An investigation of the influence of fuselage flow field on rotor

l'OUTH EROPEAN 201'0ilCRAFT AND POWERED LIFT AIRCRAFT FORUM

Paper No

8

AN INVESTIGt.TION OF THE INFLUENCE OF FUSELAGE FL01_{tl FELD ON R01'0R LOADS, AND THE SfE:CTS OF}

'JEHICLE CONFIGURATION

P G Wilby, C Young and J Grant Ro:~al Aircraft F~stablishment

England

September 13 - 15, 1978 3TctESA, ITALY

.~ssociazione Itnliana di J\eronautica ed Astronauticn

ARsociazione Industrie Aerospaziali

J:.N Hrv':G0TIC:··.'I'ION Or' '.:'HE I:rFLl!ENCE OF Fu:_:~Lf,GE FLO\~' FIELD CN ?-IO'X?. LOADS, i,ND TEE EF?SCTS OF

1.JE.H~.·:::LE CONFIGURATION

Introduction

P G ihlby C Young and

J

Grant

Royal !':.ircraft Establishment

The idea that the flow disturbances due to the fuselBge of helicopter

can affe(:t the loads and mot ion experienced by the rotor is not a new one. An enrly study of the influence of the fuselage of a Wessex helicopter on

the flapping motion of the blades1 concluded that the effect wos not

signifi::c1nt, Cur present interest in the topic started a few years ago when

predicted blade loadings for an S53 helicopter were being comnared with fliv,ht

experiments2, It \oJas found that an appreciable difference beb1een tn.eory !ind mF:<1surement existed over the inner pRrt of the blade for a narrow band of azimuth angle centred on 1;30°, This difference could be accounted for if a modific~tion to blade incidence wns made in the prediction method that was compatihle wit!! the upwash provided by a simple representation of the fuselage. At About the same time, in some flight experiments at the RAE on 8 0essex

heJ icopter, 3 rO!.lp;hnes.c:; ·~125 applied to the blade leading-edge to lower the stallin~ incide~ce. This res11lted in blade stall being provok~d in the region of 60% rotor rRdiu.s at 180° nzimuth in forward flir:;ht, indic-"ting th-?.t blade incidence can be quite high in this reg£on of tlle rotor disc. More recer.t R.:\E :-liE;ht experim<:nts on a Puma h8licopter have indicated a p~rturbatior in blade

incide~tce .ss t:.--,~ bl~de passes through 180° azir::uth, ·..,rhich mc:Jy be due to

fu3ela;;e flow e:'fect;?• ~he in-:luence Of the f'Jselage f'.Fl:-1 attracted ~he '3ttention ~f other researchers? who conclude th9t it can cause a large incre8Ge in blade incidence around

180°

azimuth, and that this mu:st be taken in':o account when calcul~ tin~ b:sde loading •

. .\s the fuselage of the modern helicopter tends to extenC furt1.~r nnd i":Jc"':"ite:- forwards, rel·3tive to the rotor r.ub, any disturbance it causes i.s

moved further outboard along the blade, to a. rep;ion of hif)her dynamic pressure, and its effect is therefore likely to be accentuated. The current trend towardo more co~pact helicopter designs, involving a lowering of the rotor and a

decrE:>ase of U:e clearance between the fuselage and rotor, will further increase the influence of the fuselage on rotor loads. A. more thorough study of

fusolege flow effects is

cle~rly

needed and some of the results of this study

are presented in this paper.

The que~:;tions to be answered are, wh;:.,t is the magnitude of the change

in

bl~de

loading due to the fuselage flow, what are the contributing factors

and do the resulting loads lead to a significant increase in rotor vibratory

loading.

2

Fuselage Flow Field Calculations

The first step in the investigation was to develop a suitable method

for predicting the flow field, due to a fuselage, in the plar.e of the rotor.

An existing panel method, uAed for wing-body calculetions, was adapted for

this purpose and modified to give velocity components norml'll to a disc at any

specified position parallel to the body axis, One problem 'Nith such a

calculation is that a large number of fuselage ordinates have to be generated,

For simplicity, it was decided that the cross section of a fuselage could be

represented in the way shown in Figure 1., The sides, top and bottom have flat surf0.ces and the corners are elliptic, 'Nith double symmetry assumed. Only four parameters (D, \.J, d and w) are needed to define any shape from a

circle (D = 'tl; w = d = o) or an ellipse (D

t.

'd; w = d

=

o) to a rectanp.;le

(D

=

d, W

=

w).

