R. & M. No.

3485

~,~:~,,-';~ :, !;'~" ~ ',

'~'

~I ~ ' h -~',,,"

MINISTRY OF TECHNOLOGY
AERONAUTICAL RESEARCH COUNCIL
REPORTS AND MEMORANDA

Flight Measurements of the Elevator and Aileron

Hinge-Moment Derivatives of the Fairey Delta 2
Aircraft up to a Mach N u m b e r of 1"6 and C o m pafisons with Wind-tunnel Results
By R. Rose, M.Sc., O. P. Nicholas, B.Sc.(Eng.) and
Glynis Vorley

LONDON: HER MAJESTY'S STATIONERY OFFICE

1967
PRICE

15S.

6d.

NET

Flight Measurements of the Elevator and Aileron

Hinge-Moment Derivatives of the Fairey Delta 2
Aircraft up to a Mach Number of 1"6 and Comparisons with Wind-tunnel Results
By R. Rose, M.Sc., O. P. Nicholas, B.Sc.(Eng.) and

Glynis Vorley

Reports and Memoranda No. 3485*

July, 1965
Summary.

Aileron and elevator hinge-moment derivatives have been extracted from flight tests on the Fairey
Delta 2 in which the dynamic aircraft response, to control pulses was recorded. As this method is not
widely used, and has proved very successful, it is described in detail. Corrections had to be made for aeroelastic distortion of the control surfaces, for inertia loads acting on the controls and for the dynamic
response of the instrumentation.
Where possible the results are compared with wind-tunnel data. There is good agreement for the
hinge moment due to control derivatives, b2, and the elevator, bl, whereas the results for aileron, bl, show
differences from the tunnel values at supersonic speeds.
bo could not be measured because of strain gauge drift; however consistent results were obtained for
the aileron hinge moment induced by the elevator deflection and vice versa. At transonic speeds these
terms are of similar magnitude to the conventional hinge moment derivatives b 1 and b 2, and must therefore
be considered by designers.

LIST OF CONTENTS
1.

Introduction

2.

Description of Aircraft and Instrumentation

2.1.

Aircraft

2.2.

Instrumentation

3.

Flight Tests

4.

Method of Analysis

5.

4.1.

Elevator pulses

4.2.

Aileron pulses

4.3.

Steady turns

Results and Discussion

*Replaces R.A.E. Tech. Report No. 65143 --A.R.C. 27 376.

LIST OF CONTENTS--continued
6.

Conclusions

Appendix A The hinge moments acting on elevator and aileron controls

Appendix B Evaluation of b 2 and cross-control derivatives from control pulses
Appendix C Determination of b 1 from stick fixed longitudinal short-period oscillations
Appendix D Evaluation of elevator and aileron hinge moment derivatives due to incidence, bl,
from flight at different values of normal acceleration
Symbols
References
Table 1

Leading aircraft dimensions

Table 2

Control surface details

Illustrations
Detachable Abstract Cards

1. Introduction.
There is considerable interest in predicting values of the hinge moments for controls. At present there
are no reliable theories 1 for estimating hinge moments at transonic and supersonic speeds, and predictions, for these cases, are usually based on wind-tunnel results. To give confidence in these predictions, it
is important that some checks are made between tunnel and flight measurements. This report presents
the results of flight measurements made at R.A.E. Bedford, of trailing edge control hinge moments on
the Fairey Delta 2, which is a research aircraft with a control configuration of current interest and a
performance well into the supersonic speed range. The flight results are compared with tunnels results z,5
made at transonic and supersonic speeds.
A dynamic flight test technique has been used, which allows the elevator and aileron hinge-moment
derivatives, with the exception ofbo, the hinge moment at zero mcldence and control angle, to be extracted.
,

2. Description of Aircraft and Instrumentation.

2.1. Aircraft.
The Fairey Delta 2 is a 60 deg delta wing tailless research aircraft capable of speeds up to at least
M -- 1.7. Fig. 1 shows a general arrangement and Fig. 2 a photograph of the aircraft. Table 1 gives the
principal dimensions of the aircraft and Table 2 details of the control surfaces.
The aircraft has full-span trailing-edge controls, which are split into separate elevators and ailerons;
the elevators are aerodynamically unbalanced, the ailerons are rigged up 3 deg and have a small horn
balance. The control surfaces are operated by irreversible hydraulic jacks. The elevator jacks are positioned in the rear fuselage and operate the elevators via torque tubes. The aileron jacks are situated beneath the outer wing, forward of the mid portion of the aileron, and operate the ailerons directly.
Control-feel forces are supplied to the pilot by simple springs, the datum of each spring being varied,
within a limited range, by an electric trim motor. To cater for the large variations of aircraft response
and control effectiveness over the flight envelope, the gear ratios between the elevators and ailerons and
the corresponding stick movement, can be varied in flight by the pilot.

