NASA CR-2276

NASA CONTRACTOR
REPORT

CN
CM

STOL RIDE CONTROL

FEASIBILITY STUDY
Technical Report
by C. K. Gordon and R. 0. Dodson
Prepared by

THE BOEING COMPANY

Wichita, Kansas 67210
for Langley Research Center

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

WASHINGTON, D. C. JULY 1973

2. Government Accession No.

1. Report No.

3. Recipient's Catalog No.

NASA CR- 2276

4. Title and Subtitle

STOL RIDE CONTROL FEASIBILITY STUDY - <TECpICAL REPORT

5. Report Date
July 1973
6. Performing Organization Code

8. Performing Organization Report No.

7. Author(s)

C. K. Gordon and R. 0. Dodson

D3-9052-1
10. Work. Unit No.

9. Performing Organization Name and Address

The Boeing Company
Wichita Division
Wichita, KS
67210

NAS1-11683

12. Sponsoring Agency Name and Address

Contractor report

National Aeronautics and Space Administration

Washington, D.C. 20546

14. Sponsoring Agency Code

741-86-99-01
11. Contract or Grant No.

13. Type, of Report and Period Covered

15. Supplementary Notes

16. Abstract

The feasibility of developing a ride-smoothing control system for a

20-passenger turboprop STOL transport was assessed. Five different ride-control
system configurations with varying degrees of complexity, performance, and cost
were investigated. Results indicate that a satisfactory ride-control system can
be practically implemented on the aircraft with minimum flight performance
degradation.

17. Keywords (Suggested by Author(s))

STOL

18. Distribution Statement

Unclassified - unlimited

Ride smoothing
Ride quality
Active controls

19. Security Qassif. (of this report)

Unclassified

20. Security Qassif. (of this page)

Unclassified

21. No. of

22. Price*

103

$3.00

For sale by the National Technical Information Service, Springfield, Virginia 22151

TABLE OF CONTENTS
PAGE
1.0
1.1
1.2
1.3
1.4

INTRODUCTION
Objective
Background
Feasibility Study
De Havilland DHC-6 Twin Otter Characteristics

1
1
1
1
2

2.0

SYMBOLS

3.0
3.1
3.2
3.3
3.3.1
3.3.2
3.4

DESIGN CRITERIA
Flight Conditions
Design Turbulence Conditions
Performance Goals
Ride qualities
Handling qualities
Safety Criteria

4.0

CONTROL SURFACE CONFIGURATIONS

5.0
5.1
5.2

5.4
5.4.1
5.4.2

CONTROL SYSTEM SYNTHESIS AND PERFORMANCESALYSIS

Mathematical Models
Longitudinal Synthesis and Performance Analysis
Acceleration Feedback
Longitudinal Synthesis Acceleration Feedback
Longitudinal Performance Analysis
Acceleration Feedback
Longitudinal Synthesis and Performance Analysis
Angle of Attack Sensor
Lateral-Directional Synthesis and Performance
Lateral-directional synthesis
Lateral-directional performance

6.0
6.1
6.1.1
6.1.2
6.2
6.3
6.4
6.5

PRELIMINARY DESIGN
Utility Systems
Hydraulic system
Electrical system
Structural Modification
Control System Mechanisms
Electronics and Control Implementation
Hardware Location and Weight Distribution

49
49
49
49
50
50
50
57

7.0

AERODYNAMIC ANALYSIS AND TRADES

61

8.0

CONCLUDING REMARKS

75

5.2.1
5.2.2
5.3

.
. .

. .

.-.'

-7
7
7
'8
8
10

10

in

.
....

11
17
17
18
18
22
35
37
37
41

TABLE OF CONTENTS (CONT'D.)

PAGE
APPENDIX
APPENDIX
APPENDIX
APPENDIX
APPENDIX
APPENDIX

A:
B:
C:
D:
E:
F:

PERFORMANCEPOLARS
. . . . . . .
LONGITUDINAL EQUATIONS
LATERAL-DIRECTIONAL EQUATIONS . .
STABILITY AND CONTROL DERIVATIVES
AIRPLANE INERTIA AND GEOMETRY . .
SIGN CONVENTIONS

REFERENCES

.
.
.
.
.
.

77
81
85
87
91
93
95

IV

LIST OF FIGURES
FIGURE
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.

PAGE
de Havilland DHC-6 Twin Otter - 300 Series
Design turbulence condition
Ride control surfaces
Wing control surfaces
Elevator surfaces
Rudder surfaces
Vertical ride control system block diagram
The effect of acceleration feedback on vertical
acceleration at the c.g
Short period root locus, closing acceleration loop only
Short period root locus, closing pitch rate loop around
fixed acceleration gain
PSD/RMS of c.g. vertical acceleration for the free
airplane at landing condition
Vertical acceleration linear analysis
Comparison of free airplane and RCS power spectra of
vertical acceleration aft passenger station
Effect of aileron actuator bandpass on vertical acceleration
Effect of aileron rate limit on vertical acceleration
Effect of lift growth on vertical accelerations
Vertical accelerations
Effect of nonlinearities on vertical acceleration
Vertical acceleration response to 1-cos gust
Pitch handling qualities
.
Surface deflections
Vertical ride control system angle of attack sensor
Vertical acceleration angle of attack sensor
Dutch roll and spiral mode root loci, closing lateral
acceleration only
Lateral ride control system
Lateral acceleration-rudder RCS
Comparison of free airplane and RCS power spectra of
lateral acceleration - aft passenger station
Comparison of free airplane and RCS power spectra of
lateral acceleration pilot station
Rudder surface deflections .
Hydraulic flow capacity
Rudder control system
Elevator control system
Aileron control system
Spoiler control system
Major hydraulic component location
Ride control hardware locations
Present flap configuration
Inboard aft flap effectiveness-40 flap

3
9
12
13
14
15
19,
20
21
21
23
24
26
27
28
29
31
32
33
34
36
38
39
40
44
45
46
47'
48
51
53
54
55
56
58
59
62
63

LIST OF FIGURES (CONT'D.)

FIGURE
39.
40.
41.
42.
43.
-i44.
45.
46.
47'.
48.

PAGE
Aileron chord extension effectiveness - 40 flap
Flap gap effectiveness
.
Three dimensional lift effectiveness for a .17 semispan,
.10 chord spoiler
Aileron and spoiler locations
Elevator required to trim symmetrical spoiler
Aileron required for manual backup - landing condition
Elevator required to trim - critical flight condition
. . . ; . . .
Minimum directional control speed -10 flap
DHC-6 Twin Otter performance polars - 0 flaps
DHC-6 Twin Otter performance polar - 40 flaps . . . . ,

VI

64
=65
66
49
70
71
72
73
78
79

LIST OF TABLES
NUMBER

PAGE

I.
II.

Feasibility study flight conditions

Angular acceleration an6^fa'te^er.f.qrmance
longtudinal control system .'".0!.-'!<?.r;V^V ? , . . . . . . . . . . . . .
III.
Angular acceleration and rate performance - " "*ftj-r'v
lateral system

IV.
Lateral handling qualities
V.
Hydraulic system design requirements . .
v . .
VI.
Electrical power requirements . . . . : .
..-.. .;. ...
VII.
Ride control weight estimates - newtons (pounds)
. . r. .
;: . .
VIII. Results summary
.,
.
IX.
Airplane stability derivatives
X.
Control surface derivatives
XI.
DHC-6 dimensional and inertial properties

VII

7
35
42
;43

49
50
57
' 76
88
,89
91

STOL RIDE CONTROL FEASIBILITY STUDY

TECHNICAL REPORT

By C. K. GORDON and R. 0. DODSON

THE BOEING COMPANY, WICHITA DIVISION

SUMMARY

Development of ride-smoothing systems is an important part of advancing STOL transport

technology. A study of the feasibility of developing and certificating a ride-smoothing system for a
passenger STOL aircraft has been performed by Boeing-Wichita, with de Havilland Aircraft of
Canada, Limited, as the principal subcontractor. The de Havilland DHC-6 Twin Otter was selected
for the feasibility study as it is the only STOL vehicle presently certificated and in use by a number
of air carriers in this country.
Five different ride control system configurations with varying degrees of complexity, performance
and cost were investigated. The total system provides symmetrical aileron, elevator, rudder and
spoiler ride control surfaces, with accelerometers and rate gyros for motion sensors. The lesser
systems eliminate the spoiler, rudder and elevator surfaces respectively. An alternate total system
was synthesized utilizing a boom-mounted angle-of-attack sensor instead of an accelerometer.
The feasibility study has indicated that an acceptable ride control system can be practically
implemented on a DHC-6 Twin Otter with minimum airplane performance degradation. Vertical
and lateral accelerations can be significantly reduced using split aileron and split rudder control
surfaces. A split elevator may be required to provide adequate handling qualities. Further, spoiler
control surfaces must be added if significant acceleration reductions are desired during landing
approach.

IX

1.0

1.1

INTRODUCTION

Objective
The overall objective is the advancement of the. technology of STOL aircraft
ride-smoothing systems. Such systems will probably be required for passenger acceptance
of any low- to moderate-wing-loading STOL vehicle due to its operational environment
and flight dynamic characteristics. The immediate objective is examination of the
feasibility of developing and certificating a ride-smoothing control system for a STOL
aircraft currently in airline use. The de Havilland DHC-6 Twin Otter was selected for the
feasibility study as it is the only STOL vehicle presently certificated and in use by a
number of air carriers in this country.

1.2

Background
Concern over the unwanted response of aircraft to atmospheric turbulence dates back to
the initial NACA reports by Hunsaker and Wilson in 1915. Indepth research regarding
the use of active control systems for ride smoothing did not begin until the late 1940's
and early 1950's when the United States commercial airline fleet started expanding
rapidly. The work at Langley Aeronautical Laboratory, NACA, by Phillips and Kraft*
exemplifies this research. The introduction of the jet transport, with its higher wing
loading and cruise altitude, led to an improved ride without an active ride control system
and thus greatly reduced the immediate need for ride smoothing research. However,
recent and proposed extensive future use by feeder airlines of STOL aircraft, which fly
low, and with low to moderate wing loading, has resulted in a ride environment inferior
to conventional aircraft, and has thus renewed interest in ride smoothing efforts.
Development of a ride smoothing system is therefore an important step in the
accomplishment of NASA's goal of advancing STOL transport technology.