In bet••een these extremes lies a whole range of shapes with

flat sides and rounded corners.

By varying these parameters along the length

of the fuselage, typical helicopter shapes can be described, provided a further parameter, the height, h, of the mean line above the datum, is prescribed.

A

simple eeneral expression can be derived to give the

co-ordinates of the cross section for any value of the angle 8

(defined

in Figure 1).

Then, by specifying the five parameters shown in Figure 1, a

complete set of f~;sel8.ge co-ordinates can be calculated for a set of val11es

of

@

and x.

Cnlculations have been made for fuselage shapes representing the Puma and Lynx helicopters, being examples of modern designs. The vertical component

of velocity in the plane of each rotor is

~iven

in Figure 2 for

180°

azimuth.

h.s might be expected, the maximum value is found near the front of the cabin roof, following the steeply inclined line of the windshield., The way in which upwAsh decays on either side of

180°

azimuth is shown in Figure 3, where •He see that at ~{ rotor radius the upwash has fallen to half its maximum value over

25°

to

30°

azimuth.

He.ving obtnined the upwash over the rotor disc it was then introduced into a program for calculating the blade loading and rotor performance. An iterative calcul.::<tion is nsed in 'Nhich the wake is represented by a series of vortex rinr;rs. 'The 0rocess converp;es onto a different solution depending on -vhether or r.ot the fuselage upwash is included. Tte

1lpW"ash is in+:roduced -?!E> a modific;:;ttion to the rotor dO'...rnwash field prior to the

cc:.lculEltion of blade lo~ding. ?L.-ure 4 gives the calcul8.ted blade incidence

di.strib'-ltion at

1il0°

a~imuth

for a Lynx helicopter at

140

knots and shows the

effect of the fuselAge flow. ~he lower part of Figure 4 gives the change in

incidence due to the fuselage upwash, together with the value that might be

inferred on the basis of upwash velocity and blade velocity alone. For this

pa~ticular case, the maximum chan~e in blade incidence is

8.5°

at about 3~~

rotor radius, giving a total incidence there of

13°,

It should be noted here

that the blade incidence still lies within the boundary (see Figure

4~

set by

the stall angle for the blade section (RAE 9615) in steady conditions ,

Although a fuselage incidence of zero has been taken in these

calculations, the flow field program will run for any specified fuselage

incidence (but of course, for potential flow).

The effect of a

4°

nose-down

attitude on the simple evaluation of blade incidence is shown in Figure 5, i'lhere an additional ., .5° in blade incidence is found at

30%

rot7r radius.

As the Lynx fuselage attitude at

140

knots is only

1°

nose-down , the effect

on blade incidence is negligible and any further results presented here are

for zero fuselage incidence, A further factor that should be considered is

that, in the calcul~tion, the rotor blade is assumed to lie in a plane that

is parallel to the fuselage datum,

Strictly, the affect of coning angle and

disc tilt should be taken into account, To assess the importance of the

simplification, fuselage upwash is presented in Figure

6

at two different

heights above the fuselage.

There is little change in upwash with height

near the blade tip, but an appreciable change over the inboard region,

Thus,

provided the plane in which upwash is calculated represents the height of the

inboard part of the rotor blade above the fuselage, then errors due to the

simplified approach will be small,

The theoretical results presented so far have been obtained with what

we refer to as the performance program. This assumes that the blades themselves

are rigid except in torsion. For the investigation of blade loads we need a

full representation of the blade dynamic characteristics and this requires the coupled modes program in which flap, lag and torsion modes are introduced.

Both these programs are based on ones developed by Westland Helicopters and are

referred to in the remainrler of this paper.

3

The Effect of Fuselage Upwash on Rotor Loads for the Lynx Helicouter

Although the fuselage upwash increases incidence considerably over the

inner part of the blade at 180° azimuth for the Lynx helicopter (Figure 4),

the dynamic pre"sure over this region of the blade is low and the effect on

blade lift distributions is fairly small, as is seen in Figure 7,

In Figure

8

(a)

the azimuthal variation of the blade lift coefficient is given and the effect

of the fuselage upwash is seen to cause only a small perturbation on the rotor

alone value.

The magnitude of the amplitude of the variation is unaltered,

Figure

8

(b) gives the azimuthal variation of blade torque, due to

aerodynamic drag, and again the fuselage flow causes only a small perturbation,

The decrease in blade torque in the region of 180° azimuth arises from the

reduction in induced drag due to the upwash of the fuselage.

rr·he coupled modes program was used to calculate the response of the

rotor to these small changes in aerodynamic loading at

140

knots, with

8

modes

( 4 flap••ise, 3 lagwise and

1

torsion) being represented.

