinal handling

was acceptable to satisfactory because the short period modewas usually

critically dampedand moderate angle of attack excursions could be permitted

without large changes in normal acceleration, airspeed, and flight path

(refs. 9-11). Figure 43 presents the boundaries of stability and damping

developed in the variable stability helicopter tests of reference 29.

Included in this figure are static stability and dampingvalues measuredfor

several STOLaircraft. The short period frequency could not be accurately

measured in flight because the frequency was low (_ < 1-i/2 rad/sec), the

damping was high (_ > i), and the control power was high. This figure shows that satisfactory handling could be obtained with near zero M_, provided adequate damping was present.

.&lrcrgfl V, knots 0 BR 941 60 [] UF-XS 55 V NC-150 70 I'x 567- 80 85 /1 CV 48 60 [] TND-4364 45-75 l :i i .: Jnaccept[[ t _' ..... ...... r ! £i:__:[[: 3 i;-:.]-_-i: ........ [-:: : .... .L...L.. ._ -L ., ....... i-- " 1.2 .8 .4 0 -.4 -.8 -I.2 -I.6 -2.0 Me ' I sec2 Figure 43.- Angle-of-attack stability and pitch rate damping. Pilot rating next to symbol.

Reference I0, tests of the BR 941, pointed out that pilot opinion improved significantly when the center of gravity was forward rather than aft, even though the dynamic motion was not greatly different. The corresponding increase in Ma, shown in figure 43, reduced the pilot effort to maintain the desired angle of attack in smooth air under VFR and IFR conditions.

Experience in rough air is insufficient to determine the effects of M_.

However, it would be expected that high levels of M_ are not desired because of the rough ride; for such an environment, attitude stability through augmentation would be preferable to angle-of-attack stability. Reference 29 showed that changes in positive speed stability, My, had only a minor effect on the pilot rating. It was noted in references ii and 13 that speed stabil- ity was related to the pitching moment produced by a thrust change; the benefits of speed stability were much less noticeable to the pilot than the corresponding adverse trim caused by a thrust change.

For most of the STOL aircraft, the phugoid motion was of low frequency (periods greater than 20 sec) with near neutral damping and caused no problem; in fact, it was usually difficult for the pilot to excite this motion. For one configuration the phugoid had a period of 12 sec with divergent damping (g about -0.15). This caused no problem in VFR flight, but the pilot anticipated problems in routine IFR operations.

Conclusions

STOLaircraft generally had low levels of stability which were satisfac-

tory provided the dampingwas sufficient and the mechanical and aerodynamic

control characteristics were good. It is necessary to have low friction,

force gradients, and control centering consistent with one-hand operation; in

addition, lags and adverse cross coupling should be minimized. Criteria for

control and stability are presented for two levels; one, which is the minimum

for satisfactory handling, and the other which is acceptable, but requires

considerable pilot workload. In addition, substantiating data are presented

that relate pilot opinion and pertinent stability and control characteristics.

The lateral control requirement was dictated by rapid correction to

disturbance by gusts; good turn-entry coordination and dampingwas necessary

for precise maneuvering. The directional control was primarily determined by

the necessity for trimming in crosswinds. None of the STOLaircraft could

be trimmed in crosswinds exceeding 40 percent of the approach speed. Longi-

tudinal control was dominated by either trim or maneuvering.

There are conflicting requirements for lateral-directional stability and

damping. Lowdirectional stability is desired to reduce the disturbance of

the aircraft by gusts; however, low stability increases problems of turn-entry

coordination and of maintaining heading. Similarly, high directional damping

is desired, but the aircraft becomessluggish to control. High lateral damp-

ing can cause the aircraft to be more disturbed by gusts. Slight spiral

instability could be tolerated, but a slight spiral stability was optimum.

Lowdihedral effect was desired provided it did not cause spiral instability.

The static and dynamic levels of longitudinal stability for the STOL

aircraft were quite low, but these levels were acceptable because they were

usually critically damped. Information was insufficient for presenting

criteria.

The conflicting requirements of low stability and damping to reduce air-

craft disturbance by gusts and of good stability and damping to maintain the

desired flight path will be best satisfied by augmenting stability and damping

about the flight path axes rather than about the wind axes.