2.2. Instrumentation.
The following parameters, relevant to the tests, were recorded on Hussenot A.22 photographic trace
recorders :
Speed
Altitude
Elevator angle (port and starboard)
Aileron angle (port and starboard)
Elevator hinge moment (port and starboard)
Aileron hinge moment (port and starboard)
Strain-gauge bridge voltage
Incidence
Normal acceleration
Rate of roll
Sideslip angle
The hinge moments were measured with four-active-arm strain-gauge bridges energised by a stabilised
voltage from Venner accumulators. In the case of the elevators the strain gauges were attached to the inner
surface of the torque tubes; for the ailerons the gatiges were attached to one end of the jack body. A typical
strain increment in the tests was 0.02 per cent. The strain gauges were carefully matched and during ground
tests were found to have negligible drift over a temperature range - 4 0 deg C to + 4 0 deg C. The bridges
have been calibrated by applying known loads to the control surfaces. Control-surface angles were measured by potentiometers attached to the controls.
Incidence was measured by one of four wind vanes mounted on the nose boom of the aircraft behind
the pitot-static head. The vanes were arranged in a cruciform configuration so as to preserve the symmetrical arrangement required to reduce position errors in the transonic and supersonic speed range.
3. Flight Tests.
The main flight tests consisted of recording the control hinge moments and the response of the aircraft
during and after either an elevator or aileron pulse. Due to the lack of control of thrust with reheat on
and the limited fuel available, it was not possible in this aircraft at transonic and supersonic Mach numbers
to stabilise speed. This caused some difficulty in trimming the aircraft, particularly at transonic speeds
where significant trim changes occurred rapidly. Nevertheless, pilots found it possible, after some practice,
to trim the aircraft satisfactorily during these non-stabilised conditions. With the aircraft in trim, the pilot
tapped the control column in either a lateral or rearwards direction and allowed it to return to its trim
position under the action of the feel spring, resulting in a control pulse of about one quarter second
duration*. This was the shortest pulse possible, and gave a good compromise between the requirements
that the aircraft should not have responded significantly by the time the peak of the pulse had been reached,
and that unsteady aerodynamic effects should be negligible (Appendix A). In some cases a combination
of elevator and aileron movement occurred due to the difficulty of moving the control column precisely
in the desired direction. For the elevator tests, both the initial response during the pulse and the ensuing
longitudinal short-period oscillation of the aircraft were used in the analysis. However, for the aileron
pulses, only the initial response has been analysed, as the hinge moments induced by the dutch roll
oscillation were too small to measure accurately. At 40 000 ft, control pulses were recorded in level flight
at Mach numbers from 0-86 to 1-6, further tests were made in level flight at 10 000 ft, at Mach numbers from
0.6 to 1.2.
Some supplementary tests were made, in which records were taken during level flight and turns at
constant normal acceleration. These tests covered a range of Mach numbers from 0.85 to 1-07 at 40 000 ft
and from 0.55 to 0-97 at 10 000 ft.
*For a fully irreversible system no further movement should occur. However, the hydraulic jacks did
allow some small residual movement of the control surface.

4. Method of Analysis.
The incidence vane readings have been corrected for the effects of boom interference evaluated by windtunnel calibrations 4, and, in addition at subsonic speeds, for estimated wing and body upwash. Calculation
showed that the effect of pitching velocity on the vane readings was negligible, and this was therefore
ignored.
The control surfaces are subject to aeroelastic distortion, especially at high speed. Flight pressureplotting tests made earlier have shown that the aerodynamic loading is reasonably uniform, and making
this assumption, the mean angular distortion of the control is directly related to the jack load. In ground
tests s the distortion of the controls was measured under uniform loading conditions. This distortion was
measured relative to the jack output point since the control surface transmitters were attached at this
position. The effective stiffnesses of the elevator and aileron are respectively two thousand lb ft/degree
and about forty thousand lb ft/degree. Two factors contribute to this large difference in stiffness. Firstly,
the jack load is applied near the aileron mid-span, whereas it is applied at the root-end of the elevator.
Secondly, in the case of the elevator, flexibility of the linkage between the jack output and elevator root
accounts for about two-thirds of the measured distortion. In the present flight tests, it has been calculated
that significant elevator distortion occurred and this has been allowed for, but that the aileron distortions
were negligible.
Appendix A derives the relations between the measured control jack loads and the aerodynamic and
inertia loadings on the controls. The jack loads were measured by strain gauges, and although these were
carefully matched some drift occurred due to differential-temperature effects. Thus it has not been possible
to determine absolute values of jack loads very accurately, but as the rate of drift was sufficiently slow in
relation to the duration of most of the manoeuvres executed during the tests, incremental values of the jack
moments could be measured with confidence.
The dynamic response of the recording galvanometers and ratiometers has been allowed for in the
analysis. Simple laboratory checks showed that amplitude ratio effects were just significant in some cases.
A number of assumptions have been made in the analysis of the flight results and these are listed below.
(a) All hinge-moment derivatives are linear within the range covered during each individual test, but
they are not necessarily linear over the full range of control angles and incidences covered in the flight
tests.
(b) The shape of a control pulse may be represented by the first half of a sine wave.
(c) The change in aircraft incidence, up to the instant where an elevator pulse has its maximum, is
small.
(d) The changes in sideslip and rate of roll, up to the instant where an aileron pulse has its maximum, are
small.
(e) Movement of the controls is small, but possibly significant, during the aircraft oscillations following
an elevator pulse.
(f) The distorted elevator can be represented by an undistorted control at an appropriate mean angle.
(g) In applying the distortion corrections the aerodynamic and inertia loadings are assumed uniform
over the control surface.
(h) Calibrations of the control jack load and the surface distortions may be extrapolated linearly to
the considerably higher flight loading conditions.

4.1. Elevator Pulses.

Fig. 3a illustrates an idealised elevator pulse test, showing time histories of the elevator and aileron
jack moments, and incidence, during and following the pulse. A corresponding actual flight record is
shown for comparison, Fig. 3b. The essential differences between the actual and idealised records are that,
in practice, a small, but negligible, change of incidence occurs before the peak of the pulse, and also the
elevator does not returri precisely to its trimmed value at the cessation of the pulse. However, these effects
are quite small and thus it is possible to use simple methods of analysis. This analysis is conveniently split
into two parts, treating first the response during the pulse and then the ensuing aircraft oscillation.