The predicted

changes in blade bending moments at 13,5% rotor radius are shown in FiF,Ure 9,

The change in flapwise moment has a strong

3

cycles per revolution content, and

the change in las•ise moment has a strong 4 cycles per revolution content.

These observations combined with the interference diagram for the Lynx rotor

(Figure 10) suggest that the 2nd flap and 2nd lag modes are being excited,

which is to be expected even though their natural freauencies are reasonably

well clear of resonance.

It could be argued that these modes are likely to be

excited because their shapes are similar to the form of the spanwise

1istribution of the forcing lORdS,

It should be pointed out here that the peak to peak variation of

flap-wise bending moment is not altered by the presence of the fuselage, but the

peak to peak lagwise moment is increased by about 30%,

It can then be noted

that the agreement between predicted and measured amplitude of the flapwise

moment was found to be very good by Westland Helicopters?, whereas the

amplitude of the lagwise moment was found to be considerably larger than

predicted.

The difference in influence on the two moment variations follows

from the fact that peaks in the lagwise moment perturbation coincide with

peaks in the lagwise moment variation for the rotor alone. This does not

happen with the flapwise moments,

Hub forces and moments are important from the vibration point of view,

and Figure 11 gives the predicted azimuthal variation of pitching moment,

axial force and in-plane force for rotor alone and with fuselage effects.

The amplitude of the fore and aft in-plane force is doubled by the presence

of the fuselage whereas the pitching moment and axial force amplitudes are

slightly

decr~ased.

In Ref

8

the agreement between predicted and measured

pitching-moments and axial force was quite good whilst the measured in-plane

force had double the predicted amplitude at

140

knots. Fuselage effects again

appear to offer an explanation for some of the discrepancy.

The results of these calculations can be summarised by saying thA.t the upwash due to the fusel8t:;e of a Lynx helicopter appears to increase the lagwise

bending moment by about 30%, and doubles the amplitude of the fore and aft

-1

in-plane force. However, this is not to suggest that the rotor cleBrance is . too small, as Fi~ure 6 shows that even

if

it

was increased by 30% the fuselage

effects would be reduced by only 20%.

4

Wind Tunnel Model Tests and Comparison with Theory

In order to check the theoretical methods that were being used to

calculate the effect of the fuselagecj a model rotor test programme ••as devised

for the 24 ft wind tunnel at the RAE • A fuselage model was pleced below an

existing model rotor1° and measurements of bending-moments made on the flap and

lag flexures of one particular blade (the 3-bladed rotor is non-articulated

in flap and lag). Tests were carried out for two fuselage shapes (Figure 12),

and a ra_nge of rotor height, rotor speed, advance ratio and thrust.

The experimental results are being compared with theoretical predictions

given by the coupled modes rotor loads program.

Five blade modes have been

included in the calculations, three flap and two lag, which are all the modes

with a frequency of less than 150hz (see Figure 13).

Some of the comparisons between theory and experiment are presented in

Figures 14 to 16.

Each figure shows the measured and calculated change in the

flatwise bending moment on the flap flexure at 9.4% rotor radius, due to the addition of the fuselage. ~he configuration for all these cases is as shown in

Figure 12 (a) with the lowest of the three rotor positions.

The bending mome~t increment for a rotor speed of 400 rev/min,

c8llective pitch of

8.7°

Rnd advance ratio of 0.3, is virtually undamped at

6

\l

whicil is coincident with the second flA-p mode frequency (see Figure 13),

The C-".llculated oscillMtion is more damped than the measured one on both the

r8tre,~ting and advancin~~ sides of the disc, but is c1gain essentially of

6

0

frer;;uency. ;,:

:=;co

rev/~in (Figure 15) we have basically a 50, oscil:;_ation for both experimental and theoreticsl results but with different deerees of damping. ,\gain, we note that 50 is very close to the 2nd flap mode frequency. For this case, the Measurements show a virtually undamped oscillation on the retreAting side of the disc which is rapidly damped in the advancing sector, while theory predists a pro;;~essive damping all round the disc. With a further increase

of rotor speed to 600 rev/min, the m"asurements show a mildly damped

oscillation with theory predicting slightly greater damping.