1.3

Feasibility Study
A five month feasibility and planning study was conducted by the Wichita Division of
The Boeing Company under NASA Langley Research Contract NAS1-11683 during.the
second half of CY-1972. The de Havilland Aircraft of Canada, Limited, assisted as the
principal subcontractor. Results of the study are contained in this report.
Synthesis and preliminary design were accomplished to modify the existing DHC-6 Twin
Otter flight control system by incorporating a control system for ride improvement in
the vertical and lateral axes.
Several system configurations were designed to provide trade studies of ride quality,
airplane handling, and complexity. Initially turbulence response in all six motion degrees
of freedom were examined and compared with tentative criteria proposed as goals for
the ride-smoothing system performance. Only vertical and lateral aircraft motions were
found to need ride smoothing to meet these tentative criteria.

1.4

De Havilland DHC-6 Twin Otter Characteristics

The Series 300 Twin Otter has a maximum gross weight of 55,600 N (12,500 IDS.) and
carries 20 passengers and a crew of two. It has a nominal wing loading of 1436 N/m2 (30
Ib/ft2), making it susceptible to ride quality degradation in severe turbulence. The Twin
Otter is powered by two Pratt and Whitney PT6A-27 turboprop engines. Maximum
cruising speed is 92.6 m/sec (180 KTAS), while the maximum altitude is 3049 m
(10,000 ft), due to the unpressurized cabin. Maximum range is 1399 km (755 n.m.).
However, the average flight time is less than 45 minutes for normal commuter service in
the United States. Approximately 90 of the 366 Twin Otters in service operate in the
United States under FAA Part 135 rules. A three-view illustration of the Twin Otter is
shown as Figure 1.

3.81M
(12 FT 6 IN)
19.8 M
(65 FT 0 IN)

5.79M
(18 FT 7 IN)

I
4.50M
I
r (14 FT 9 IN)H
15.77M
(51 FT 9 IN)

6.30M
(20 FT 8 IN)

IG720473-38C

Figure 1: de Havilland DHC-6 Twin Otter - 300 Series

2.0

SYMBOLS
b

wing span, m (ft)

Tf

mean aerodynamic chord, m (ft)

CD

drag coefficient, dimensionless

C|_

lift coefficient, dimensionless

(C^ ,CM C|\j)

aerodynamic moment coefficients about the (x, y, z) body axes,

dimensionless

Cj

thrust coefficient, dimensionless

(C^, Cy.C^)

force coefficients along the (x, y, z) body axes, dimensionless

normalized acceleration

dxx, \ yv , \ zz )

moments of inertia with respect to (x, y, z) body axes, kg-m^

(slug-ft2)

lxz

product of inertia with respect to x and z axes, kg-m^ (slug-ft^)

^~T, unit imaginary number

gain constant

turbulence scale length, m (ft)

p, q, r

perturbation angular velocity vectors along (x, y, z) body axes,

rad/sec

q~

dynamic pressure, N/nrr (Ib/ft2)

reference wing area, rrr (ft )

LaPlace operator

U0

steady flight velocity along x body axis, m/sec (ft/sec)

(u, v, w)

perturbation linear velocities along (x, y, z) body axes, m/sec (ft/sec)

total airplane velocity, m/sec (ft/sec)

v_

component of gust velocity along body y axis, m/sec (ft/sec)

weight, N (Ib)

w_

component of gust velocity along body z axis, m/sec (ft/sec)

^sensor

body station of sensor, m(ft)

Xc g

body station of center of gravity, m(ft)

inertial acceleration along y body axis, g or m/sec^ (ft/secz)

. . .

")

r y

Z'

Inertial acceleration along z body axis, g or m/sec^ (ft/sec )

ct

airplane angle of attack, rad

/?

airplane sideslip, rad

6^

aileron deflection, rad

elevator deflection, rad

flap deflection, rad

rudder deflection, rad
spoiler deflection, rad
roll, pitch, and yaw Euler angles (Reference 8), rad
f\

power spectral density of vertical turbulence, (m/sec)

rad/sec
damping ratio

incremental

root mean squared

frequency, rad/sec
undamped natural frequency, rad/sec

SUBSCRIPTS:
AIL

aileron

command

FUS

fuselage

gust

max

maximum

SS

steady state

tot

total

3.0

DESIGN CRITERIA

3.1

Flight Conditions
A survey was conducted of existing data on operational profiles of the DHC-6 Twin
Otter. Typical climb, cruise and landing approach conditions were selected for design
flight conditions and are tabulated in Table I, below. Performance polars are presented in
Appendix A.
, TABLE I
FEASIBILITY STUDY FLIGHT CONDITIONS

Condition
Airspeed, m/s
(KIAS)
Altitude, m
(ft)
Gross weight, N
(Ib)

Climb
51.5
(100)

914.4
(3,000)

50040
(11 250)

Cruise
77.2
(150)

1 829
(6,000)

50040
(11 250)

Landing
Approach
36.0
(70)

152.4
(500)

50040
(11 250)

Flap position, rad

(deg)

0
(0)

0
(0)

.698
(40)

CG location, % MAC

30

30

30

2.54
(500)

Rate of climb, m/s

(ft/min)
Glide slope, rad
(deg)

*.

-.1309
(-7.5)

The cruise condition is representative of both cruise and descent for a ride control study,
since airspeeds are similar and the cruise altitude is relatively low. The DHC-6 altitude
ceiling is 3048 m (10,000 ft), and the typical cruise altitude is 1829 m (6,000 ft).
Typical flight durations are approximately 45 minutes, and therefore the constant
weight and center-of-gravity locations are representative of all flight phases.

3.2

Design Turbulence Conditions

A random turbulence intensity with an exceedance probability of 0.01 was selected as a
design condition for this study. Figure 2 shows rms gust velocity with 0.01 exceedance

probability, as a function of altitude"^' . This probability level corresponds to an rms

gust velocity of 2.1 m/sec (7.0 ft/sec) for all the design flight conditions, since the
altitude is 1829 m (6,000 ft) or less for all conditions.
Atomospheric turbulence was modeled with a von Karman spectrum having the
following spectral density for digital computation of airplane response power spectra.
JL
-k
2
*
* W 9 L 1 4 3 (1-339^)
(m/Sec)2
rad/sec
^

where

aw

rms gust velocity, m/sec.

U0

airplane forward velocity, m/sec.

turbulence scale length, m.

(*>

frequency, rad/sec.

A scale length (L) of 762 m (2500 ft) was used for the climb and cruise conditions, and
152 m (500 ft) for the landing condition.^
For hybrid simulation, an analog of random turbulence velocity was generated by
filtering the output of a random white noise source so that the power spectral density of
the filter output approximated the von Karman spectrum. The LaPlace transform of the
von Karman gust filter is:

F(S)

.372

3.3

> < s + 1-372

><S + 1 7 . 7 9 - ) ( S + 2 6 4 . 8 - )

Performance Goals
Positive, well-defined ride-quality criteria for aircraft do not presently exist. However, a
Symposium on Vehicle Ride Quality held at the NASA Langley Research Center, July 6
and 7, 1972, did produce indications of approximate human comfort motion
boundaries. Motion levels felt to be conservative based on the symposium discussions
were suggested by NASA to the contractor for use as ride-smoothing performance goals,
rather than as precise performance requirements. Several ride-smoothing systems with
varying degrees of complexity were to be examined with the intent that performance of
these various systems bracket the given performance goals.

3.3.1

Ride qualities. The primary ride quality goals were to reduce vertical acceleration in
each flight condition to 0.030 g rms or less and lateral acceleration to 0.015 g rms or
less, at all crew and passenger stations, while subjecting the airplane to the design
random turbulence.

IS.OrX 10'
40.0

rXlO

0.01
EXCEEDANCE
PROBABILITY

10.0
H
PL4

'

'

g 20.0

Q
H 5.0

2.1 M/SEC
(DESIGN
CONDITION)

1
2
3
GUST VELOCITY - RMS, M/SEC

GUST VELOCITY, RMS FT/SEC

Figure 2: Design turbulence condition

IG720473-22B

In addition, angular accelerations and rates were not to exceed existing free airplane
values. Specific maximum angular rates of 4 deg/sec for pitch and yaw and 7 deg/sec for
roll were suggested by NASA; however, the existing values were used as goals since they
were smaller for all conditions.

3.3.2

Handling qualities. Pitch short period handling qualities were evaluated qualitatively
by comparing pitch rate and normal acceleration of the aircraft plus ride control system
with those of the free airplane in response to control column step inputs. A minimum
damping ratio of 0.04 was selected as a design goal for the phugoid mode.
Lateral-directional handling qualities from MIL-F-8785B (Reference 7, Level 1 for light
airplanes) or existing values, whichever were less, were used as design goals. In general,
the quantities specified are dependent upon characteristic root locations. Specific values
are tabulated and compared with actual performance in Paragraph 5.4.2.2

3.4

Safety Criteria
The ride control system was designed to provide (1) adequate handling qualities and
safety for continued flight following a single engine failure, total hydraulic or electrical
power failure or a single ride control surface hardover, and (2) safe maneuvering and
landing capability following two engine failures.

10

4.0.

CONTROL SURFACE CONFIGURATIONS

For this feasibility study, the ailerons, elevators and rudder were each split spanwise into
two segments for ride control and manual flight control, as follows:
.

'?'>.

Aileron
Elevator
(

Manual Flight Control

Segment
Percent Span
60

40

80

20

Rudder

Ride Control
;
Segment.
'
Percent Span

70

--

,30

Manual segments are controlled through existing mechanisms. Ride control segments are
controlled by electrohydraulic power actuators that receive electrical position command
signals from both pilot manual and ride control commands.
Spoiler control surfaces, operating from a biased position, were added to augment the
ailerons for direct lift control during landing approach.
Figure 3 defines five surface/sensor combination options that were selected to provide
performance, complexity and cost trade data to aid in selecting a specific configuration
for subsequent implementation and demonstration phases. The cross hatched areas are
electrohydraulically actuated surfaces that accomplish both ride control and manual
control functions (spoilers excluded). The remaining control surfaces are manually
actuated. Under normal operation, the control authority for the modified airplane is the
same as the basic airplane.
Option II uses an angle of attack feed forward sensor while the other options use an
accelerometer as the primary motion feedback sensor.
The aileron and spoiler configurations are illustrated in Figure 4. The existing aileron
trim tab is retained on the left wing and the present geared tabs are moved to the manual
portions of the ailerons. The geared tabs are adjustable to achieve aerodynamic balance
of the manual aileron.
The selected configurations of the elevator and rudder surface are compared to the
existing configurations in Figures 5 and 6; respectively. The split configurations require
aerodynamic balancing of the manual portion of the surface since the ride control
portion eliminates the horn balance. Aerodynamic balance is achieved with more
overhang area in front of the hinge line and adjustable geared tabs. This requires addition
of a gear tab on the right elevator. The present elevator trim tab and flap interconnect
tab are retained. A portion of the present rudder geared tab is retained. The rudder trim
tab is moved to the manual rudder surface.
Aerodynamic analyses supporting the choice of spoilers to supplement the ailerons and
the areas allowed for the ride control surfaces are discussed in Paragraph 7.