From the increments in the control jack loads and the elevator angle between the trimmed conditions
and values at the peak of the elevator pulse, according to equations B.7 and B.8 of Appendix B, the hinge
moment derivatives b2~ and OCm,/Otl may be derived, b2E is the elevator hinge moment due to elevator
angle derivative, and OCHA/Orlis the aileron hinge moment due to elevator angle derivative. This latter
derivative is not normally considered, but it was found in these tests to be of a very considerable magnitude.
The inertia term, in equations B.7 and B.8, amounts typically to 5 per cent of the total measured hinge
moment.
The elevator pulse excites the longitudinal short period oscillation of the aircraft. Appendix C, equations
C.4 and C.5, shows how measurements made during the oscillation allow bl~ and b ~ to be determined.
bl~ is the elevator hinge moment due to incidence derivative, and b~., is the aileron hinge moment due
to incidence derivative. In this case the inertia term amounts typically to 15 per cent of the measured total
hinge moment. The analysis uses values of the recorded quantities only at the peaks of the oscillation.
In practice it was found that reliable results could be obtained at supersonic speeds; however, at subsonic
speeds, due to increased residual control motion following the pulse, and increased damping of the aircraft oscillation, the method could only be applied in some cases. In this regime an alternative technique
was required to obtain measurements of b ~ and bt~, and this is discussed in Section 4.3.
4.2. Aileron pulses.
Figs. 4a and b show idealised and actual flight records of an aileron pulse. The initial response of the
aircraft is a rapid acceleration in roll with practically no change in sideslip and rate of roll. The aileron and
elevator hinge moments closely follow the aileron angle input.
Equations B.10 and B.11, of Appendix B, show how b2~ and OC~JO~ can be obtained from measurements of incremental control loads and the aileron angle between the trimmed conditions and the peak of
the pulse, b2~ is the aileron hinge moment due to aileron-angle derivative, and OCn~JO~is the elevator hinge
moment due to aileron-angle derivative. The latter derivative represents the effect on the elevator hinge
moment of aileron angle deflection, which was found to be considerable in the present tests. The inertia
terms in these cases amount typically to15 per cent of the total hinge moment.
4.3. Steady Turns.
At subsonic speeds the damping of the longitudinal short period oscillation of the aircraft is high, and
as mentioned earlier, under these conditions it is not possible to extract bl~ and b~A from elevator pulse
tests. However, by measuring the elevator and aileron jack loads in steady flight at the same Mach number
but at two different values of normal acceleration, it is possible to evaluate b ~ and b~A as shown in
Appendix D. The required tests were made by accelerating the aircraft in level flight through the desired
range of Mach number, and then decelerating the aircraft in a steady turn. By comparing values recorded
at the same Mach number during the two parts of such a flight test, the data required for analysis are
obtained. It should be noted that in this case absolute rather than incremental jack loads are required,
and as an appreciable time interval is involved between the recording of corresponding test points,
strain gauge drift may become significant,and as a consequence the accuracy is reduced. Also any two flight
points used in this analysis differ not only in incidence, but also in the elevator and aileron angles required
to trim. The contributions due to control movement terms are generally larger than those due to bl~ and
b~., which are to be extracted from the tests. As a consequence the accuracy of the final result is further
reduced.

5. Results and Discussion.

Fig. 5, 6 and 7 show the ranges of aileron angle, elevator angle and incidence explored during the tests.
These envelopes are presented because tunnel tests show that, in some cases, hinge moments vary nonlinearly with incidence and control angle; thus values of derivatives depend on the range of control angle
and incidence used in a particular test. As an example, Fig. 8 shows some tunnel measurements of aileron
hinge moment at M = 1-02, and the range of elevator angle and incidence covered in the flight tests at
10 000 ft and 40 000 ft. It can be seen that bl~ and 8Cm,/~tl would be expected to differ at the two
altitudes.