At this rotor speed

the oscill.:ttion is atter'lpting to achieve a frequency of about 4.50. Once

again, it is the 2nd flap mode that is being excited even though the 3rd flap

mode !requency is closer to resonance.

In an attempt to reduce the differences beh.reen measured and predicted results, the structural damping of the modes in the calculation was changed hut this l;.ad no sir,nificant effect. It is possible that the diff<Jrences in d<'lmping are

of

aerody~amic

origin and may be related to differences in fuselage flow field

at the rear of the rotor disc, where inviscid flow calculations are not expected to be accurate.

For the experimental results, it was found that the effect of the

fuselage on the magnitude of the peak to peak flatwise bending moments depended

upon the rotor speed.

As we have already seen, the frequency of the bending

moment perturbation varies with rotor speed, thus the phasing between

perturbation peaks and rotor-alone peaks will vary ·with rotor speed. At 400 rev/min we have a coincidence of ;·Jeak values (Figure 17), givine; '1n increase in peak to peak variation, b;--tt at 600 rev/min this does not occur.

'ile thus have the ~esult that although the perturbntions Dt 600 rev/min nre l:::~.rr;er,

in proportion to the undisturbed amplitude, than 3.t 400 rev/min, they rtctually lead to a reduction in peak to peak load as apposed to an increa~e .~t 400 rev/min.

:.t

the time of writing this paper, the experimental lag bending moments had not been analysed. However, on the basis of the flotwise moments comFarisons, one can conclude that the theory predicts the effects of the fuselage

reasonably well but with an underestimate of the amplitude of the oscillatory perturbation.

5

Features that can Increase Blade Incidence at 130° Azimuth

The calcul8tions so far presented for the Lyf'l..x helicopter indicate that the blade is well below stalling incidence over the forward part of the rotor disc at 140 knots at sea l~vel. The possibility of large fluctuations in control loads caused by blade stall does not appear to be a problem with this helicopter. Flip;ht tests at the

R.>~E

on A Puma helicopter4, which provides

a similar fuselage upwash field to thRt of the Lynx (see Figure 2), have also indicated an absence of blade stall at the front of the disc. In these tests, pressure was measured at t~e leading and trailing-edges of a blade at 17

sp~nwise stations over the outer rzlf of the blade. The leading-edge pressures give an indication of blade incidence, and show a disturbance centred on

180°

azimuth ir, Figure

18.

There was no sign of trailing-edge pressure

divergence, ·which is used as an indic3tion of flow separEltion, even ;~t a speed of 140 knots at 2500 m altitude~ Howev~r, this r.ioe.s not r1.:le out the possiblity of blade stDll at

1B0°

azirr.uth for oti1er helicopter configurations Rnd f} ight conditions. ':iake distortion, increased forward speed, increased blade b.rist and different f11selage shapes can all provide increases in 'clade incidence and

rna!:e stall more likely. The importance of these factors will now be co~sidered, using the performance program.

5.1 \lake distortion

Tf,e ~.,.!'Ike model u~ed in the ce.lcul2.tion consists of an a":"r?.y of vortex rinr:;s, in planes normal to the rotor shaft but displaced longitudinally and vertically according to forward flight and downwash velocities. The vertical displacement 6z, between each vortex ring is given by

2

-J;;;f = JfV

R

b()R

where

v

is the mean downwash velocity and b the number of blades. In the real c.<tse, with the influence of the fuselage, the distribution of downwash velocity over the rotor disc is v~ry uneven and the vortex generated by the blade tip at 1~0° azimuth is likely to stay very close to the rotor disc as it is transported r~arwarrts. To evaluate the possible effects of such a feature of the vortex wake, the first vortex ring in the wake model has been distorted by taking

2.:v

where k ca~ be sp8cified as desired. Advance ratio p is included in the distortion ter:n Cec:Juse the fuselage upwash, and hence the vortex displacement,

is directly proportional ::o forward speed. The distortion term is chosen to ) , . give a variation 'l'lith &zimuth that is similar to the form of the variation of

fuselage upwash, as se•~:1 in Figure 19. Using this expression, the vortex distortion is eff8ctively zero outside the range 120°<:..+ <240o.

1_{ilith this moJified form of the vortex ring wake model, the aerodynamic}

loading on a Lynx rotor at 140 knots has been calculated, and the blade

incidence distributions at 130° azimuth are shown in Figure 20 for k = o and 2.5. 't/ith the resulting close proximity of the vortex from the previous blade, the predi~ted blade incidence at mid span is increased by about 5°, giving a value of incidence slightly i~ excess of the static stall value. It is clearly important to have an accurate representation of the effect of the fuselage flow on the vortex wake

if

a reliable pr~diction of blade incidence is to be

attained.