CONCLUDING REMARKS

This report summarizespreviously reported NASA flight and simulator

data on STOLaircraft, vehicles that derive a large portion of their lift and

control from the propulsion system. Data are extracted and presented in a

form that should be useful for the designer and operator in evaluating new

designs and for regulatory agencies for ascertaining the airworthiness of com-

mercial STOLtransport aircraft. The main emphasis has been to provide

information for satisfactory performance, operational characteristics, and

handling qualities during approach and landing, because these characteristics

are required to provide safe and consistent operation during routine flying

in a wide variety of weather conditions. The data are primarily addressed to

STOLaircraft operating at 40 to 80 knots, with descent and ascent angles of

6 ° and greater, multiengines, and gross weights from 30,000 to I00,000 lb.

It is concluded that STOL aircraft utilizing power to develop lift can safely operate with smaller speed margins than conventional aircraft. The speeds chosen cannot be singularly related to the stall speed by some factor such as 1.3 times the power-off stall speed. A method is given whereby an operating envelope can be developed to estimate a safe operating speed consid- ering maneuvering margins, operating restrictions, and powerplant failure.

The margins and restrictions required are discussed, and illustrative examples are given.

At present, rational field length factors cannot be developed because flight data are insufficient for assessing consistency in STOL performance over a range of runway and atmospheric conditions. It is recommended that performance measurements be made with restraints imposed to simulate the operational environment and to expose adverse handling. With such a method, rational field lengths can be ascertained for each STOL aircraft. Field length factors will have to be developed for different types of STOL aircraft to account for their unique characteristics and operational techniques.

Handling qualities criteria for different parameters are presented for two levels: one that should provide satisfactory handling under a wide operat- ing environment including IFR; and the other, the lowest level of an individ- ual parameter that can be tolerated in some task, but would still provide a satisfactory rating for the overall landing task. It is concluded that with the generally low level of stability and damping present on STOL aircraft, the mechanical control characteristics assume a larger importance in overall han- dling than they do in conventional aircraft. The control friction, gradients, harmony, sensitivity, lags, etc., are as important as the basic stability and damping of the aircraft. In fact, in most cases, these are indistinguishable by the pilot and must be included in evaluating aircraft stability and control.

Insufficient systematic work has been done to define acceptable mechanical control characteristics for STOL craft; however, some preliminary guidelines are given. It is concluded that conventional stability and damping present conflicting requirements with handling in gusty environment. That is to say, that high levels of aerodynamic stability and damping at STOL speeds are not necessarily desired because they cause the aircraft to be more disturbed in gusty air. Consequently, augmentation with respect to the flight path will be more desirable than augmentation of conventional aerodynamic parameters.

It should be noted that the proposed methods, margins, and criteria presented are a first cut and will require further verification in a system- atic manner with different types of STOL aircraft. Like other flying quali- ties specifications, requirements, and standards, the recommended levels of margins and criteria will have to be reviewed and revised as more experience is gained.

Additional research must be performed to define the gust, wind shear, and crosswinds that are encountered in STOL operation. Statistical data are

needed to determine the effect of these environmental conditions on perfor-

mancemargins, field length factors, and obstacle clearance angle. The

effects of gusts and wind shear on handling qualities must be evaluated fur-

ther. A systematic study should be madeto relate control system character-

istics (friction, gradient, harmony, lags, etc.) to control power, control

sensitivity, stability and damping in IFR conditions with representative

turbulence levels. It is necessary to define the desired levels of longitu-

dinal stability and damping in relation to flight-path tasks when power is

used for control. Tests should be madeto determine how attitude stabiliza-

tion about the lateral and longitudinal axes affects the handling qualities

and the control power requirements. More flight experience is needed to

define methods of efficiently operating STOLaircraft under IFR in the

terminal area and to define the guidance and displays needed.

AmesResearch Center

National Aeronautics and SpaceAdministration

Moffett Field, Calif., 94035, June 19, 1969

REFERENCES 1 .

Stein, Kenneth J.: Avionics Play Major Role in Eastern STOL Evaluation.

Aviation Week and Space Technology, vol. 89, no. 18, Oct. 28, 1968, pp. 162-169.

.

Loebelson, Robert M., ed.: Airline Interest Grows. Vertical World, December 1967, p. 4.

° Anon.: Tentative Airworthiness Standards for Verticraft/Powered Lift Transport Category Aircraft. Flight Standards Service, Dept. of Transportation, Federal Aviation Agency, Washington, D. C., July 1968.

.

Innis, Robert C.; and Quigley, llervey C.: A Flight Examination of Operat- ing Problems of V/STOL Aircraft in STOL-Type Landing and Approach.