Figs. 9 to 14 show the variation with Mach number of the measured values of bl~, b2~, aCHJO~, bl~,
b2~ and OCm,/OrI. The flight results include data from tests at 10 000 ft and 40 000 ft using both pulse and
steady turn techniques. Where possible comparable wind tunnel data 2'3 are also shown. In the cases
where the tunnel tests showed non-linear variations with incidence and control angle, the derivatives
were extracted by considering the incidence and control angle appropriate to the corresponding flight
conditions (Fig. 8).
The flight values of the elevator hinge moment due to incidence derivative b~, Fig. 9, show some
scatter, particularly at transonic speeds where as noted in Section 4.3 the accuracy of the analysis is particularly poor, but the variation with Mach number is well defined, b ~ shows a rapid increase at transonic
speeds and a gradual decrease at supersonic speeds. Thereis no detectable difference between the results at
10 000 ft and 40 000 ft, and this suggests that the fairly simple method used to correct the results for the
aeroelastic distortion of the elevator has been adequate, and confirms the tunnel results which show no
non-linearities over the range of elevator angle and incidence explored. The comparison with windtunnel results is good.
For the elevator hinge moment due to elevator-angle derivative b2~, Fig. 10, the same general remarks
as made for bl~ apply, b2E is roughly twice as large as bl~.
Fig. 11 shows the variation of the elevator hinge moment due to aileron-angle derivative, OCnJO~, with
Mach number. The results show very little scatter, and the derivative is well defined. OCH~/c3~ is small at
subsonic and supersonic speeds, but increases rapidly at transonic speeds. This derivative has normally
been ignored by designers, but clearly its effect on the control hinge moments of this configuration at
transonic speeds is very significant. The wind-tunnel tests do not provide data on this derivative.
The flight results for the aileron hinge moment due to incidence derivative b~a are shown in Fig. 12.
In the subsonic range results from the two alternate test procedures are shown. They differ quite significantly. From the previous discussions on the steady turn tests, it would appear that these are generally less
reliable, and as a consequence more weight has been given when drawing a mean line through the test
points to the results of the pulse tests. There is a rapid increase in btA at transonic speeds followed by a
rapid decrease at supersonic speeds. No aeroelastic corrections have been applied since the calculated
distortion of the ailerons was negligible. However, there are quite significant differences between the results
obtained at 10 000 ft and 40 000 ft. These differences could be due to either an aeroelastic distortion of the
wing, or a non-linear variation of hinge moment over the different ranges of elevator angle and incidence
appropriate to the two test heights. At transonic speeds wind-tunnel results 2,a show marked non-linear
variations of hinge moment with incidence and elevator angle, for example as shown in Fig. 8. At supersonic
speeds wind-tunnel values of the derivative extracted for the 40 000 ft flight conditions are considerably
higher than the corresponding flight measurements. No corresponding wind-tunnel data are available for
the 10 000 ft flight conditions as the appropriate ranges of aircraft parameters were not covered in the
tunnel programme. However, extrapolation of tunnel data in the range M = 1.1 to 1.2, suggests that bta
at conditions appropriate to flight at 10 000 ft is approximately 0-6 less than at 40 000 ft. This difference
is almost identical to that between flight results at the two test altitudes. This suggests that the apparent
effects of altitude in the flight results shown in Fig. 12 are mainly due to non-linearities in the hinge
moment characteristics.
The results for the aileron hinge moment due to aileron-angle derivative bz,,, Fig. 13, show very little
scatter. The results at 10 000 ft and 40 000 ft are different. Again it is thought that this is probably caused
by non-linear effects, but no wind-tunnel results are available for comparison.
Fig. 14 shows the variation of the aileron hinge moment due to elevator derivative, aCnJarl, with Mach
number. This derivative has been generally ignored in the past; however, in this configuration, although
it is small at subsonic and supersonic speeds, it is significant at transonic speeds and is indeed then of
the same magnitude as b2~. Wind-tunnel results show a transonic peak in good agreement with the flight
measurements, but at supersonic speeds, whereas the variation is similar, the agreement is not so good.
The effect of the inter-control gap on 3CnA/~?r1 must be considerable, hence it is possible that Reynolds
number effects may be causing this discrepancy. The present tests have shown that at transonic speeds the
influence of one trailing edge control on the hinge moment of an adjacent control is considerable.

6. Conclusions.
The elevator and aileron hinge-moment derivatives of the Fairey Delta 2 have been determined, using
dynamic techniques, up to Mach numbers of 1.2 and 1.6 at 10 000 ft and 40 000 ft respectively. Where
possible, comparison has been made with wind-tunnel results 2'a obtained with a 1/9th scale model.
The results are summarised in Figs. 8 to 13.
All the derivatives show a charadteristic rapid increase at transonic speeds followed by a reduction at
supersonic speeds. For the elevator, it has been necessary to correct for the aeroelastic distortion of the
control; a relatively simple method of analysis has been used with success, and the results of tests at 10 000
ft and 40 000 ft agree well with one another. On the basis of ground tests, it was calculated that no significant aeroelastic distortion of the ailerons occurred. In this case, the 10 000 ft and 40 000 ft results show
considerable differences; wind-tunnel tests suggest this is probably due to non-linearity of the hingemoment characteristics over the range of elevator angle and incidence appropriate to the two test heights.
The results show that at transonic speeds the influence of one trailing-edge control on the hinge moment
of an adjacent control can be considerable. The influence of this derivative on control hinge moments has
normally been ignored by designers in the past; it would appear that this effect should be taken into
account in appropriate cases.
Although some of the flight results show considerable scatter, the derivatives are reasonably well
defined andthe agreement with wind-tunnel results is reasonable in view of the difficulties involved in the
test techniques and the non-linear variation of some of the derivatives.

LIST OF SYMBOLS

Aircraft rolling moment of inertia

OCL
Oa

a -~----

Lift-curve slope

Aircraft pitching-moment inertia

slugs/ft 2

Constant in equation (A.2)

boE
b1 -

slugs/ft 2

OCn
Oa
OCu~

b2~- aC~
Mean chord of control, aft of the hinge line

ft

Wing mean aerodynamic chord

HA

C ~ - Po V~ SA cA
HE

CH. = Po V?SgcE

CL

Aileron hinge-moment coefficient

Elevator hinge-moment coefficient

Lift coefficient

ft/sec 2

0 = 32"18
Aerodynamic hinge moment of a control, nose up positive

lb/ft

Hj

Elevator jack moment, nose up positive

lb/ft

Hk

Aileron jack moment, nose up positive

lb/ft

Moment of inertia of a control about the hinge line

slugs/ft2

Circular frequency of the aircraft longitudinal short-period

oscillation

sec- ~

Aileron rolling power

A
iA "-~ m s 2

in =

B
m ~2

mA

Aileron mass

slugs

mE

Elevator mass

slugs

Aircraft mass

slugs

Normal acceleration

'a' units

Aircraft rate of roll, positive to starboard

rad/sec

LIST OF SYMBOLS---continued
Aircraft rate of pitch, positive nose up

(r~/~)

rad/sec

Amplitude ratio of rate of pitch to normal acceleration in the longitudinal

short period oscillation

1"A

See Fig. 15

ft

l"g

See Fig. 15

ft

SA

Aileron area, aft of hinge

ft 2

SE

Elevator area, aft of hinge

ft 2

Wing area

ft 2

Wing semispan

ft

Duration of control pulse

sec

true airspeed

ft/sec

Vi

Equivalent airspeed

ft/sec

XA"