11

09

a>

ioi

I
o
o
0)

ACCEL

ACCEL

ACCEL

ALPHA

ACCEL

RIDE
ONTROL
SENSOR

CO

^^

B ~

>

K
u

tfQ
en 'S
a.
cc
O
j5

5. *
u

2"3

sz
a

12

0)

fe

KQ

ow

.-B

WH

sS

- g

Ofc
fc W
W u
bC

nS

13

K
O
H

Q<
W >
HW
U J
WW

CQ
d)

CO

CO

CO

J-l

i
ffi

<

H
H
U

K
W
H
fc

CO

K
Pi

SCQ
K<
HH

14

IT5
0

CO

TO

Q
CO

0)

co

(0

s:

T5

CO

P
W

15

5.0

CONTROL SYSTEM SYNTHESIS AND PERFORMANCE ANALYSIS

Ride control system synthesis was accomplished by digital computation with the linear
equations presented in the Appendices, without lift growth effects. Actuator bandpass,
position limit, and rate limit studies, and final ride quality performance evaluations were
accomplished by hybrid computation with lift growth included.

5.1

Mathematical Models
Small perturbation, linear, rigid-body equations of motion were used for the airplane
mathematical model. A statement of the equations and tabulations of stability and
control derivatives, inertial properties, and geometric properties are contained in the
Appendices B and C. The general form of the equations is derived in Reference 8, in
stability axes. Conditions and assumptions required to derive linear, small perturbation
equations are stated in the referenced manual. Sign convertions are presented in
Appendix C.
The function used to simulate lift growth of the circulatory aerodynamic forces resulting
from airplane positions and rates is given in Equation 3. The functions is in LaPlace form
and is derived from the Wagner function**' .
.1655-.335 5

where

g = .0455 _
~c

(3)

f = .3 _^~c

Similarly, the function below, derived from Kussner's function, simulates the lift growth
of circulatory lift resulting from gusts.

where

.500
.500
"S + .130B Q " S + B Q

<4>

BQ =

Airplane motion variables were filtered, one at a time, through W(S) before multiplying
by the aerodynamic coefficients in the hybrid simulation. The effect of the Wagner
function filter on acceleration responses was negligible, except when applied to the
symmetrical aileron deflections for vertical ride control. The Wagner function filter was
employed only on the aileron deflections to obtain the acceleration data presented.
The turbulence generated using the turbulence model (Paragraph 3.2) was passed
through the Kussner filter, ^(5), and then multiplied by each of the gust coefficients in
the hybrid simulation. The gust coefficients include fuselage effects, which are
noncirculatory. However, fuselage forces are small relative to wing and tail forces.

17

5.2

Longitudinal Synthesis and Performance Analysis Acceleration Feedback

5.2.1

Longitudinal synthesis acceleration feedback. Control surfaces available for vertical

ride control are shown in Figure 3. Ride control systems were synthesized utilizing the
three combinations of vertical control surfaces with acceleration feedback listed in that
figure. Configuration numbers referred to in the following paragraphs are defined in
Figure 3.
Figure 7 is a block diagram of the vertical RCS configuration I. Symmetrical ailerons
and spoilers are driven by eg vertical acceleration, and the elevators are commanded by
pitch rate.
The primary concept was essentially to reduce translational acceleration with feedback
eg acceleration to control surfaces near the eg, and to control angular acceleration and
pitch handling qualities with pitch rate feedback to the elevators.
Effectiveness of eg acceleration feedback to the ailerons was determined for each flight
condition as shown in Figure 8 for the cruise condition. The figure presents vertical rms
acceleration at the eg as a function of acceleration feedback gain. The data was digitally
computed by power spectral methods for acceleration feedback, without feedback filter
or actuator dynamics. A gain of 96.6 deg/g (3 deg/ft/sec^) was selected for the cruise
condition. This work was accomplished for full span aft flaps and ailerons. Later, when
the configuration was changed to partial span ailerons, the gain was increased
appropriately for the decrease in control surface lift authority.
The acceleration gain for climb was set by the acceleration goal also. However, the gain
for landing was set to prevent frequent and prolonged saturation of the ailerons and
spoilers, since they are not effective enough in the landing condition to meet the
acceleration goal.
The spoilers, active only in the landing phase, are commanded by the same sensor and
filters as the ailerons. The feedback gain to the spoilers is half of that to the ailerons,
since the spoilers have only about half as much deflection available from their biased
position.
Feeding back acceleration decreases the natural frequency and increases the damping
ratio of the short period mode as shown in the root locus plot in Figure 9, closing only
the acceleration loop at the cruise condition. This feecback makes the airplane response
to pilot commands sluggish. Angular rate feedback is required to "restore handling
qualities. Angular rate feedback through a lag filter (pseudo integration) essentially
accomplishes a blend of rate and position feedback, the conventional methods of
controlling damping ratio and frequency, respectively.
Figure 10 is a short period root locus with the acceleration gain fixed, and closing the
pitch rate feedback loop to the elevator through a 2/(S + 2) lag filter. With K^ =11.5
deg/rad/sec, the ride control system short period root location was made almost
coincident with the free airplane root. Again, K^ was appropriately adjusted when the
system configuration was changed to partial span elevators. A high pass (washout) filter

18

4
to

o
w
CO

rsi
r>
O

O"

"y

<!>

be
u

ft

.2
-O

ca

w
W

o
3

(M

CO

CO

^
O

Cd

4-

s
(U

-l_>

CO

CO

CO
o

o
ca
O
fc
Q

0
HT CONDI'

O
ii
J
PH

<
B

CM

CO
ct>
&

5
j
CRUISE

i-i

CO

CO

f^

CNI

CM

1-1
in

CO

CD

CO(

*j

o
t^

0)

w
m
S
j

0)
73

<jj

CO

q? :N :N
P ^J M

o
u
19

NOTES:
CRUISE CONDITION
FULL SPAN AFT FLAPS
ACCELERATION FEEDBACK ONLYNO FEEDBACK FILTER
NO ACTUATOR DYNAMICS

GOAL

SELECTED
GAIN
50

100

150

200

ACCELERATION FEEDBACK GAIN, DEG/g

Figure 8: The effect of acceleration feedback on

vertical acceleration at the e.g.
20

250

FREE
AIRPLANE

NOTES:
CRUISE CONDITION
6 = K
Z e.g.

Kg = 96.6DEG/g
0

-2

-3

Figure 9: Short period root locus, closing acceleration loop only

5=11.5 DEG/RAD/SEC

FREE
AIRPLANE
NOTES:
CRUISE CONDITION

"

ACCELERATION
LOOP ONLY

96 6

' sTTs

-2

-3

-1

Figure 10: Short period root locus, closing pitch rate

loop around fixed acceleration gain
21

was included in each feedback to eliminate feedback of steady state acceleration or pitch
rate. The washout break frequency was later changed to 0.25 rad/sec to accommodate
landing approach and climb condition phugoid stability requirements.
Vertical RCS Configuration III is identical to Configuration I, except that the spoilers are
omitted. This affects the landing condition only. Gains and filters in the aileron and
elevator loops are identical to Configuration I.
Vertical RCS Configuration V utilizes only the aileron control surfaces. The acceleration
feedback gain was increased only to a point where the short period root location was still
acceptable (reference Figure 9) without pitch rate feedback. The same high pass filter
was used, with a feedback gain of 13.1 deg/m/sec2 (4 deg/ft/sec2) at all flight
conditions.

5.2.2

Longitudinal Performance Analysis Acceleration Feedback

5.2.2.1

.Longitudinal ride qualities acceleration feedback: For the landing approach and climb
conditions rms accelerations of the phugoid mode alone exceed the 0.03 g rms goal and
cannot be effectively reduced by pitch rate or acceleration feecback. Figure 11 shows
that the rms acceleration resulting from the free airplane phugoid mode in the landing
condition is approximately 0.006 g's, for a 0.3049 m/sec (1 ft/sec) rms turbulence
intensity. The rms acceleration would be approximately 0.042 g's for the design
turbulence of 2.1 m/sec (7 ft/sec).
Attempting to control the very low frequency phugoid mode with acceleration or pitch
rate feedback interferes with even long term pilot commands. It was felt that passengers
would not be sensitive to such low frequencies (30 seconds or more per cycle). Boeing
and NASA agreed to exclude the phugoid contribution to rms acceleration for
comparison with the ride quality goal.
Referring to Figure 11, the cumulative rms at a frequency, w^, is the square root of the
integral of the PSD from CL= 0 to w = u)^. The rms acceleration to be compared with
the goal is then Jo tot
- 0"phu - wnere <?tot 's tne cumulative rms over the total
frequency range, and <7pnu is the cumulative rms over the phugoid range. The value
of tfpnuis taken at the plateau in cumulative rms at the frequency where the PSD
returns to zero from the spike resulting from the phugoid mode.
Figure 12 shows rms vertical accelerations computed digitally by power spectral
methods. The PSD's were generated from linear equations, without lift growth effects.
Accelerations at the pilot, eg and aft passenger (fwd, mid,aft) stations are shown for the
three flight conditions for a 2.1 m/sec (7.0 ft/sec) rms gust velocity. The linear analysis
indicated that the ride quality goal of 0.03 g's or less would be met with partial span
ailerons and elevators for cruise and climb conditions at all stations. In the landing
condition acceleration at the aft passenger station is reduced by 42 percent with spoilers
added and 20 percent without spoilers. A major part of the reduction accomplished by
RCS ailerons and elevators can be accomplished with RCS ailerons alone on the DHC-6
as defined for this study.