5.2 Elade twist and for"'ard speed effects

An increase of blade twist naturally tends to increase blade incidence over ~::e irwer p;:n·t of ~\",e iisc. f'igure 21 shows the predicted changes in bJ.E-Jde incld~nce :, t 1 S0° azimut~, if ::.he twist of the Lynx rotor were increased

f'ro::. :~0 to 12° n.nd 16°. Doubling the twist would increase incidence by 2.5° over -l:.!'.e inner [-Art of t!"le blade at a forward spe~d of 140 knots.

Also ,:,howr1 in F'i~ure .21 is ~he increcse in blade incidence resulting from an incresse of' forward sDGed :rom 140 to 160 knots. This also amounts to

?.)0 O','S>: t\---:-:: L'!ncr >::.3rt of tl1e blade at 180° azimuth. Although the blade

inci'"!·?ncP. ~-or :: .... '2 i.Jr'C< '..:o1.~ld t.her: :.;li:-z:htly -::xceed the static stall incidence for t>.e 'ol.qj_e section over a small extent of blade, no serious stall effects would

0<:> Anticipated.

5.3

Fuselage sha~e

?!·ie fusel3ge shape and its !)Osition relative to the rotor disc have a direct influence on up·.vash in the plane of the rotor and hence on blade incidence. To give some indi('ation of the sensitivity of upwash to fuselage configuration, the predicted upwash at 1S0° azimuth is given in Figure 22 for several variatlons of fusela~e shape. The plane in which upwash is

calculated is shown in Figure 22. Summarizing the results, the steepness of

the cabin windshield line hAs little effect on upwash, but the form of

f.::~irings on top of the fuselage has a considerable effect. For fuselage shape 5 in particular there is a consider.'3ble reduction in upwash over the inner blade

if

the rotor is ra!sed from position •a• to position

'b'

as shown in

figure 23.

The effect of verying the forward extension of the

fusel,~ge

nose

relative to the hub

ca~ b~

assessed simply by imposing a transverse displacement

on the upwash distribution.

A

further

possibl~

change in fuselage shape is a change in width, and

Fi~re 24 ~rives the upw;1sh distribution for a fuselage of half the width but with the same sideways profile as shape 1. A considerable reduction in

5.4 A combination of features that will produce blade stall ~t

180°

azimuth On the basis of the results obtElined so far, a con:'iguration wrt;:, selected

that is likely to produce blade stall at 180° azimuth,

Fuselage shape

5 was

chosen v1ith the rotoy- in close proximity, as shown in Figure 22. A rotor of the same dimension

as

the

Lynx

rotor

but

with

16°

twist

(in

place of this normal

3°)

was taken, and the tip vortex from the previous blade 'lias

distorted in the way outlined in Section 5,1, taking k

=

2.5. The blade

incidence at

\80°

azimuth for a speed of

140

knots as given by the rotor performance program is plotted in Figure

25.

Incidence is seen to be '"'ell in excess of the static stall incidence over the inner part of the blade. ·rhe performance program includes a representation of dynamic stall as developed by ':Jcstland Helicopters11, and the predicted azimuthal variation of root

pitching moment is shown in ?igure

26,

compared with the v~lue for rotor alone. In the top part of Figure

26

.~;~_re the corresponding variations of root pitching moment for the standard Lynx rotor. It is seen that the blade does in fact stall at

180°

azimuth for the selected configuration, p~oducing a considerable

oscillatory increment to the pitching moment,

This will of course be reflected

in an oscillation superimposed on the control loads. Stall will also produce consider0ble variations in blade drag and an extra aerodynamic forcing

for additio~l oscillatory leg bending moments and in-plane hub forces. In such a case as this, ~here would be considerable benefits in raising the height of the roto~·.

6

Conclusio~s

.-.. 1t!:.01..:.g=1 the .st'.ldy of fuselage effects on rotor loads is not comDlete :;,pd i.s -:ertainl;;: not ex!la'...lstive, various points have emerged that are worth emrhasj.sinss.

a. Ca~cu:l.."tions indicate that fuselage upv1ash can provide a

pert1.:.rbation. to ::~.erodyn~mic forces wftich lead to a significant hlade and hub response in a form depending upon the stiffnesses of the rotor blades, even when blade stall is not precipitated. For the Lynx helicopter

the effect shows up as an increase in blade lagwise bending moment and in-plane hub force. In such a case, where blade stall is not present, r~ising the rotor height would not have a large effect on the loads gen~!"ctted.

b.