See Fig. 15

ft

XE

See Fig. 15

ft

Ya

See Fig. 15

ft

YE

See Fig. 15

ft

Aircraft incidence

radians

Elevator pitching power

~cd~n

Incremental values of parameters

Sweep of control hinge line

degrees

Elevator angle, trailing edge down positive, measured in a plane parallel to

the aircraft centreline
radians
Aileron angle, trailing edge of either surface down positive, measured in a
plane parallel to the aircraftcentreline
radians
Po

= 0.002378

NOTE:

Air density at sea level

A dot above any of the above quantities denotes differentiation with respect to time.

Suffices
A

Aileron

Elevator
Conditions in steady turns at different normal accelerations

slugs/ft 3

REFERENCES
No.

Author(s)

1 W . E . A . Acum

Title, etc.
..

The comparison of theory and experiment for oscillating wings.

A.R.C.C.P.681. March 1962.

P . G . Hutton and D. M o r t o n

Results of pressure plotting and control hinge moment tests on a

1/9 scale model of Fairey Delta 2 in the A.R.A. transonic tunnel.
A.R.A. Model test note J12/1.

G . F . Moss and R. W. Hayward

Unpublished M.O.A. Report.

F . W . Dee and D. G. Mabey

Wind tunnel calibrations of incidence vanes for use on the Fairey

ER 103.
Jl. R. aeronaut. Soc. Vol. 67, No. 628, p. 267. April 1963.

R. Rose, A. A. Woodfield and

C. S. Barnes

Flight measurements of the lift, longitudinal trim and drag of the

Fairey Delta 2 at Mach numbers up to 1.65 and comparisons
with wind-tunnel results.
R.A.E.T.R. 67036. June 1967.

M. Shinbrot

On the analysis of linear and non-linear dynamical systems from

transient-response data.
NACA T N 3288. December 1954.

W . E . A . Acum

F . W . Dee

. . . . . .

Flight measurements at subsonic speeds of the aileron rolling

power and lateral stability derivatives lv and yv on a 60 deg
delta wing aircraft (Fairey Delta 2).
A.R.C.C.P. 739. June 1963.

M . D . Dobson . . . . . .

Wind tunnel tests at supersonic speeds of the Fairey Delta 2.

A.R.C.C.P. 672. October 1962.

10

D.J. Kettle

. . . . . .

8 ft x 6 ft transonic wind tunnel tests on a 1/24 scale model of the

Fairey Delta 2 (ER 103).
A.R.C.C.P. 656. May 1962.

11

S. Neumark

. . . . . .

Analysis of short-period longitudinal oscillations of an aircraft:

interpretation of flight tests.
A.R~C. R. & M. 2940. September 1952.

12

D.R. Andrews . . . . . .

Measurements in flight of the longitudinal stability derivatives

of a 60 deg delta wing aircraft (Fairey Delta 2).
A.R.C.C.P. 639. April 1959.

. . . . . .

....

Aerodynamic forces on a rectangular wing oscillating in a supersonic air stream.

R. & M. 2763, A.R.C. August 1950.

10

APPENDIX A

The Hinge Moments Actin9 on Elevator and Aileron Controls.

Fig. 15 shows the relevant geometric data for the control configuration under discussion. The hinge
line of both surfaces is swept, and this introduces certain difficulties in the definition of the various quantities involved. Unfortunately there is no clearly established rule to guide us here, and it is therefore
proposed to use the following definitions which have also been adopted for the Royal Aeronautical
Society aerodynamic data sheets. Control angles and chords are measured in a plane parallel to the plane
of symmetry of the aircraft, whilst the hinge moments are measured about the swept control hinge line.
A further difficulty arises as the conventional definition for positive aileron deflection is trailing edge
down for the starboard aileron and trailing edge up for the port surface, whereas trailing edge down is
normally considered positive in hinge moment notation; the latter definition is used here.
At any instant in flight the sum of the control jack moment, the aerodynamic hinge moment and the
inertia reactions is zero. Using the definitions shown in Fig. 15, it can be shown for the starboard elevator
that the control jack moment is given by:

- H s = HZ +r E me gn-(x e r~ mt+ Iz cos )~)g1-Ie sec 2 # -(Ye re m e - I e sin )~)p

(A.1)

where the small term due to the aircraft acceleration along the flight path has been ignored. Hj is the jack
moment, H E is the aerodynamic hinge moment, both measured trailing edge down positive, and IE the
control inertia about the hinge axis. Similar equations may be written for the other three control surfaces.
The aerodynamic hinge moment is a function of control deflection and the aerodynamic quantities
describing the aircraft motion, these relationships not always being linear. In theory such a situation
can be dealt with by an 'equations of motion' technique such as that proposed for the extraction of
aircraft stability derivatives by Shinbrot 6, but this would require a resolution of the test data well beyond
the accuracy available from the present flight tests. As the control pulses and consequent aircraft responses
used during the flight tests were generally rather small, it was assumed that the hinge-moment characteristics could be linearised for the range of parameters covered in each test, and the formal analysis has
been made using linearised equations.
Consequently for the elevator, H e is expanded as