22

tl V-

.030

X 10"

RMS Wg = .3048 m/sec (1 ft/sec)

.025

.020
O
El

0)

CUMULATIVE RMS

"bo

.015

1.0

.010

0.5

.005

PSD

2 0 00

4.00

6.00

8.00

10.00

12.00

FREQUENCY, RAD/SEC

Figure 11: PSD/RMS of c. g. vertical acceleration for

the free airplane at landing condition

23

0
14.00

S
K

C) RMS

CQ

C (7.0 FT

w
o
0)

.3

Pn

u
NCE INTENS

3
tf

a
rt
o
u
rt

73

O
^
CO

BU

fi
tf
U

s,3 swa

24

0)

The PSD and cumulative rms of vertical RCS accelerations are compared with those for
the free airplane in Figure 13. The accelerations are aft passenger station responses to
0.3048 m/sec (1 ft/sec) rms turbulence at the cruise condition. The ride control surfaces
are ailerons and elevators. Cumulative rms at a frequency, U)^, is equal to the square
root of the areas under the corresponding PSD from w = 0 to w = o>j. In addition to
reducing the power (or rms) of the response, the power is concentrated at very low
frequencies, which is typical of the RCS at all flight conditions. Practically all of the
power exists below 0.20 Hertz.
Performance analyses were conducted with a hybrid computer simulation that included
lift growth as described in Paragraph 5.1 and actuator dynamics and limits as described
in the following paragraphs. Only two rigid body degrees of freedom were included in
the hybrid simulation. Effects of the phugoid mode were omitted.
The effect of aileron actuator bandpass on acceleration reduction was evaluated by the
hybrid simulation at the cruise condition and is illustrated in Figure 14. Actuator
bandpass is defined as the break frequency of the first order lay simulating actuator
dynamics. An aileron actuator break frequency of 20 radians per second or more
provides adequate performance. This value was selected as the design break frequency
and was used in all further analyses. This data was produced without actuator rate or
position limits. Aft station accelerations is presented in these parametric studies because
it is the maximum acceleration among the various stations.
The effect of aileron actuator rate limit on acceleration reduction, shown in Figure 15,
was evaluated at the cruise condition also. Data was obtained with the aileron actuator
bandbpass set at 20 radians per second and the aileron deflection limit at 20 degrees.
One hundred degrees per second provides satisfactory performance and-was used in all
further control analyses.
Development and maintenance of actuators with this relative high rate may be difficult
and expensive. Increasing aileron authority would reduce this rate in direct proportion to
the aileron area increase. The split surface approach was selected as the most practical
initial design. The scope of this feasibility study did not allow aileron authority design
trades. Using the full aileron for both ride control and manual flight control should be
considered in future work.
Similar analyses were conducted to determine the elevator actuator requirements, which
were found to be quite low. The elevator actuator bandpass and rate limit were rather
arbitrarily set at twenty rad/second and fifty deg/second, respectively, for the following
performance analyses. These were considered practical to implement, and would not
affect the performance.
The effect of lift growth was determined with the actuator dynamics and limits set as
discussed above. The simulation of lift growth is discussed in Paragraph 5.1. Figure 16
shows that the Kussner filter applied to the turbulence decreases acceleration, and the
Wagner filter applied to the aileron deflection increases acceleration. Stated differently,
the Kussner function makes the gusts less effective and the Wagner function makes the
ailerons less effective (at high frequencies). The data discussed below reflects the effect

25

FREE AIRPLANE
AILERON & ELEVATOR RCS
3.0 X10

-4

.030

RMS Wg = .3048 m/sec (Ift/sec)

CRUISE CONDITION

.025

.020
CO

"bD

.015 w
CUMULATIVE RMS

.010

.005

2,00

4.00

6.00

8.00

10.00

12.00

14.00

FREQUENCY, RAD/SEC

Figure 13: Comparison of free airplane and RCS power spectra

of vertical acceleration - aft passenger station

26

NOTES:
CRUISE CONDITION
AILERON & ELEVATOR SURFACES
TURBULENCE INTENSITY, 2.1 M/SEC (7.0 FT/SEC) RMS

. It

FRE E AIR PLAN E

aE-t

.10
1
1
.08 1
1
.06 1
I
\
.04

2
W
J

w
u
u

2
E-i

SE: ^ECT]ED
RT?
Di\ EAK

FR EQUE NCY

GOAL

~ "- -

.02
0

10

20

30

40

50

ACTUATOR BREAK FREQUENCY, RAD/SEC

IG720473-54B

Figure 14: Effect of aileron actuator bandpass on vertical acceleration

27

NOTES:
CRUISE CONDITION
AILERON & ELEVATOR SURFACES

TURBULENCE INTENSITY, 2.1 M/SEC (7.0 FT/SEC) RMS

"bD
03

Q
H
W
J
W

u
u

.12 \
\
\
.10 \
\
\
\
.08
\
\<
.06

<

fc

.04

02

.02

FREE AIRP LANE

'
1

SEL ECTE D
RATE
LIMIT

X
^

GOAL

^ ^^
"

e;n

inn

i ^ n ' orSn ' OR

AILERON RATE LIMIT, DEG/SEC

IG720473-55B

Figure 15: Effect of aileron rate limit on vertical acceleration

28

NOTES:

- . - , . .

CRUISE CONDITION
AILERON & ELEVATOR RCS
TURBULENCE INTENSITY, 2.1 M/SEC (7.0 FT/SEC) RMS

FREE AIRPLANE
CO

RCS

w
u

FWD

MID

AFT

PASSENGER LOCATION

Figure 16: Effect of lift growth on vertical accelerations

29

of both functions. Actually, control deflections result in somewhat noncirculatory

forces, the noncirculatory portion not being subject to lift growth. Consequently the
data are conservative in this respect.
Vertical accelerations at the forward, mid and aft passenger stations are shown in Figure
17 for three flight conditions for a 2.1 m/sec (7.0 ft/sec) rms gust velocity. At cruise,
climb and the landing approach conditions, aft passenger accelerations for the
aileron-elevator ride control system were reduced to 31, 44 and 84 percent, respectively,
of the free airplane accelerations. With spoiler feedback in the landing condition, aft
acceleration is reduced to 64 percent of the free airplane level.
The ride quality goal of 0.03 g's rms is achieved at the forward passenger station for the
cruise condition and at the mid to forward stations for the climb condition.
Actuator break frequencies were 20 radians per second. Aileron and spoiler rate limits
were 100 deg/sec, and position limits were + 20 degrees and + 10 degrees, respectively.
Reasonable elevator limits had no effect on the results.
As previously discussed, linear analysis indicated that the vertical ride quality goal would
be met. Nonlinearities degrade performance slightly, exceeding the desired maximum
acceleration. However, Figure 18 shows that the system meets the vertical ride quality
goal of 0.03 g's rms at the cruise condition for all turbulence intensities up to
approximately 6.2 feet per second, which has a probability of exceedance of only 0.012.
Table II compares the RCS pitch angular responses with those of the free airplane. All
design goals were met at the cruise and climb conditions. At the landing condition pitch
angular acceleration and angular rates were approximately 15 to 20 percent more than
existing airplane values. However, the accelerations and rates are very small, and this
slight increase is considered insignificant.
The ride control system is designed to reduce relatively small accelerations, much less
than structural limit load levels. However, the system does reduce large gust-induced
accelerations, as shown by the 1-cos transient response in Figure 19, and no instability is
caused by system saturation. In this particular case, the gust frequency was selected to
maximize free airplane acceleration. The nonlinear simulation included actuator rate and
position limits.

5.2.2.2

Longitudinal handling qualities acceleration feedback: Short period handling

qualities were evaluated at all flight conditions by comparing ride control system vertical
acceleration and pitch rate responses. for column step inputs to free airplane responses.
The comparison is shown for Configuration I at the cruise condition in Figure 20, fora
two-degree deflection of the full span elevator (manual and RCS). The ride control
system causes a loss of effectiveness of the elevator for relatively sharp inputs. A
crossfeed from column input to ailerons corrects the problem. Initially, the crossfeed
signal cancels the ride control signal opposing the acceleration and then washes out at
the same rate as the ride control signal. The crossfeed transfer function is:

_ ic

S
5 + 25

2 <?p
5+2
Command
30

(5)

LANDING
APPROACH

w
o
CO

^
fn

S5

PQ

O
O
03

<
U

.3
W
O
co
O

G-

'

fclgi ;
>

i
K
O

s,S

'NOLLVH31300V

31

U
CO

co

33
fn

ff
l>
0)

AILERON AND ELEVATOR RCS

CRUISE CONDITION

09
"fafi

CO

. 1OU

*.'/
jj^

55

8
W
J

w
u
u

.100

_l

. '-

.075

55

8
I.025

GOAL /
/

^ ** *

---"^"
]//_.
0

; **>

/^

<

H
fe

^"'
. i

243
DESIGN CONDITION
GUST VELOCITY, RMS M/SEG

, i
C>

,
i
. 1
i
2 4 6
8
GUST VELOCITY, RMS FT/SEC

i
1 0
IG720473-69A

Figure 18: Effect of nonlinearities on vertical acceleration

32

FREE AIRPLANE
RCS

CRUISE CONDITION
Wg = 30 (1-COSTt)

TIME,

SEC
IG720473-71A

Figure 19: Vertical acceleration response to 1-cos gust

33

K
O
H

S
w

J
w

CO

O
CM

.3

55
O

-C

0
w o

CS)

CQ

W
H

"8

ss.O'D,
'O'O,

34

TABLE II
ANGULAR ACCELERATION AND RATE PERFORMANCE LONGITUDINAL CONTROL SYSTEM
RIDE CONTROL SYSTEM OPTION
Free
Airplane

I
Ailerons
Elevator
Spoilers

Cruise

2.6

1.2

1.2

1.5

Climb

2.2

1.7

1.7

1.9

Landing Approach

7.7

8.7

7.6

7.7

Cruise

.7

.4

.4

.5

Climb

.8

.8

. .8

.8

1.3

1.5

1.2

1.3

Condition Mode

III or IV

Ailerons
Elevator

Ailerons

Angular Accel-Pitch, deg/sec2

Angular Rate-Pitch, deg/sec

Landing Approach

Whether the loss of elevator effectiveness for sharp inputs would present a problem,
requiring the crossfeed, should be determined by piloted simulation in Phase II.
The damping ratio of the phugoid mode was also evaluated and was kept greater than
0.04 at all flight conditions.

5.2.2.3

Longitudinal control surface requirements - acceleration feedback: Figure 21 shows

the RCS control surface rms deflections required by a 2.1 m/sec (7.Oft/sec) turbulence
intensity. Aileron deflections are approximately 8 degrees rms for climb and cruise
conditions and 6 degrees for landing approach. Maximum deflections are approximately
three times these levels.
The relatively small partial span elevator deflections indicate that the elevator may not
be required to optimize short period root locations. The elevator feedback does not
improve acceleration responses greatly, as shown in previous vertical acceleration data.
Studies should be conducted during the Phase II design to investigate reducing, or
possibly eliminating, the elevator segment used to improve handling qualities.