NASA TN D-862, 1961.

, Quigley, Hervey C.; and Innis, Robert C.: Handling Qualities and Operational Problems of a Large Four-Propeller STOL Transport Airplane.

NASA TN D-1647, 1963.

6.

Quigley, Hervey C.; and Lawson, Herbert F., Jr.: Simulator Study of the Lateral-Directional Handling Qualities of a Large Four-Propellered STOL Transport Airplane. NASA TN D-1773, 1963.

.

Quigley, Hervey C.; Innis, Robert C.; Vomaske, Richard F.; and Ratcliff, Jack W.: A Flight and Simulator Study of Directional Augmentation Criteria for a Four-Propellered STOL Airplane. NASA TN D-3909, 1967.

8, Turner, Howard L.; and Drinkwater, Fred J., IIl: Some Flight Character- istics of a Deflected Slipstream V/STOL Aircraft. NASA TN D-1891, 1963.

. Quigley, IIervey C.; Innis, Robert C.; and Holzhauser, Curt A.: A Flight Investigation of the Performance, Handling Qualities, and Operational Characteristics of a Deflected Slipstream STOL Transport Airplane Having Four Interconnected Propellers. NASA TN D-2231, 1964.

I0.

Innis, Robert C.; llolzhauser, Curt A.; and Gallant, Richard P.: Flight Tests Under IFR With an STOL Transport Aircraft. NASA TN D-4939, 1968.

ll. Holzhauser, Curt A.; Innis, Robert C.; and Vomaske, Richard F.: A Flight and Simulator Study of the llandling Qualities of a Deflected Slipstream STOL Seaplane llaving Four Propellers and Boundary-Layer Control. NASA TN D-2966, 1965.

12.

Feistel, Terrell W.; and Innis Robert C.: Results of a Brief Flight Investigation of a Coin-Type STOL Aircraft. NASA TN D-4141, 1967.

13. Vomaske, Richard F.; and Drinkwater, Fred J., Ill: A Simulator Study to Determine Pilot Opinion of the Trim Changes With Power for Deflected Slipstream STOL Airplanes. NASA TN D-3246, 1966.

Ames Research Center: Conference on V/STOL and STOL Aircraft.

14.

NASA SP-ll6, 1966.

1S. Hall, Albert W.; Grunwald, Kalman J.; and Deal, Perry L.: Flight Investigation of Performance Characteristics During Landing Approach of a Large Powered-Lift Jet Transport. NASA TN D-4261, 1967.

16.

Lovell, J. Calvin; and Lipson, Stanley: An Analysis of the Effect of Lift-Drag Ratio and Stalling Speed on Landing-Flare Characteristics.

NACA TN 1930, 1949.

17. Anon.: FAA Airworthiness Standards: Transport Category Airplanes.

FAR Part 25, U.S. Government Printing Office, Wash., D.C., 1965.

18. Anon.: FAA Certification and Operations: Air Carriers and Commercial Operators of Large Aircraft. FAR Part 121, U.S. Government Printing Office, Wash., D.C., 1965.

19. Anderson, Seth B.: An Examination of Handling Qualities Criteria for V/STOL Aircraft. NASA TN D-331, 1960.

20. Kroll, John, Jr.: Initial VTOL Flight Control Design Criteria Development - Discussion of Selected Jlandling Qualities Topics First Yearly Report. AFFDL-TR-67-151, Feb. 1968.

21. Anon.: Recommendations for V/STOL llandling Qualities (With an Addendum Containing Comments on the Recommendations). AGARD Rep. 408A, 1964.

22.

Cooper, George E.: Understanding and Interpreting Pilot Opinion.

Aero. Eng. Rev., vol. 16, no. 3, Mar. 1957, pp. 47-51, 56.

SS 23.

Cooper, George E,; and tlarper, Robert P.: The Use of Pilot Rating in

the Evaluation of Aircraft Handling Qualities. NASA TN D-5153, 1969.

24.

Anderson, Seth B.; Quigley, Hervey C.; and Innis, Robert C.: Stability

and Control Considerations for STOLAircraft. AIAA Paper 65-715, 1965.

25.

Condit, Philip M.; Kimbrel, Laddie G.; and Root, Robert G.: Inflight

and Ground-BasedSimulation of Handling Qualities of Very Large

Airplanes in Landing Approach, NASACR-635, 1966.