Results from wind tunnel tests on a model rotor have shown that the

method for

p~edicting

the effect of fuselage flow on rotor loads is

reasonaoly accurate, with perhaps a slight underestimate of the true

magnitude.

c,

The theoretical study of the factors that tend to increase blade

incidence at

180°

azimuth shows that it is possible with some

helicopter configurations to produce stall over the inner part of the

blade in tiie forward sector of the disc. Stall can be sufficiently

severe to cause large oscillations in blade root torsional load and

a much increased variation of blade drag,

The latter is likely to

amplify further the blade bending moments,

d.

The calculations show the importance of developing an accurate

representation of the way in which fuselage upwash distorts the

vortex wake when modelling the wake for the calculation of rotor

e. A general conclusion is that the influence of fuselage upwash

is important and must be taken into account when calculating blade

and hub loads.

References

1

M A P "illmer. Sffect of flow curvature due to the

fuselage on theflapping motion of a helicopter rotor,

RAE Tech Note Naval 61 (1963).

2

J Scheiman.

A tabulation of helicopter rotor blade

differential pressures, stresses, and motions as

measured in flight.

NASA TM X-952 (1964).

3

M

J

Riley,

A flight investigation of the spanwise lift

reauirements of a helicopter rotor blade by measurement

of the control loads arising from locally applied roughness,

ARC R&M 3812.

4

P Brotherhood, M

J

Riley.

Flight experiments on aerodynamic

features affecting helicopter blade design.

Vertica,

'lol 2 pp 27-42 (1978).

5

A

J

Landgrebe, R C Moffitt, D R Clark. Aerodynamic

technology for advanced rotorcraft, Journal of the

American Helicopter Society, Vol 22 No 3 July 1977,

6

N Gregory, P G Wilby.

of aerodynamic data.

~PL

9615 and NACA 0012; a comparison

ARC C:P No 1261 (1973).

7

K T Mckenzie, D A S Howell,

The prediction of loading actions

on high speed semi-rigid rotor helicopters, AGARD CP No 122 (1973).

8

V A B Rogers.

The design of the WG 13.

The Aeronautical

Journal, January 1974.

9

R

J

Marshall, 'ilind tunnel tests on the influence of

rotor-to-fuselage proximity on helicopter blade loads.

R~E

Technical Report to be published.

10

A Anscombe, A P Cox, R J Marshall,

'!lind tunnel testing of model

rotors at RAE Farnborough.

Proceedings of the 2nd European

Rotorcraft and Powered Lift Forum, Buckeburg, 1976.

11

T S Reddoes.

A synthesis of unsteady aerodynamic effects

including stall hysteresis. Proceedings of the 1st European

Rotorcraft and Powered Lift Forum, Southampton, 1975,

/

._L -~ ~-"'-

=---w --3 -~, '"" 0 >i>

t

C.\8 ' c :) . 5 .),

'

Ra-ji-.~s R L ·1 nx ~

'

'

/r?.-ma

Fig 1 Parameters to be specified for the calculation

of fuselage co-ordinates

---0.1 s ;.,., '). ~ 0 0

•

> < ~ :) 0 s

'

Q 0 080 L y n~ Pur~·;a : ~c· \ - - \ 160" ' . 1 ~)'

/''\

''"'

/ I ' '

' '

_'

'

_'

J.5

'

_{' '}

' '

_'

\

'

0 B \

'

\ \

'

'

'

I I I I I. 0

Fig 3 Predicted azimuthal variation of fuselage upwash in plane of rotor at 40% radius for Lynx and Puma helicopters

Fig 2 .d ~

.

"

..

.

"

0 iii

.

0 ;,

•

:0""

•

=

.

0

.

~~ 0

•

0 £ ~ u

Calculated upwash in plane of rotor at 180° azimuth for Lynx and Puma fuselages

,,

I 0 c Tp 0 8 5 fi "

'

0 T.;l :), c / / -W1:h luse;a~e

-:u

.. _t· r'-;__,Q

Fig 4 Predicted effect of fuselage upwash on blade incidence at 180° azimuth for Lynx

20

•

~ 10 :0 ~

sf----r-•

~ u ---,---.---, ' i

'

;;

Fig 5 Predicted effect of fuselage attitude on change of blade incidence due to fuselage upwash at 180° azimuth

I 0 Slide ilf: per un1t lengl!:!