5.3

Longitudinal Synthesis and Performance Analysis Angle of Attack Sensor

Synthesis and performance evaluation of the angle-of-attack system were accomplished
completely by hybrid simulation, and for the cruise condition only. The angle-of-attack
sensor was located on a boom on the nose of the airplane. The equation for the sensed
parameter is:
tot

<xsensor ' Xcg>

35

(6)

2.1 m (7 FPS) RMS TURBULENCE

9:0
^^M

8.0

o
w

7.0

CO

tf

6.0

PH

^^^Mi

co"

1 1

H
O
j
En
P
W

u
<J
EH

5.0

tcD

CQ

4.0
3
U

I
rn
u

3.0

_IT-

0
K
PH
PH

PH

jfL

2.0

^5

CO

1I

1.0

CO

<

J_4

B
p^

^M

nA

ELEVATOR

AILERON

Figure 21: Surface deflections

36

SPOILER

Each term in the numerator is an increment of vertical velocity at the sensor: airplane
velocity, gust velocity, and the velocity resulting from pitch rate. Sensed ot tot was then
filtered and used as a feedforward command to symmetrically controlled ailerons. Figure
22 is a block diagram of the angle of attack system.
The time delay between sensing the gust and the resulting force on the wing was
simulated. The second order portion of the PCS filter provides the required phasing, of
ailerons relative to the time delay, to minimize acceleration response. The second order
filter also attenuates high frequency gust signals which command unnecessarily high
aileron actuator rates. The natural frequency and damping ratio were varied to optimize
the filter regarding acceleration reduction. A Pade'linear approximation for a time delay
was tried in place of the second order filter, at a number of time delay periods.
Acceleration reduction was as good, but it did not attenuate the high frequency
commands as well. Several combinations of Pade' approximations and second order filters
were also tried, without improvement in performance. The filter would have to be tuned
to the best compromise among the various flight conditions, if implemented, since the
time delay between the angle of attack at the boom sensor and lift on the wings is a
function of airplane velocity.
A high pass (washout) filter eliminates feedback of steady state angle of attack. Actuator
dynamics are siulated with a first order lag 20/(S + 20). Pitch rate feedback to the
elevators was tried, but it did not improve performance.
Ride quality performance for the system is shown in Figure 23. At the cruise condition
the angle-of-attack ride control system reduces aft passenger acceleration to 36 percent
of the free airplane level compared to 31 percent with acceleration feedback. The same
actuator limits and bandpass were used for both systems.
5.4

Lateral-Directional Synthesis and Performance

5.4.1

Lateral-directional synthesis The rudder is the only existing control surface that can be
used for reduction of lateral accelerations. Forces applied by the rudder are not effective
in reducing translational acceleration of the eg, although lateral accelerations aft of the
e.g. resulting from angular acceleration can be reduced.

,9

The rudder has much more rotational authority than translational. The problem is
evidenced in a root locus as shown in Figure 24. The Dutch roll and spiral modes are
shown, closing only the acceleration loop for the cruise condition. The spiral mode
rapidly goes unstable as Ky is increased, and is highly unstable at Ky gains required to
reduce translation acceleration significantly. The problem is even more severe at the
othe; flight conditions. The spiral modes goes unstable faster if the washout frequency is
.decreased (or the washout is deleted), and slower if the frequency is increased. However,
the washout begins to interfere with control of accelerations in the Dutch roll range of
frequencies when the washout break frequency is increased above 0.50 rad/sec.
The criterion used for the spiral mode root location was a minimum time to double
amplitude of 20 seconds7. State differently, the spiral mode root must be less than plus
0.035. The gains shown in Figure 24 reflect the authority of a full span rudder. In the
final configuration, with a partial-span RCS rudder, the Ky used corresponds to
approximately 1.57 deg/m/sec^ (0.48 deg/ft/sec2) in Figure 24. Feedback of yaw rate

37

r.

p'
8

P "

. g
.>
Q

CO
00

o
CJ

Z CO

Qtf
WW

<g

^ Q

g ,'
_

<3

B "

0
CM

CO

Ctf

CO

bfi

< ^i

^
K?

WHEEL INPUT

S 2 +20S+400

RCS FILTER

^i

S+.2

CD
B

to

CO
11
0

c
o
0
0)
T3
|H
(4
11

rt
o

'

>

OO

exi
CM

QQ
irD

S>

c3
r

I-H

O
H

~g
- s
w
M

Wed

5S
_i
gM
rt
38

'

CRUISE CONDITION
TURBULENCE INTENSITY, 2.1 M/SEC (7.0 FT/SEC)RMS

.15
CO

"bD
W

.10

i
.05
W

GOAL

o
0

FWD

MID

AFT

PASSENGER LOCATION
IG720473-79A

Figure 23: Vertical acceleration-angle of attack sensor

39

= .46DEG/M/SEC
^ = 1.74 DEG/M/SEC2
N.

= 3.38 DEG/M/SEC2
= 6. 56 DEG/M/SEC

NOTES:
.50

INSERT
3.0

CRUISE CONDITION
GAINS ARE FOR FULL
SPAN RUDDER

2.0

1.0

SEE INSERT

-2.0

-1.0

\ **h

-*J^

1.0

2.0

Figure 24: Dutch roll and spiral mode root loci, closing
lateral acceleration only

40

(with proper filtering) tends to restore the spiral stability and increase Dutch roll
frequency. However, all attempts to control the spiral mode at higher acceleration gains
were unsuccessful. Figure 25 is a block diagram of the final lateral-directional ride
control system.
A number of other combinations of feedback filters were tried, as well as feeding back
roll attitude to the PCS ailerons, in order to achieve a higher acceleration gain.
Generally, in making the spiral mode sufficiently stable with higher acceleration gains, a
new, low frequency mode of oscillation is created. The rms accelerations for a full
spectrum of frequencies may be increased from those of free the airplane for such
systems. However, the power is concentrated at very low frequencies. The lateral
acceleration goal could be met, or very nearly so, if accelerations below 0.10 Hertz were
excluded. Since this artificially created low frequency mode is unconventional, and its
effect on passenger comfort is unknown, the system was not recommended.

5.4.2

Lateral-Directional Performance.

5.4.2.1

Lateral-directional ride qualities: Lateral acceleration responses to 2.1 m/sec (7.0

ft/sec) random turbulence are presented in Figure 26. The lateral ride control system
significantly reduces aft passenger accelerations by reducing rotational acceleration. The
eg cannot be translated effectively by a force introduced at the rudder, although
translational accelerations are reduced somewhat in the landing condition. A lateral force
surface located near the e.g. would be required to achieve the design goal of 0.015 g's.
The additional complexity of such a surface was considered impractical for this program.
Figures 27 and 28 compare lateral accelerations for the free airplane and RCS. These
figures show aft passenger and pilot station acceleration responses, respectively, to
0.3048 m/sec (1 ft/sec) rms turbulence at the cruise acceleration along with the
reduction in rms. Even though the forward station acceleration is increased (reference
Figure 26), the frequency is still shifted downward, which may be an improvement in
comfort.
Angular acceleration and rate design goals were to maintain airplane responses with the
ride control system below corresponding free airplane responses. Table III shows that the
RCS roll and yaw rates and accelerations met the goal for all flight conditions.

5.4.2.2

Lateral-directional handling qualities: Existing handling qualities or the requirements

of MIL-F-8785B' Level 1 for light airplanes, whichever were more stringent, were used
as design goals for lateral handling qualities. Table IV presents the comparisons. All root
locations meet the MIL-F-8785B'levels except the spiral mode time-to-double at the
landing condition, which meets existing values.

5.4.2.3

Lateral-directional control surface requirements: Figure 29 shows the RCS control

surface rms deflections required by a 2.1 m/sec (7.0 ft/sec) turbulence intensity. The rms

41

rudder deflections are approximately 3.0 degrees at landing approach and approximately
1.5 degrees for cruise and climb. This indicates that the rudder ride control segment
could be reduced, and should be investigated during Phase II.

TABLE III
ANGULAR ACCELERATION AND RATE PERFORMANCE LATERAL SYSTEM
FREE
AIRPLANE

CONDITION -MODE

RUDDER
RIDE CONTROL
OPTION I, II & III

Angular Rate- Yaw, deg/sec^

Cruise

4.0

1.0

Climb

2.9

1.4

Landing Approach

3.5

2.7

Angular Rate- Yaw, deg/sec

Cruise

1.5

.6

Climb

1.4

.8

Landing Approach

2.2

1.2

Angular Accel-Roll, deg/sec

Cruise

4.4

2.9

Climb

2.9

2.1

Landing Approach

3.0

2.3

Angular Rate- Roll, deg/sec

Cruise

1.0

.9

Climb

.9

.6

Landing Approach

1.5

.7

42

'

*-"c
CO

ll

f rt

t S2

01

CO

i>

O)

in

(S

^5
^ 0

"to Q

CO ^~

01
c _

.2

*j~:

<n

0)

+^ o
0 CM

CM

03

in

E "o

CO

F CO

1^

CO
LU

"to

00

CM

1 r-

'x ^

LU ili

r^

*-m

CM

CD

-M

i {s

8
a

_ a
OQ X

13
Lit
DC
LU
QC

c
3

in

H
^<
-1
cc

.2
*t

1
O V

1
C>
4-1

O
i-;

O)

CM

<

CO
LL

LU

a
CO
LU

0)
TJ
O

O1

'o.

E
Q
"5

(0

'C
(U

CJ

LO

LT5

Lf)

AI

AI

AI

C^
LO

^f
-<t

co

i-

DC

To
CO
irt
~
_g **
ra
DC
01
C

(O

'5. 'C
en
<D
-;
E
(D
Q 6 A|

o>

co

r-

C3

A|

A(

c
I
I
ii

01

3
l-

CJ

43 .

-o

-1

^^
c\j

c
(O

CM

0)
CO

0)
03

faD
73

<N
CO

CO

CO

w-"

+
co

co

fctD

~o
in

CO

**

-'

^.

5
w

0)

CQ

CM

CO

Si

O O
CO
CM

II

Irt

CO
CM

D
Q
W

u
CO
00

in

O
U

44

o
CM

51
u"

CO

w
o
w

CO

o
ft

I!
U

o
o

1
W

H
S
W

CO

CO
CM

0)

^
J
W

s,S

'NOLLVH3130DV

45

FREE AIRPLANE
RUDDER RCS
.007

3.5rX10

RMS Wg = .3048 m/sec (1 ft/sec)

CRUISE CONDITION
CUMULATIVE RMS
U

CO

bfi

2.00

4.00
6.00
8.00
10.00
FREQUENCY, RAD/SEC

Figure 27: Comparison of free airplane and RCS power spectra

of lateral acceleration - aft passenger station

46

FREE AIRPLANE
RUDDER RCS
1.4rX10

-5

.007

RMS Wg = .3048 m/sec (1 ft/sec)

CRUISE CONDITION
.006

.005

.004
bfl

CO

.003

CUMULATIVE RMS
.002

.001

2.00

4.00 6.00 8.00 10.00

FREQUENCY, RAD/SEC

Figure 28: Comparison of free airplane and RCS power spectra

of lateral acceleration - pilot station

47

12.00 14.00

2.1 M (7 FPS) RMS TURBULENCE

ftoU

-,

W
Q
CO

3,0

i
K
U

2
1

W
rff*
CM

fc

i
m

10

^ "

O
g

u
u
^

CO

s
3

Figure 29: Rudder surface deflections

48

6.0

PRELIMINARY DESIGN
In addition to the control surface and control system synthesis, preliminary designs were
initiated for utility systems, structural modifications, control system mechanisms and
electronics.