26.

Ashkenas, I. L.: A Consolidation of Lateral-Directional Handling

Qualities. AIAA Paper 65-314, 1965.

27.

Kelly, James R.; Garren, John F., Jr.; and Deal, Perry L.: Flight

Investigation of V/STOLHeight - Control Requirements for Hovering

and Low-SpeedFlight Under Visual Conditions. NASA TN D-3977, 1967.

28.

Bisgood, P. L.: A Review of Recent Handling Qualities Research, and its

Application to the Handling Problems of Large Aircraft. R.A.E. Rep.

Aero.-2688, 1964.

29.

DiCarlo, Daniei J.; Kelly, JamesR.; and Sommer,Robert W.: Flight

Investigation to Determine the Effect of Longitudinal Characteristics

on Low-SpeedInstrument Operation. NASA TN D-4364, 1968.

,_ 0 0 C; 0 _ ,-, E,,; I 0 0 _ oo Z O4 O4 0 0 0 C "_ 0 0 0 0 Lq 0 ',O I'_ Lq 0 0 0 l-q '4:) U Lq 0 Lq 0 _ _ _ _ q-_ [j Lq C'-.I 04 04 _ _ ,--I 0 C; 0 0 _ C.j 00 _q 0 0 00 I'--. e4 _ ,H 0 .< F_ d'J 0 aO 0 0 0 0 u"_ o4 , ,._ *1--__ I _ _ ,"_" _ -- _ o _ _ l,q Lq I_- iO'_ Ch _, _ 0 U.q P.-, U.-1 0 0 0 0 >< 0 0 0 0 _ "_ I 0 0 Lq 0 Lt_ 0 I'_ _-_ _ _ _ O0 I ;..!q ,--1 O0 0 0 0 0 _ .'_ 0 ..-_ "_ "a _ _ o_ o L,Q Z LD r", r'-" M.J o L.J, © r", .,< '0 p.- 1, i r_j ,.,., I i i i i 5g O O c'q c,-IO O i ;> _D r--.

I ,,D d .I X . O O G5 _, I 0 c,3 U U O {D U U Z _ _D O O .< _4 0 -n C,O .._ .,-_ ._ .S o c_ 0 .,-_ 0 t."3 F_ 0 c8 t_('M _-_ _ ._ z b_ ¢xl 0 U u .,--_ o/_ 0.0 i i 0_ 0 ,.c: +-_ _D c8 u Lt_ '.D {3'5 r..-I _ '_0 • 00"C_ O _ _--_ _'_ 0 0 _ _ +._ 0 0 0 0 t_ _ ct_ _::_._ 4_ 0 _ ".D rU _ ',.D _ t'_ {'-1 2 0 0 _._ _ r_ t¢3 m U "_ I U < o rj ,_3 b'3 U % X c8, 0 _0 Z _ i < m c_ Z "J I O 4J '2 I-..-t r...)

r,1 O < O O £ O O O 0 0 .,-I _ '_ '_'_ m 0 c_ _ • ,...t 0 _ ._ I::_._ 0 q_ • q-_ "_0 "0 0 0 _ ,.c_ .,-4 0 Ln ¢_ +i u_ i o u_ us +, i i r_ i ,j Ca _I o tO © i i o LD I o i o_ _ +' _ oo z q4 to _0 _ o ca ___ +I i 0 i _ +i _J O.._Z z o ,_ , _ (J o _'i ÷i o _ ._ o _D i u_ _r • 00 b_ • uJ E- _.a + _ I (J c,l ,._ _,.D LD 0 'm Z _D uJ I u I *a _J _ US _J L_ o_ i < W4 +I _o 6O © o i p..

,,D +1 I I'-- ! 0 _D ©,--q I +1 <

T

.,-I c_ _J u_ r..)

I mm © + 0 0 +1 O ! (.J rm < I-- Lq 0 O0 I_ Cq ('xl U +1 ,.q .t.-_ >-, ,-_ c_ ,,- (_ Z

o

O0 © U _ 00 0 _ Lr_ _I" 0 _.D -- 0 +1 ,--I O_ I.q '--q +1 "0 O _ c_ c_ .,-I 0 _ _ 0 ,--I O0 I'_ O'_ C',l ',_ +1 ,--t +1 r_ .,--1 • r...q I I _ .,-4 I .r-t © 0 0 ._ © ,--1 < L_ I q 0 ._ _- _ _ "_ I._ I._ •-_ _._ _ 0 r...)