He = Po V~Sece

bo~+b~c~+b2~tl+--~-~

qCeV +
0

+ OCn~ OcE

OC~

&c~ ]

(A.2)

The term OCn~/O~ represents the elevator hinge moment induced by aileron deflection; in the present
tests this term was found to be comparable in magnitude with primary control derivatives bl~ and bat.
As the strain gauges used to measure the jack moments were unable to give reliable absolute values,
almost all the analysis was made using incremental values. As a consequence b o could not be determined.
The frequency and amplitude of the control movements used were considerably larger than those of
the resulting aircraft motion and thus one would expect any unsteady aerodynamic effects to be more
important for the former. Estimates based on Acum 1'v suggest that at supersonic speeds unsteady contributions are ver~small compared with those from the steady aerodynamic effects ; while at the subsonic
speeds tested they might produce up to 2 per cent change in the peak hinge moment, and a phase displacement of 12 deg between the peaks of the control input and resulting hinge moment. In fact, the flight

11

records showed that the peak of the hinge moment did not lag by more than 10 deg behind the peak of
the control input. Since this is equivalent to less than 2 per cent in amplitude, it has been assumed that
unsteady effects have a negligible influence on the level of the hinge moment peaks.
Thus equation (A.2) is reduced to
AHE = Po V~ SE cE bl~ Aa + b2E

. aCH~A~)\
At/+---~

(A.3)

where the symbol A denotes incremental values.

For the starboard elevator equations (A.1) and (A.3) give

-AHj = Po V2 SECE (bl~ Ac~+b2~ A

OC/tE

+ rE mE g A n - (xE rE mE +IE cos 4) A 0 - Ig sec 4 A/~- (YEre mE- Ig sin 4) A/~.

(A.4)

A similar expression may be written for the starboard aileron incremental jack moment.

--AHK=PoV:SAcA(bl.4Ac~+b2AA~ + -0CHA
- ~ - At/) +
+ r A ma g A n - (xA rA ma + Ia cos 4) A0 - 1 A sec 4 A~ -(YA rA rnA--IA sin 4) Ai0.

12

(A.5)

APPENDIX B
Evaluation of b 2 and Cross-control Derivatives from Control Pulses.
If a control pulse is of short duration, at the instant of maximum control application the aircraft will
not have responded sufficiently to affect significantly the aerodynamic hinge moments. Consequently
at that instant, incremental values measured from the initial trim condition, Aa, Ap and Aq are negligible,
but in the present tests their time derivatives are significant.
In the elevator pulses analysed no significant aileron movement occurred, and consequently A/~ was
zero. The incremental values of the elevator and aileron jack moments, for both the port and starboard
sides are given by:
- A H s = Po V 2 S ~ c E b 2 ~ A t / + r e m e g A n - ( x ~ r ~ m e + I ~ c o s 2 ) A g l - I E s e c 2 A i i

(3CH.4At/+ rA mA g

- AHr = Po V2 Sa cA -~q

A n - (x A r A mA + IA cos 2) A~.

(B.1)
(B.2)

These equations can be solved for b2~ and OCHA/Ot/, the remaining terms being either known aircraft
parameters or variables of motion. In fact, only At/was measured, A/] and A ~ were not measured and An
was measured by an instrument too coarse to give the required resolution. As we are dealing here with the
initial response to a control pulse before rate of pitch and incidence have had time to build up to any
extent, the effects of aircraft damping may be ignored, giving:
A~ =

2g OC,,
--At/
CL iB ~ at/

(B.3)

An - OCL At/
Ot/ C L"

(B.4)

Values of OCm/Ot/and OCL/Otl have been obtained from earlier tests made on the aircraft and reported
in Ref. 5.
The most plausible procedure for estimating the value of A# is to assume that the elevator movement
can be approximated to a sine wave in this region, say

At/(t) = At/maXsin(l'I T)

(B.5)

where T is the duration of the control pulse.

The value coinciding with the maximum t/is,
A# = -At/

(B.6)

Having thus conveniently defined all the variables in terms of At/, the peak value of elevator, equations
(B.1) and (B.2) can now be solved for b2~ and aCHA/0t/ for both port and starboard controls
-AH s
b2E = l po V2 SEcEArl

S FrEmE OCL 2(x~rEmE+I~cos2) OC m

StCE L m
Orl
rtli B~
Ot/

/ seo ( )21
13

BT,

0rt

1 Po V~2 SA CAArt

SA ca L m

m :tB c:

art

(B.8)

Ort ]

Considering the aileron pulse, the acceleration in roll is determined as for ~ in equation (B.3).
2g

P - CL iA~Sl~ A~.

(B.9)

Flight s and tunnel 9'10 values of l~ show reasonable agreement, and where possible flight values have been
used. Applying a procedure identical to the elevator case, equations for the evaluation of b2. and the cross
derivative OCmJa ~ can then be obtained, giving

b~ = po v ~ s ~ , ~

OC~_._
04

-AH,
P0 Viz SE c~ A~

s~ ~

r. sec

L~o~k

b2EA~+
ac

T2(yArAmZAS

2(yEr~me-I~sin2t~

SE cE

m iA S

l~

(B.10)
(B.10)

where the alternate signs apply to starboard and port respectively. In the present tests, a small correction
for At/has been retained in equation (B.11) as this term is significant, although A# is negligible.

14

APPENDIX C

Determination of b 1from the Stick Fixed Longitudinal Short Period Oscillations.