To~al rotor lilt

- - W1th !ustiag~ ---Rotor alone 0 I

--0.~--~----~--~--~~--~ Ttp 08 05 o~ 0.2 Hub

'

;; G 2'J ;'!::.:' 0 II u 0

•

> ~

.

"

•

~ ~ 0 10 005

"

L.ow rotor I / I I

'

H1gh

'---

__.. ro \or 1----j-- . ./_{1 L;.._. ·} -/ ~ 0 5 r 0 ' 0 1 0 R Ti!_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ HvD

Fig

6

Predicted variation of fuselage upwash with height above the fuselage at 180° azimuth

L1 t I

2 0 _{, - - - c - - - ! __:-w~-~~~Js.el~·g·;-~-;-i-;·~--1}

1_{---~olor aton<' 1}

0 I

0

'BLode i1ft .as a muU pie o"t

j mea" vJ.lue fo: ro:or alone

90 180

"

170 ~50 2 or----~---·---.---~---1 - - W 1 t h fuH!age effect : ---Rotor alone

Fig 7 Predicted effect of fuselage upwash on blade

'"q"'

loading at 180° azimuth for Lynx rotor at 1 o

f----~L..--i---'\----+--140 knots

0 I

0

BIOide aerodynam1c torc;ue ;as a multiple of mean v;~.lue tor rotor alone

180

<I!

270

Fig 8 Predicted effect of fuselage upwash on azimuthal variation of blade lift and torque for Lynx helicopter at 140 knots

"

E 0 E ~ c 0 c 0 a 25 20 II 10

'

I I I I I Fi ap

-/

I I I I I I

'

I I I

'

I

'

I

"

• - I ~ c

.

G I I _I v -\0

'

' I I I

_{,_}

'

'

Fig 9 Predicted change in flapwise and lagwise bending moments for Lynx rotor due to fuselage flow, taken as a percentage of peak to peak variation for rotor alone

10

"

•

_E 0 E I ~ 0 i. a_ 0 D

,

I _o .:'S

.

0

.

_{: :)O}

.

a

_,

I 0 91

'

I I ~ 0 0 ;; c 0 n

.

c I 0 \

•

_;; a

_,

-I

'

'

::) ?0

'

_'

) 0

,,

I I ! I I ! I ' / / I

'

Wdh luse1~ge Rotor ;~to·'"' I I / / I eo / / 90

'

_'

\ IQ 20 30 'G SO 60 ~.) 80 90 ! / / / / ~

' '

20 30 40 ')0 60 .. 1 80 90 ~

Fig 11 Predicted hub moment and forces with and without fuselage flow effects, as a multiple of mean value for rotor alone for Lynx

N I c

.

,

~ ~ ;; ; ;; z sc

"

]0 2 0 10 so l 1 ~" L2 w>!h l'a;"'SmlSS•Ofl .:lf n a-., ,: ~ 3 ,')

!:)0 tSO 200 250 lCO. E:J

I

Fig 10 Interference diagram for Lynx rotor

-~----

--::::.

b Snort fuselage

ton 9n 90 '0

+

-F'

1

-I

0 E z

c

€

0 E ~

"

~ c

•

n > ~

•

'

"

•

~ c '

•

u

8 r-E'P"'meo>

---Theory 11=03.~₀-:87' 0

_,

-8

---~

I _ j _ _ _ _ _ _ _

l_~--Fig 14 Predicted and measured effect of fuselage on

300

'"

500 600 100 model blade flatwise bending moment at

400 rev/min Rotor s ::> e e d • ? v' m.o

Fig 13 Interference diagram for RAE model rotor

8

'

z F

€

0 E ~ c

"

_•

'

~

.

0 ~

•

•

I

'

_,

I

•

~ c

•

•

u -8 Fig 15 --- r----~---~---· -· - - E~per,menl ~-- Theory '

'

I I I ' I

I

1 I 270 ' ¢ I I

I

'

:

I

I ~~r--- -!---+---~

--+

__ , ____

j_~

__

____________j__l - · E z F > E 0 E --~---~ - -E~per1ment ---Theory

I

--~

0 I I I I \

_,

\ , j

. \ __ / r

r--- --

1

_j ___

j

-8 L _ _ _ _ __L__

Predicted and measured effect of fuselage on Fig 16 model blade flatwise bending moment at

Predicted and measured effect of fuselage on model blede flatwise bending moment at 600 rev/min