+1 +I ,.0 0 "_ _ _= Z I_ _ I._ O0 O) (.J oo I._ 0 i h,,-i © © u-/ +1 LM > (1) u_ u_ .r-I _ ,._ o 0 ,--_ _ ,_.

o C_ > I_,0 %0 _,-_ I,_ _ o •_ ._1 _ I_ I_ 1_ .1_._ '_ _1 ._ 0 © 0 _ .t--_ © 0 _'_ .,_ _ ._ 'J _

.5

< o °r-t 0 _) Xq-_ N_-4 _ _ _ _, I._ 0 _ ;:;_ "_ 0 m t'_ ',D t_ r--4 r'--4 O OO I I O ¢',,I ',D O4 t'_ I O O i OO I t_ t_ t_ O O t"- O'J t.-.t o o O ,--4 ,--4

A

I r'--4 Lt_ t_ Z O O o4 0 I" c;o o I C) r._ .< (_; ("4 E-_ >_ c_ L_ _ 0 _0 eq Ch _0 0 0 '--_ _0 0 _ I g O4 t'-- _ L_

'7

c/3 Z 1 t-t5 < (,0 .r-4 0 oo m L._ p-. I .-1 e-- O ,.c: C_ X Z _ t'M ° ° O c,,l 0 d _ 0 _,_ .-_ I I I 0 0 % L_ I -.1 L_ t_ I ,< 0 0 % o.1 0.0 .,-I • ,--i .,-_ I_1 4_ 4J o4 q_ % m _" _ ° U 0 • ,_ _ 0 0 (D II E"a ul b.O qb o_ .it c_ ._ < "a:) .r..I < zoMod io_uo 3 -!suo_ o e r.i ,_ o ,-q oo . o '.o _ O o o o ',0 t._ oo o o _o _1 o o0 d >.

_ d t'M o o > U o .el L_ Z r-- _ o d o c, o (:D m o ,'.-4 O t¢3 O .-.1 o t_ t--q .-'1 6 o0 c_ •,D t'-- o _E_ o •H ..'-4 (30 o _D O (D _2 0 c_ O rj ",O O c_ ,-.4 0 0 c_ O t

o, ,o

r_ o_ c-i rj _._ .r_ b_ •_ o I o o ,--t ,---i O C_ c_ 4-_ ::S ,_ _D o

loo

0 0 t,_ rj 4J t_ 0 ._ O I c_ t_D < o0 rj 6 _ ...1 oo _D uD 0 • _t ._1 r--- m_ o 'o,d o {-- Z t_ rj o CD o o ° > i % > ,el O0 uJ ',o I 0 % ,-1 _d i _._ _ c_ _4 O ,--4 I_ _: ,-'_ "_ 0 < U m [-, .oo, O O m ,-..i u3 O0 m t--- o > C,8 O m

:7 % m

c_ c_ _D q-_ m _-_ 0 (I) 0.) 0 0 o •r_ _ 0 % 4J c.) _) _ "_ _ _rj o c_ ._ _ .-'4 m _ +J _'_ 0 :_ _ r_ 0 •,._ _ ;_.

m _ O _od I0_u0D Z < xo C_l t_ . ° c_ o t"q c'q o i c'q 00 0 , E •_ _ ,-_ ,_ c_ _D ,...-I • ° o o _ c'q L_ i O0 0 q._ o c'q Lr_ o .,._ 0 0 L,_ _ _ 0 I E U Z _ 0 o0 _ 0 U I m % c'q o % 0 b_ 0 o C_ L_J t'¢3 % oo t__ ,,0 "," 0 Lf% o _ o ,._ ,---t o c_ ,-4 d _ o m 4-_ i 0 ,< C13 ._ d ._ c8 o r..Ll "--" • 0 r¢ X t_ i uh 0 o o o C_ Oi 0 ,-.I t'¢3 _V C'q Z 0o o g 0 ._ ,.o '-.0 5 & I O_ = 0 -,_ .c: V ,--_ _ _ 0 _:_ _ ._ _._ ,_ .-_ • _ ,-_ 0 c_ cd _:_ -- _ _ u o _ _ _ _ _'_ .,-I X'_ 0 0 _ _ _ < _o_od io._:_UOD X_!A!_!suos D.a Z ¢J U4 d 2_ ,--.1 I > d .<