For this motion incremental values are measured between successive peaks of the aircraft oscillation.
In the present tests, Ap, AiO,A# and A~ are negligible. Following similar reasoning to that of Appendix B,
it can be shown from Ref. 11 that the aircraft motions in normal acceleration and pitching are given by,
An -

a Aa
CL

A4 =

-~

(C.1)

~A<~.

(o2)

This relation has.been derived ignoring the effect of the damping derivatives, since the damping was small
in the flight tests.!J is the frequency of the longitudinal short period oscillation and (77/~)is the amplitude
ratio of pitching velocity to normal acceleration in this oscillation. Both these parameters, and the untrimmed lift curve slope, were measured in previous flight tests 1z.
The incremental jack moment for the elevator is, from equation (A.4),

-AHa = Po VgSEC~ b 1EAc~+b2EA i ? +OCH~

~-A~) \

(DI

rem~g+(x~reme+Iecos2)<~'

~-~LAa.

(C.3)

From this equation ble may be derived, as b2E and OCHE/O~are known from Appendix B. In the present
tests, the term involving A~ is negligible. Thus

=AH s
A~? aS [rJ~mm~+(x~remE+i~cos2)Z_(7~
bl~ = Po V~ SEc~Ac~ bz~-~ Sece
rag\nil

(C.4)

In the case of the aileron, the terms in both Aq and A~ are negligible in the present tests. Thus,

mg \n) J

i5

(c.5)

APPENDIX D

Evaluation of Elevator and Aileron Hinge Moment Derivatives due to incidence, bl,from flight at different
values of normal acceleration.

In flight at constant normal acceleration, the angular acceleration terms can be ignored and equations
(A.1) and (A.2) reduce to,

c~C~7

-Hj = Po V~SEc~ bo~+b~+bB~q+-U(- j +rEm~gn

(D.I)

a similar equation may be written for the ailerons.

Taking tests at the same Mach number, but different normal accelerations, say, nl and n2, we get,

l -b2E -

blE-kOoSEcEA

where suffices 1 and 2 refer to the different test conditions and A signifies the increment between test conditions, b2E and aCnJO~ have been derived in Appendix B.
A similar expression can be derived for the ailerons.

16

TABLE 1

Leading Aircraft Dimensions.

Gross wing area

S = 360 ft 2

Semi-span

s = 13-42 ft

Nominal centreline chord

2 5 ft

Tip c h o r d

1"83 ft

Mean aerodynamic chord

= 16"75 ft
4 ~ Symmetrical, max tic at 29.5 ~ chord

Wing section
Trailing-edge angle

5.2

Leading-edge sweepback

59.9

Trailing-edge sweepback

Twist

Dihedral

Incidence with respect to fuselage datum

+ 1"5

Mean weight at test condition

12800 Ib

Pitching moment of inertia coefficient, iB

0.269

Rolling moment of inertia coefficient, i a

0.0558

17

TABLE 2

Control Surface Details.

Aileron
rn A, mass

2"38 slugs

IA, m o m e n t of inertia a b o u t the hinge

2"77 slugs ft z

SA, area aft of hinge

16"04 ft 2

CA, m e a n chord aft of hinge

2"70 ft

rA
XA I~ Fig. 15

0"65 ft
8-52 ft
10-01 ft

YA
2, sweep of hinge

9.7

H o r n - b a l a n c e area

0-57 ft 2

Rigged up angle

Elevator

rg/E,mass

3" 15 slugs

I~, m o m e n t of inertia a b o u t the hinge

6"17 slugs ft 2

SE, area aft of hinge

20-18 ft z

c~, m e a n chord aft of hinge

3"69 ft
0"75 ft

xe

Fig. 15

7"57 ft

Y~

4"53 ft

2, sweep of hinge

9.7

18

IOj

L2 ~ J "~j

SCAL~ - ~

LL

FIG. 1.

v_

G e n e r a l a r r a n g e m e n t o f the F a i r e y D e l t a 2.

-'';

~::,:~
. , ~ f

. ~ .

r - - ~
?'~r

4"

"(t"

~.'

~" * :

Flo. 2. The Fairey Delta 2.

(3111
I-.a

Iij
Z
Qw

t~

<

ul<
,_1
m
I11
.A
Z
ul
Cl
j
:w

MACH
NUMBER
= 1-16
ALTITUDE
= 4-1,600 St

O0_
_ 4

<

ul
bO

vtj

Z
ul

Z
I~tu

O
i
+

+4"

v
o

Ill
.J
IiJ

ll:uj

J
v
<A I_
"3tt1
ZE
ul
,J
<

4-

TIME

FIG. 3a.

Idealised time history of an elevator

pulse.

FIG. 3b.

SEC

Flight record of an elevator pulse.

_A

~
o
n-

<

MACH
NUMBER=I'~
ALTITUDE
= 4 | , 0 0 0 ~t

_7 .
ILl

no

_5 -

LU Z
eL_____ <
<

-- 3 " -

P_

n,,

ILl

__i
~ ,~

-5%

l..u

-3 -

X<

~
Z

_7 -

V:

z~
0
LU
-J

n~z~
0
ILl
-J

<

3:=
<tn,, n-- w

--I
ILl

Q-gs-

- i#oo'

-1
ILl

+Io
b_
0 j
-I

h
(:3 j
<n."

%ec
o

-I0
cff

0~:

- P-O

_3 ~,

0
_

- ~ '

2~ .

u3

TIME.

FIG. 4a.

Idealised time history of an aileron pulse.