E z C tOr---,--- ----r---~---,

•

E 0 E ~ 0

i

o f---f-'---:'-:-:-li"---4L::...."----:':::'-C:~:---~ n

•

E z

c

\0

•

E 0 E ~ c 0 ;; 0 n L.OOrevlm1n - - - W1th fuselag~ - - - - R o t o r atone , - - - - ---,- ---1

' '

_'

•

~

'

'

•

' : 600 1ev/mon'

• Fig 18 Measured variation of leading-edge pressures on a Puma blade in forward flight

u.. -\0

Fig 17 Measured variation of blade flatwise bending moment for model rotor at two rotor speeds for iJ. = 0.3 I • . 08 0 6 0 4 G 2 90

Fig 19 Tip vortex distortion function compared with azimuthal variation of fuselage upwash at 40% rotor radius for Lynx

d

•

c

•

~ u c

•

~

•

"'

~ : 2 5' IS

i%i

(\~~ ~\\ I 10

we--

<>~ - <, \

''

'

,

'

'

~---~-- -- -+---i

i

o_L---~~--~----~--~----~ Tip 0 8 0 6 0 ' 0 2 Hub I O•stance along bl<lde R

Fig 20 Predicted effect of vortex wake distortion on blade incidence at 180° azimuth for

"I

I

I I

T--.

, ~w·st ''5'1 '' 2'1 :t.J ~r"l oL---~----~--~~--~~---_{r,D ca ·J5 o .. 02 ·<u~} J,s!ance cl''!~g :.>:aae R

Fig 21 Predicted effect of increased forward speed and blade twist on blade incidence for Lynx rotor at180° azimuth

"T---.

--~

---... '

'

"

---~---~- --~i _ _ _ _

)201-;·:;

01\r---;"-~

; v I \. ~b " , t \ ' ] 'J

-4

\

---Fig 23

;

»

1 Ca! Cl s 0 5 0 ' 0 1 Hu:

:_

--_-_-_-_-_-_

-_-_-_-_-_-_-~----=-=---.-~-~

~I

Predicted variation of fuselage upwash with height above fuselage at 180° azimuth for two fuselage shapes

0 2 3 ,_ 0 > :> () 20

"

' 0

'

~ "(} :o

•

~

•

•

' ~ 0 OS --,----T-~--T·---,-·-~

,,

0 8

y:::,

-·-

!'-:2_'.'...~\-

---.

~~2

0 6 c 0 1 Hob

Fig 22 Predicted upwash in plane of rotor for various fuselage shapes at 180° azimuth

_, > ;? . 0

•

> ~ '

'

~

'

.

~ ) 20 !5

..

; c cs '

'

~ '8

c

' · - - " r - - - - --- - - · ·

'

'0f,ae fus~tace

'

.

'

_'

_., ---~- ~

---"-+---

---'

'i<l •• ~~-~ lust?iage '

'

_'

'

' ~ G 0

. Wtde !o.JS<:.>ia;e 1 prol,te I: .--- · Narrow !use:age ·' 1 6-::·

~~~o·

I ' ,

:

'\..

'

I I \ Pianv,ew , ~ J R , o s

-Fig 24 Effect of fuselage width on upwash velocity in plane of rotor at 50% radius

25 10

·,

> _{, 5} >

:::::.:-+ -

I u 0 0

•

_"

u

•

jj 0 L__-:"": _ _ _ _ ---:_;__ _ ___,J-:--_...J :,p ·')8 06 0~ ~2 hut ' ;;: ~otor ;)I an~

C_,,_._~

___

_...J Fig 25 Predicted blade incidence at 180° azimuth

for rotor of Lynx dimensions with 16° twist with fuselage configuration shown at 140 knots Az,muth ._, ~r---9TO~---~~s~o---~~~~OL_ ______ ~J~so 1

"

•

_{c -0 5}

g

g

-l 5 . < 2 0

-I

l

-~·

a Lynx fuse-lage and rotor J.l 140 knots

0

•

E 00 E-o s

•

v :-10 - - · ~ g -15

"'

-2 0 90 ! . . L Allmulh ~ ·---W1th ;---Rotor

i

I JGO

--,

fuselage (k ·25) d. I one

b Fust>lage 5 and 16' tw•st rolor at 1~0 knots

Fig 26a&b Calculated blade root pitching moment as a multiple of mean value for rotor alone, showing effect of blade stall at 180° azimuth for fuselage 5 and high twist blade