6.1

Utility Systems
The utility system load requirements were based on those required to supply the more
complex Option I configuration. The reduced supply requirements for the less complex
configurations would not have a significant impact on the design or costs.

6.1.1

Hydraulic system. Hydraulic system flow rates and capacities were determined by
actuator piston area, pressure, monent arm, surface loads, surface rates, number of
actuators and number of .servo valves. For the preliminary design, average rather than
peak rates were utilized for the pump selection since the accumulator can make up short
duration demands. The servo valve leakage was based on using an Abex SV2 valve on the
aileron system since it requires the maximum rate. Pilot inputs were not included with
the ride control inputs since they are expected to be relatively low and are not
necessarily additive. The design requirements shown in Table V resulted in the selection
of an Abex AP05V-7 engine driven pump. Two such pumps feeding a common reservoir
through appropriate check valves will provide undegraded ride control performance
following a single pump or engine failure at all but the flight idle power.condition. The
hydraulic pump capacities are shown on Figure 30.
TABLE V
HYDRAULIC SYSTEM DESIGN REQUIREMENTS
Design Characteristic
Number of actuators
or valves
Actuator area - m2 x 10
(in2)
Moment arm - m
(in)

Aileron
2
180.6
(0.28)

0.0889
(3.5)

Elevator

Rudder

Spoiler SV Leakage Total

180.6
(0.28)

180.6
(0.28)

180.6
(0.28)

8.0828
(3.26)

0.0889
(3.5)

0.066
(2.6)
i

Average rate - rad/s

(deg/s)
Average flow - m^/s x 1 0
(gal/min)
6.1.2

0.873
(50)
27.9
(0.433)

0.262
(15)
7.88
(0.125)

0.436
(25)
7.06
(0.122)

0.524
(30)

12.4
66.2
(0.197) (1.05)

121.5
(1.93)

Electrical system. The current DHC-6 electrical power requirements, exclusive of

optional equipment, are shown with the estimated ride control requirements in Table VI.
Ample DC power is available. However, there is no margin available on the AC power.

49

Two additional 75 volt-amp static inverters are required to provide redundant power for
the ride control electronics.

TABLE VI
ELECTRICAL POWER REQUIREMENTS
type

Existing

28V DC

2-200 Amp
Generators

Present
Requirements
1 50 Amps

RCS
Requirement
1 Amp

115V AC

6.2

6.3

126V A

1-175 VA
Inverter

49 VA

* Structural Modification
Preliminary structural modification in the form of layout drawings were completed for
.the aileron, elevator, rudder, and spoiler control surfaces. Surface splits were made at
logical separation locations such as ribs and struts. Existing hardware was retained and
utilized in the modified system as much as practical. Modified configurations are
;
' illustrated in Figures 4, 5, and 6.

Control System Mechanisms

The split surface control system concept minimizes changes to the flight control linkage
and components. Principal features of the required changes have been incorporated into
the structural design layouts. The split surfaces were chosen for the feasibility study to
provide a simple inexpensive means of manual backup following a system failure.

6.4

Electronics and Control Implementation

The manual mechanical control authority is sufficient to counteract any electrical or
hydraulic failure of the powered segment, including hardovers, without switching off the
failed system. In addition, electrically separated dual channel electronics are included in
this preliminary design to provide fail-soft capability. Trade studies including safety,
system reliability, and cost should be accomplished during subsequent design activities to
determine redundancy requirements,
f
Power surfaces will float to a neutral position following loss of hydraulic power, and will
be flutter free. Negative structural feedback is provided for all actuators.
The ride control system is connected in series and therefore does not move the pilots'
control. Trim systems, connected in parallel, move the primary controls, as in existing
airplane configurations.

50

o
o

tf
U

Q
W
W
W

25
W

o
ci
o

l-

o
CO

W
J
Q
I-H
EH
ffi
O

CO

o
o

_r

eg
o
o

o
o

- axvH MOT a aovnaAv

T
CO

NIFI/TLVO - 3XVH

3OVH3AV

51

Figures 31 through 34 are implementation block diagrams of the rudder, elevator,

aileron, and spoiler control systems, respectively. The primary control surfaces are split
spanwise into two segments, with RCS feedback to one segment and with pilot authority
over both. The manually controlled segments are controlled through existing
mechanisms. Each. RCS segment is controlled by an electrohydraulic actuator that
receives electrical position command signals from two sources: pilot manual commands
and ride control commands. These two signals are summed in the valve drive amplifier.
Pilot manual commands to the RCS surfaces are sensed by LVDT's attached to manual
:
control linkages. .
i

Yaw rate and lateral acceleration, sensed at the aft passenger location, are gain
scheduled, filtered and summed to form the lateral ride control command signal to the
30 percent RCS rudder segment. The fail-soft preliminary design control system contains
dual signal channels with two stages of monitoring between channels for failure
detection. An electronics monitor channel is compared to the primary channel, and an
unfavorable comparison switches off the ride control signals. Following such a failure,
the pilot retains control of both rudder segments.
A second logic circuit compares actuator position to an electronic model of the actuator
dynamics. A failure detection in this circuit switches off hydraulic power to the
actuator. The pilot then controls the airplane through the lower, 70 percent span rudder
segment.
The inboard 80 percent segments of. the elevators are bussed together are controlled
through existing elevator mechanisms. The elevator control system, similar to the rudder
control system, consists of the segmented surfaces, dual channel electronics, and two
stages of failure detection. Pitch rate constitutes the ride control feedback to the
elevators. Left and right hand powered segment positions are compared; therefore, an
actuator model is not required.
The aileron control system is similar to the rudder and elevator systems with manual and
ride control commands superimposed on the electrohydraulically powered segment.
Manual inputs to the ailerons accomplish roll control with differential deflections.
Vertical ride control is accomplished by symmetrical deflection of the 40 percent
powered aileron segments commanded by vertical acceleration sensed at the eg. Manual
inputs are subtracted for comparison of left and right hand actuator positions.
Spoiler control surfaces are added for vertical ride control in the landing approach flight
phase. When active, the spoilers deflect, from a biased position. Spoiler bias is
commanded by the flap actuator, with a dead-zone between flap deflection and the bias.
The specific amplitude of the dead-zone will be determined in subsequent work. An
override activated by the throttles may be required to prevent spoiler bias during takeoff
or "go-around". The spoilers are deflected symmetrically to augment the ailerons for
direct lift ride control during landing approach and are commanded by the same
feedback. Feedback gain is varied as a function of spoiler bias so that commanded
deflections are smaller than the bias.

52

09

T3
5
TH

CO
CU

53

5 K

3s

O w

0)
4->

03
>.
CO
1-4

1
Cvl
CO

cu

SDIWVKAQ 3NVldHIV

54

o
IM

A-

SDIWVNAa

55

-^^^^ "

L!

nr

o
o
n
r*

V(V(V

>

\
\
1

IT* '^^./J

1
I
I

<o
to
CQ
14

1
o

CO
0>

J-l

56

6.5

Hardware Location and Weight.Distribution

The ride control hardware locations are shown in Figures 35 and 36. Conservative weight
estimates for the ride control system are presented for the various options in Table VII.

TABLE VII
RIDE CONTROL SYSTEM WEIGHT ESTIMATES - NEWTONS (POUNDS)
OPTION

Description

1
Aileron
Elevator
Rudder
Spoiler

II

III

IV

Same as
1 with
a Sensor

Aileron
Elevator
Rudder

Aileron
Elevator

Aileron

93.4 (21)

Structural

200 (45)

200 (45)

133 (30)

Actuators

31 1 (70)

311 (70)

222 (50)

178 (40)

Hydraulics

245 (55)

245 (55)

222 (50)

200 (45)

Electrical

142 (32)

142 (32)

133 (30)

116 (26)

98 (22)

Electronic
& sensors

129 (29)

129 (29)

125 (28)

107 (24)

98 (22)

Angle of attack
Boom

Total

1 029 (231)

66.7 (15)

1 094 (246)

57

835(188)

694

66.7 (15)

89 (20)
178

(40)

(156)

530

(119)

u
o

c
o

t
o
o
o

t
J3

f^

CO

58

co

.3
0>
fn

!H
OJ

C
O

CO
CO

59

7.0

AERODYNAMIC ANALYSIS AND TRADES

The full span aft flap segment and aileron were initially considered for vertical ride
control surfaces. However, the aft flap segment was ineffective as a ride control surface
during the landing approach with 40 degrees of flap deflection. Preliminary estimates
indicated that the authority of .the ailerons was adequate to meet the acceleration goal
during climb and cruise, but not during landing. Aerodynamic trade studies were
conducted among methods of supplementing the ailerons during landing. Methods
investigated were (1) increasing the aileron chord, (2) varying the fore flap gap and (3)
addition of spoiler surfaces. Spoilers wwere chosen in the preliminary design.
The existing DHC-6 Twin Otter flap configuration is shown in Figure 37. The Twin Otter
has double-slotted flaps extending to 44.2 percent semispan. Outboard of this station are
the droop ailerons attached to a slotted fore-flap which deflects when the flaps are
extended. In the extreme 40 degree flap position the inboard fore-flap is deflected 40
degrees and the aft flap segment about 60 degrees with respect to the wing chord plane.
The outboard fore flap and aileron are deflected 26 degrees.
The inboard aft flap effectiveness, including effects of the propeller slip stream on the
flap at the low thrust level during landing, is shown in Figure 38; for the landing
condition.
When the flap deflects beyond the nominal position the incremental lift begins to
decrease, so that this surface is unacceptable for ride control during this condition. Since
the outboard aileron is deflected only 26 degrees at this condition, it does not have this
characteristic. However, the effectiveness of the aileron is not sufficient to meet the ride
quality goal during landing, so that an auxiliary ride control surface is required.
The effectiveness of a large chord on the outboard aileron was considered. The effect of
an aileron chord extension is shown in Figure 39. The present aileron is a 17 percent
chord surface. Increasing the aileron chord to 34 percent will gain about 20 percent
effectiveness above the present aileron. To meet the vertical acceleration goal at landing
will require two to three times the present aileron effectiveness.
The effect of varying the fore flap gap was also investigated as a possible auxiliary force
producing method. Effectiveness of this method from reference 11 is illustrated in
Figure 40. Varying the Twin Otter fore flap gap about the nominal does not produce
appreciable incremental lift, so that this method was unacceptable for a ride control
surface.
A biased hinged flap type spoiler was then considered for an auxiliary ride control
surface. The spoiler has good effectiveness and incremental lift characteristics when
deflected from a bias position. The incremental lift characteristics of the spoiler chosen
for the Twin Otter ride control system is shown in Figure 41. During the landing
approach condition the biased, 10 percent chord, 17 percent semispan, hinged flap type
spoiler is required with the aileron to approach the 0.03 g vertical acceleration goal. The
spoiler is biased at about 12 degrees and deflected + 10 degrees for the ride control
system inputs. During the climb and cruise conditions the spoiler is not required.