FIG. 4b.

SaG

Flight record of an aileron pulse.

O
I.U
td
0E

-8

iJj
I

Z
z

O,

tu
..I

LMACH

ci
I:O .-1-4

NUMBERz_.___

FIG. 5. Envelope of aileron angles covered in the flight tests.

ta

-8

i.u

./

bj
,.-I - 4 z

I' 9' TRIM tIHE

n,,

1"2

ILl
-J
UJ

I-4
1.6
HACH NUMBER

I'$i TR#H ktNl,

+4

FIG. 6. Envelope of elevator angles covered in the flight tests.

+8
tu
I/d

lal

0-6 1 0"8
I
/
t
I'~' TRIM LINE

>
Z
i

1.0~-""-"~,J,/~Z_//A/~,///,,.,~
~
g-4t.6
''"
MACH NUMBER

m~

FIG. 7.

Envelope of incidence covered in the flight tests.

23

- I -4-

INTERPOLATION TO
F L I G H T CONDITIONS

-I -2
-I .0

F L I G H T A T 4-0j00oT'+..
INITIAL TRIMMED CONDITION.
ELEVATOR PULSE.
NIBSEOUENT OSCILLATION.

~ 0 -

TUNNEL
R E g U L T S F'OR
4"0,000 "if,CON :::IlTION5

X 40,O00ff, PULSE
+ I0~000~ PULSE
0 40~000r% T U R N
[] lO,O00rrl".T U R N

: xx x
x'~X ~x ><

-O-B

El

blE

-2JO

2-0

8-0

4.0

I0"O
06.

-0-10

-0'~

k-A.

N,x/

I V

.v

o.

0"6

0-8

I. 0

1"2
1"4"
MACH NUMBER

0-~

El
0-4

~'1.1

~5"0'1

O-G
0,8

FIG. 8.

A.R.A. wind-tunnel results at M = 1.02

for aileron hinge moments.

\\

-0-~

l,J
4~

fJ

- 0 "6

FLIGHT

y.,~/
/ N,\x

FIG. 9.

Elevator hinge m o m e n t due to incidence

derivative v e r s u s Mach number.

1"6

+ +

-2- 2

+ I0~000~ PULSEI

"1+
X

-1-5

+.

-I .6

- 0 "8

-I ,4

- 0 -6.

i'l

[ :,o, ooo

c3CH~

y. ^

-1"2

~2E

~
\

- I "0

+4*.
-0-g

-0"2

-I-/ /
"f

-0"6

I
t
I
I
I

0-4-

T U N N E L RESULT~ FOR /
4-070OO,,Ct CONDITIONS

-0.2

O-A.
0.8

I'O

I-P..
I-4NUMBER

FIG. 11. Elevator hinge moment due to aileronangle derivative versus Mach number.

-0"4

0.6

0.8

MACN

0.4-

0-6

1.0
I-P
I-4MAC...H NUMBER

FIG. 10. Elevator hinge moment due to elevatorangle derivative v e r s u s Mach number.

1.6

I-6

-~-4,

4o,0o0 c~, PULSE

+ i0,000~ PULSE
e 40.000Ft TU RN
I0,000~-~, TURN

-I -8

_lm

-2"0

- I "6

A
~q ~

-I -i

+ 10,000"f'k PULSE I

"TUNNEL RE~UI-" S FOFt

40.~0~]0-F+. CONOITI0~S

~,

-I

I\1

I\I

--I -I

/~- . . . .

~.

"',1

"'"

-beA
-I-4-I-0

6~ A

o,/,11, 7,c.

I
z

- 0 "g

-0"6

IO,OOOfl:

40,000~

Ix.)

-I-0

--t , l ~ . .

"

l
.f

=! I,' ~ + +
I Jr! '\

-0-

ll,l

~I I
--0",

'ti

.~+~rI
+

I
"

' ' ' ~ 1+

"'T
40,OOOFt
-0-4

,o, o o o
-0"2

l
Oi-A
0-4-

-0"2

0"6

0-8

1.0

I-P1.4MACH NUMBER

FIG. 13. Aileron hinge moment due to aileronangle derivative v e r s u s Mach number.
.AA

0.4-

FIG. 12.

0-6

0 -~

I-0
1"2
1"4MAEH NUMBER

Aileron hinge moment due to incidence

Derivative v e r s u s Mach number.

1"5

I-6

-2-0
Fo

-1"6

r~

-1.4

AIRCRAFT
CEI;tTRE OF_
GRAVITY

-1"2

-I .0
A

-0.8
:)R

~A

~E

-0"6
R
-0-4

- 0 -2

0
"I"

~-~;~

v"

~::P

I'U

I "r"

I'~1

"

MACH HUMBER

FIG. 14. Aileron hinge moment' due to elevatorangle derivative v e r s u s Maeh number.

I ' C{:)

ELEVATOR CENTRE
OF G R A V I T Y

FIG. 15.

AI L E R O N

OF

Control geometric data.

CENTRE

GRAVITY

R. & M. No. 3485

Crown copyright 1967

Published by
HER MAJESTY'S STATIONERY OFFICE
To be purchased from
49 High Holborn, London w.c.l
423 Oxford Street. London w.1
13A Castle Street, Edinburgh 2
109 St. Mary Street, Cardiff
Brazennose Street, Manchester 2
50 Fairfax Street, Bristol 1
35 Smallbrook, Ringway, Birmingham 5
7-I 1 Linenhall Street, Belfast 2
o1 through any bookseller

R. & M. No. 3485

S.O. Code No. 23-3485