61

o
o

w
wW

62

09

O>
^i
PL|

FLAP GEOMETRY

H
S3
W
ii

U
KM

78

fe
W

o
u
H
fe
J
J
H

.74

.70

^
W

S
u
-20

-10

10

20

FLAP DEFLECTION
IG720473-12D

Figure 38: Inboard aft flap effectiveness - 40 flap

63

PROPOSED AILERON EXTENSION

.016
O

w
p

-PRESENT
CONFIGURATION

.012

DOUBLE
PRESENT
-CHORD -

CO
CO

.008

fc
H
O

.004

w
H
fa

0
0

.10

.20

.30

.40

.50

AILERON CHORD RATIO, C Ar*lT Tl <

IG72047>GD

Figure 39: Aileron chord extension effectiveness - 40 flap

64

s
o
.w

8
DHC-6
FORE
FLAP
GAP

I
O

1.0
2.0
3.0
4.0
FLAP GAP PERCENT CHORD

5.0
IG720473-7B

Figure 40: Flap gap effectiveness

65

.625c

HINGED FLAP
SPOILER CONFIGURATION

fc

E9
w
o
u

W
g

10

20

30

SPOILER DEFLECTION, <5.

40

50

DEC.
IG720473-34C

Figure 41: Three dimensional lift effectiveness for a . 17 semispan,

. 10 chord spoiler

66

The spoiler incremental lift and moment characteristics on the Twin Otter wing were
estimated using the technique in Reference 12 and wind tunnel data from Reference 13.
The two-dimensional spoiler characteristics with zero degree flap were estimated using
the method established by Barnes . The two-dimensional lift characteristics with 40
degree flap were then obtained by using wind tunnel data in Reference 13, and
correcting the zero flap estimate. The three-dimensional data was obtained by strip
integration of the two-dimensional data over the spoiler span. Propeller slipstream effects
on the spoiler were not considered due to the very low thrust setting during the landing
approach condition.
The location of the outboard edge of the spoiler was limited by the inboard edge of the
aileron so that the aileron effectiveness would not be disturbed. The inboard edge of the
spoiler was determined from buffet and pitch trim effects due to the spoiler wake
affecting the flow over the horizontal tail. Location of the spoiler on the Twin Otter
wing is shown in Figure 42.
The elevator required to trim symmetrical spoiler deflection is presented in Figure 43.
Data are presented for the forward eg limit (0.2c), the aft eg limit (0.36c) and the eg for
the three study conditions. The pitching moment due to the spoiler deflection was
determined from the spoiler incremental lift center of pressure and drag estimates. For
the extreme eg conditions, 0.5 to 0.6 degrees of elevator will be required to trim the
pitching moment due to the symmetrical spoiler deflection at the bias position.
The effect of the 12 degree spoiler bias on the Twin Otter stall speed will be from zero
to a 3.5 percent increase install speed, depending on how the biased spoiler affects the
stall characteristics of the Twin Otter wing. At most, the Ciu
will be decreased by the
max
incremental CL at the bias position in Figure 41. This corresponds to a 3.5 percent
increase in stall speed for the 40 degree flap landing condition.
The aileron authority required for manual backup was determined using MIL-F-8785B'
as a guide. This reference specifies roll authority requirements in terms of time of bank.
Using the Twin Otter roll time constant at the landing condition, the Level 1, 2 and 3
requirements were converted to roll helix requirements. These levels are shown in the
plot in Figure 44.
For the landing flight condition the present Twin Otter aileron meets the Level 1 roll
helix requirement. Roll authority data were not available for a portion of the present
aileron span, thus the partial span control authority was estimated. Although the
calculated roll helix for the total aileron does not coincide with the actual value (analysis
data substantiated with flight test data), the computed roll helix is conservative. To meet
a Level 3 requirement, the outboard 60 percent of the present aileron span is required for
manual backup.
The amount of elevator authority required for manual backup was determined from trim
requirements at the critical flight conditions. The elevator required to trim an aft eg
(0.36c), 40 degrees flap, power off condition and a forward eg (0.25) flaps up power off
condition is shown in Figure 45.

67

The present elevator required to trim the flaps up, forward eg, power off, Ciu
is
rhax
about -20 degrees, trailing edge up. The critical condition for the the trailing edge down
deflection is 40 degree flap, aft eg, power off, and minimun C|_. This requires about 15
degrees trailing edge down. The present elevator deflection limits are -25 degrees trailing
edge up and 16 degrees trailing edge down. Increasing the maximum trailing edge down
deflection to 20 degrees allows 80 percent of the present elevator for manual backup.
A rudder structure assembly break and the ride control system lateral acceleration
requirements determined the percentage of present rudder span to be used for manual
backup. Following a hydraulics failure the reduced rudder authority will decrease
crosswind capability from 12.35 m/second (24 knots) to about 8.75 m/second (17
knots) for the 40 degree flap, 35.68 m/second (70 knots) landing approach condition.
The minimum directional control speed for the total rudder is 33.64 m/second (66
knots) and 39.63 m/second (77 knots) for the manual portion of the split rudder as
shown in Figure 46. Takeoff speed with 10 degree flap maximum gross weight is 41.69
m.second (81 knots). However, a double failure (hydraulic power and engine) would
have to occur to increase the minimum control speed from 33.64 m/second (66 knots)
to 39.63 m/second (77 knots).

68

(0

a
(0

^5 W
W U

o<

K H

S)

69

50

SPOILER /

BIAS

1.2

ELEVATOR DEFLECTION, 6 DEG.

e

Figure 43: Elevator required to trim symmetrical spoiler

70

1.6
TEDOWN
IG72O473-20C

.16
.14

MIL-F-8785B
LEVEL 1, FLYING QUALITIES CLEARLY
"ADEQUATE FOR MISSION
\\\\\\\\\ \\\\\\\\\\\\\\\ \\\\\\\\\\\\ v j

.12
ACTUALW

.10

LEVEL 2, FLYING QUALITIES /

ADEQUATE

.08
LEVEL 3, SAFELY /
CONTROLLED

.06
MANUAL
AILERON
SPAN

.04

.02

.2

.4

.6

.8

1.0

FRACTION OF PRESENT AILERON SPAN

IG720473-18

Figure 44: Aileron required for manual backup - landing condition

71

a
'oaa ~ 9

HOXVASTS

72

n ax)

Ol

o
o
O
W
CO

8
GO

Q
W
W
OH

a
Q
W
H
<

U
ii

"o
I

auo u

XNSIMOW o^aMVA

73

8.0

CONCLUDING REMARKS
Results of the feasibility study indicate that an acceptable ride control system can be
practically implemented on a DHC-6 Twin Otter with minimum airplane performance
degradation. Vertical and lateral accelerations can be significantly reduced using split
aileron and split rudder control surfaces. In addition, a split elevator may be required to
provide adequate handling qualities. Further, spoiler control surfaces must be added if
significant acceleration reductions are desired during landing approach conditions.
Primary data for the various design options are shown for direct comparison in Table
VIII. Design goals for the vertical and lateral accelerations are 0.03 and 0.015 RMS g's
respectively for a 7 fps RMS turbulence. Vertical acceleration varies with system
complexity. The weight penalty is modest for all configurations and airplane range
performance penalty is minimal.
During Phase II design a major effort should be made to use the full aileron span for
both ride control and manual control. Increasing aileron authority would reduce the
aileron actuator rate in direct proportion to the aileron area increase, and would provide
sufficient direct lift control force to achieve the acceleration reduction goal at most
flight conditions and stations.
Studies should be conducted during Phase II design to investigate reducing, or possibly
eliminating the elevator segment. The feasibility study has shown that the elevator may
not be required. Data also indicates that the area of the rudder ride control segment can
be reduced.
The scope of this program did not include implementation trade studies. Further, design
criteria have a strong impact on system complexity. Trade studies should be conducted
to define the most practical combination of design turbulence conditions, airplane
acceleration criteria, and ride control system implementation.
Although the ride control system was designed to endure failures of the powered
segments, dual channel electronics were included in the preliminary design configuration
to provide fail-soft capability. Trade studies of safety, system reliability, performance,
and cost should be conducted during Phase II to define the most efficient
implementation method.

75

05

'

^ O t-"
^4

LO t**

000
' '

'C3

'

<

, "."
05
O
_,

55
fc

..

1-1

Vjj

^"""^

OS

1-1

r\
^J >ri
>"i

52; u ;

QJ
^^ ^^^
^d

<rf 13w '

CO .

X -.

Lft

.1,

o
^V

;.*;. i ;.

w
-i m o
CO CO tooo

- -^

CO

^ &

<-,-;
05
D
0
fe. "

,
-:; ..' - ' .

''

5 H 05
H^

S;S$:

'

PH
O

/', '

CO
1t
1

in

.0
"'

hH

< W 05

<

co
05
^ffi

to

in
o*
fc

&o

<iIB^ w^

OJ C- CM
'-H CSI CO
000

> - -

Q
s

^^

H
CO
CO

co

IH in in
co co in
ooo

^ ^^ Ed ^^

S > Q. J
j W Q O
<* p^ 05 co

d
CM

76

W
>
*

RFORMANCE
NALTY

*"*

CLIMB
CRUISE
LANDAPP

W
Q

^ ,

Q-

^o

Ip

co"

T ^^

05
^j fn p5 05
11

'.

CO
CO

VERTICAL ACCEL.

CQ

H ITS tO

CQ i-i 0

a>

'

' .'

I'

IIGHT PENALTY
WTONS (POUNDS)

tf

'

CLIMB
CRUISE
LAND APP

LATERAL ACCEL.

w>Q

SCRIPTION

s
s

0'

~1

W U
APH

- .

APPENDIX A

PERFORMANCE POLARS
.
*> *
i
s
,

Figures 47 and 48 present performance polars for a flaps up and flaps down 40 DHC-6 Twin Otter
configuration. The data were obtained from an internal de Havilland Report Number 71-6. The
longitudinal-force coefficient, C'x, is given by

){ cT - CD

(7)

where trie thrust coefficient, Cy, is normalized with respect to the freestream dynamic pressure and
wing area.
Data for three flight conditions, cruise, climb and landing approach, are indicated on these polars.
Note that at the landing condition the thrust coefficient is negative. During approach conditions
negative' thrust coefficients of the order -.1 to -.2 are encountered, due to the propeller pitch
control drag being larger than thrust during these conditions.

77

APPENDIX A

0
.3

.2

.1

-.1

-.2

-.3

LONGITUDINAL THRUST COEFFICIENT, C

Figure 47: DHC-6 Twin Otter performance polars - O9 flaps

78

APPENDIX A

u
V

H
Pn

O
H

LANDING
APPROACH

O
U

.2

.1

-.1

-.2

-.3

LONGITUDINAL THRUST COEFFICIENT, C

Figure 48: DHC-6 Twin Otter performance polar - 40 flaps

79

APPENDIX B
LONGITUDINAL EQUATIONS
The LaPlace transformed longitudinal equation are presented below in matrix form:

S + Xu

u(S)
w(S)

0(S)Wg(S)

<J E (S)+

"

Definition of longitudinal equation coefficients:

m\U0

U0

_slm
W,,

~W

"W.

81

m
2

I sp
qS

')

Zw

Mu

vy
IVY \2U

lyy

qSc ,c

( M (5

lyy

yy

wg

"

Mw

82

/ong/t.

n
0

du
*-Sw

fis
dct

83

Page Intentionally Left Blank

APPENDIX C
LATERAL-DIRECTIONAL EQUATIONS
The La Place transformed lateral-directional equations are presented below in matrix form:
S + YV

SYp-g

'xx

S2 + N_S S2 + N S

N.

if, +

*R

_NR_

lj>'(S)-

S
_NV1

Definition of lateral-directional equation coefficients:

qSb

. CY

"

N.

'zz

m2U 0
lzz2U0Nr

Nv

.g

qSb
L

xx

XX

0Q

c
^

85

N,

v.(S)
y

APPENDIX C
Definition of stability derivatives, lateraWirectional equations:
N0

CI"N r

/ n

/ pb

86

acN

APPENDIX D
STABILITY AND CONTROL DERIVATIVES
Longitudinal and lateral-directional stability and primary control derivatives were established for the
cruise, climb and landing flight conditions. The derivatives were obtained from an internal de
Havilland Report Number 71-6. The data were presented as a function of angle of attack, thrust
coefficient and flap position. Airplane stability derivatives are tabulated in Table IX, and control
surface derivatives are listed for both manual and powered surfaces in Table X.

87

APP.ENPIX-D

TABLE IX
AIRPLANE STABILITY DERIVATIVES

" Derivative
'

Cj_j '

...

nondim..

, .

nondim;

C
,una

Cn

Crs

"- - U

Flight Condition
Climb

' Cruise
.35.c<v,i

.788

Landing "
-- :

'.. *,-V

1.61

.047 ......

.076

,:: -.

.191

/rad. '

.16 . .

. -435

t- :;'

1 20

nondim.

.139

.228

' .

.;.'.; J

- .

~ . is

'

-57

CL

/rad.

5.9

6.13

5.62

CL

/rad.

1.72

1.93

V-1&

CL

nondim.

G- :

/rad.

."Gfia

/rad.,.. .

-1.16

-..CMa

/rad.

-6.62

CM
., u
CM

nondim.
' :<.
, . /rad.

Cy

, : /rad.

'CY

, : /rad/ '

5.50 * ''"'":

r
Cp
*P

Cp
' - P
Cn

-.85

-.io'vv<

/rad. :

/rad. .

-.12-.,^

/rad.
'" ',
/rad.

.55 .t

'

.13

>

v '" ,-T
C

N0

CM
IN
r

-1.34
-6.49

" o

-22.75

-22.75

-.87

-1.03

-.10

' ! ; \ l *.^ib

.50

:-11

;v.v. ,: ..50
^:7

/-?

-.55

-.55

.185

'

.124

- " .140

.446

: -'
'

/rad

.122
{

CM

-7.45

.50. ,.'.

,5.50

-""'

. '

-22.75

5.50

' ',^b

i - 1 - 61

P
CY

; ry:

.
/rad.
../
/rad.

.007

-.185

88

...'-^002
.19

""-.21

APPENDIX D
TABLE X
CONTROL SURFACE DERIVATIVES
'-> : i -.-. .

Cruise

Derivative

Flight Condition
Climb

..

Landing

(RCS),

/rad;

.063

.0688

.063

CL

(Man),

/rad;

;401

.435

.39

CLj

(RCS)

/rad;

.464

.464

'.241
-.974

CLj

*E

- .-'A

CL
,
-i. sp
CM, (RCS)
E
CM< (Man)

/rad.

/rad.

-.246

-.269

/rad.

-1;587

CM

/rad;

-.074

-.074

-.0367

/rad:

^_

.0106

CY
(RCS)
. R

/rad.

.1t7

.117

CYj

/rad,

;273

.273

.273

(RCS),

-1.73

-.241
-1.53

^CM
sp

CN. (RCS),
*R

/rad.

-.047.

-.407

-.047

-GNj

(Ma'h).

/rad.

-.105

-.105

-.105

Cj|^(R&)

/rad.

.017

.014

..017

Cn

(Mart),

/rad,

.023

.015

.025

C
(RCS),
* *A
Cn
(Man),

/rad.

.086

.086

.083

.12

.12

.117

x a

(Man), ;

v7

* A

/rad.

89

APPENDIX E
AJRPLANE INERTIA AND GEOMETRY
Airplane dimensional and inertia! properties used in the equations of motion are listed in Table XI
below. All other data required is listed in, or can be derived from Table I.

TABLE XI
DHC-6 DIMENSIONAL AND INERTIAL PROPERTIES
Flight Conditions
Landing

Airplane Property

Cruise

Climb

lxx, kg-m 2
(slug - ft^)

22370
(16 500)

22370
(16500)

22370
(16 500)

kg-m 2
(slug -ft 2 )

36060
(26600)

36060
(26 600)

36 060
(26 600)

1 z, kg-m 2
(slug - ft2)

53960
(39 800)

53960
(39 800)

53960
(39 800)

lxz, kg - m2
(slug - ft2)

1630
(1200)

1630
(1200)

1630
(1200)

s,

m2
(ft2)

39
(420)

39
(420)

39
(420)

b,

m
(ft)

19.8
(65)

19.8
(65)

19.8
(65)

c,m

m :
(ft)

1.98
(6.5)

1.98
(6.5)

1.98
(6.5)

91

APPENDIX F
SIGN CONVENTIONS
Sign conventions for forces, moments, angular rates, and linear velocities are shown in the sketch
below. The positive sense of each variable is shown.

P,L

Or,N

Z.w

Sign conventions for control deflections are lis'ted below:

Control Surface

Positive Deflection

Elevators

trailing edge down

Ailerons (symmetrical)

trailing edge down

Spoilers (symmetrical)

trailing edge up

Rudder

trailing edge left

93

REFERENCES
1.

Hunsaker, J. C. and Wilson, E. B., "Report on Behavior of Aeroplanes in Gust

Turbulence," NACA TR-1 (MIT), dated October 1915.

2.

Phillips, W. H., Kraft, Christopher C., "Theoretical Study of Some Methods for increasing
the Smoothness of Flight Through Rough Air," NACA TN-2416, dated July 1951.

3.

NASA TN-4332, "An Approach to the Problem of Esimating Severe and Repeated Gust
Loads for Missile Operations," September 1958.

4.

ASD-TR-61-235, "Optimum Fatigue Spectra," April 1962.

5.

Ml L-A-8861 A(USAF), "Airplane Strength and Rigidity, Flight Loads," March 1971.

6.

NASA TM-X-2620, "Symposium on Vehicle Ride Quality," October 1972, pages 107,
108, 113, 193.

7.

MIL-F-8785B, "Flying Qualities of Piloted Airplanes/'August 1969.

8.

BuAer Flight Controls Systems Manual, Volume II, "Dynamics of the Airframe,"
September 1952.

9.

Fung, Y. C., "The Theory of Aeroelasticity", John Wiley and Sons, Inc., 1955.

10.

Bisplinghoff, R. L., Ashley, H., and Halfman, R. L., "Aeroelasticity," Addison-Wesley

Publishing Co., 1957.

11.

Foster, D. N., "Flow Around Wing Sections with High-Lift Devices," -Journal of Aircraft,
Vol.9, No. 3, March 1972.

12.

Barnes, C. S., "A Developed Theory of Spoilers on Airfoils," Aeronautical Research

Council, C. P. No. 887, 1966.

13.

Fischel, J., and Ivey, M. F., "Collection of Test Data for Lateral Control with Full-Span
Flaps," NACA TN-1404, 1948.

<V. GOVERNMENT PRINTING OFFICE: 1973-739-029/60

95

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

WASHINGTON, D.C.

2O546

OFFICIAL BUSINESS
PENALTY

FOR PRIVATE USE $3OO

SPECIAL FOURTH-CLASS
BOOK

RATE

POSTAGE AND FEES PAID

NATIONAL AERONAUTICS AND
SPACE ADMINISTRATION
131

POSTMASTER :

If Undeliverable (Section 158

Postal Manual) Do Not Return

"The aeronautical and space 'activities of the United States shall be

conducted so as to contribute . . . to the expansion of human knowledge of phenomena in the atmosphere and space. The Administration
shall provide for the widest practicable and appropriate dissemination
of information'concerning its activities and the results thereof."
NATIONAL AERONAUTICS AND SPACE ACT OF 1958

NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS

TECHNICAL REPORTS: Scientific and
technical information considered important,
complete, and a lasting contribution to existing
knowledge.

TECHNICAL TRANSLATIONS: Information

published in a foreign language considered
to merit NASA distribution in English.

TECHNICAL NOTES: Information less broad

in scope but nevertheless of importance as a
contribution to existing knowledge.

SPECIAL PUBLICATIONS: Information

derived from or of value to NASA activities.
Publications include final reports of major
projects, monographs, data compilations,
handbooks, sourcebooks, and special
bibliographies.

TECHNICAL MEMORANDUMS:
Information receiving limited distribution
because of preliminary data, security classification, or other reasons. Also includes conference
proceedings with either limited or unlimited
distribution.
CONTRACTOR REPORTS: Scientific and
technical information generated under a NASA
contract or grant and considered an important
contribution to existing knowledge.

TECHNOLOGY UTILIZATION
PUBLICATIONS: Information on technology
used by NASA that may be of particular
interest in commercial and other non-aerospace
applications. Publications include Tech Briefs,
Technology Utilization Reports and
Technology Surveys.

De/ai/s on f/ie availability of these publications may be obtained from:

SCIENTIFIC AND TECHNICAL INFORMATION OFFICE

NATIONAL

AERONAUTICS

AND

SPACE ADMINISTRATION

Washington, D.C. 20546