7 AD-A28 624 NEW FLIGHT CONTROL TECHNOLOGIES FOR FUTURE NAVAL /

I AIRCRAFT(U) NAVAL AIR DEVELOPMENT CENTER UARNINSTER PA

I ARIRCRAFT AND CREWd SYSTEMS TECHNOLOGY DIRECTORATE
p NCLSSIFIED N W KANIUKA SEP 82 NADC-82248-60 F/6 1/3 N
 • No

.4

UnIIIIs,
IIIII-

1.25 1.

MICROCOPY RESOLUTION
TEST CHART
NATIONAL BUREAU OF
STANDARDS-1963-A

* * *00
oSS
.
 REPORT NO. NADC-82240-60

NEW FLIGHT CONTROL TECHNOLOGIES

FOR FUTURE NAVAL AIRCRAFT
00
W.W. Kaniuka et al

Aircraft and Crew Systems Technology Directorate

* NAVAL AIR DEVELOPMENT CENTER
Warminster, Pennsylvania 18974

SEPTEMBER 1982

Interim Report

Sub Project No. WF 41-423

SAPPROVED FOR PUBLIC RELZ'AE. DISTRIBUTION UNLIMITED

LUL
for .Prepared
' -- NAVAL AIR SYSTEMS COMMAND
Department of the N jvv
Washington, D.C. 20361

83 05 ZG 016
- - - - - - - - - - - -
 N 0 T ICES

- REPORT NUMBERING SYSTEM - The numbering of tecnnical prclect recort-sssued by 'he

Naval Air Deveiopment Center 's arranged for specific identification purposes. Each number con-
' sists of the Center acronym, the calendar year in which the number was 3ssigned, -re sequence
number of the repcrt within the specific calendar year, and the o.ficial 2 digit correspondence
code of the Command Office or the FunctjonaI Directorate responsible for the report. For exam-
* -" pie: Report No. NADC-78015-20 indicates the fifteenth Center report for the year 1973, and
prepared by the Systems Directorate. The numerical codes are as follows:

CODE OFFICE OR DIRECTORATE

00 Commander, Navai Air Deveiooment Center

01 Technical Director, Naval Air Development Center
02 Comptroller
10 Directorate Command Projects
20 Systems Directorate
30 Sensors & Avionics Technology Dirertcrate
40 Communication & Navigation Technofogy Dir,-torate
50 Software Computer Directorate
60 Aircraft & Crew Systems Technology Directorate
70 Planning Assessment Resources
80 Engineering Support Group

PRODUCT EN DO RSEMENT - The discussion or instructions concerning commercial products

*herein do not constitute an endorsement by the Government nor do they convey or imply the
license or right to use such products.

4 ,Ic
APPROVED3BY K SN
'' DATE: /x________

a.

a
 UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)
REPORT DOCUMENTATION PAGE READ INSTRUCTIONS
BEFORE COMPLETING FORM
1. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER
•NADC-82240-60 T
4. TITLE (and Subtitle) S. TYPE OF REPORT & PERIOD COVERED

New Flight Control Technologies for Future Naval Aircraft Interim Report
6. PERFORMING ORG. REPORT NUMBER

7. AUTHOR(s) 8. CONTRACT OR GRANT NUMBER(s)

W. W. Kaniuka et al

9. PERFORMING ORGANIZATION NAME AND ADDRESS SO. PROGRAM ELEMENT. PROJECT, TASK
AREA & WORK UNIT NUMBERS
Aircraft and Crew Systems Technology Directorate (Code 60)
NAVAL AIR DEVELOPMENT CENTER Subproject No.
Warminster, PA 18974 WF41-423
II. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE
September 1982
Naval Air Development Center
13. NUMBER OF PAGES
Warminster, PA 18974
183
14. MONITORING AGENCY NAME &ADDRESS(if different from Controlling Office) IS. SECURITY CLASS. (of this report)

UNCLASSIFIED
ISa. DECLASSIFICATION, DOWNGRADING
SCHEDULE

16. DISTRIBUTION STATEMENT (of this Report)

Approved for Public Release; Distribution Unlimited.

17. DISTRIBUTION STATEMENT (of the abstract entered in Block 20, If different from Report)

I8. SUPPLEMENTARY NOTES

19. KEY WORDS (Continue on reverse side if necessary nd identify by block number)

Fly-By-Wire Fiber Optics

Digital Flight Control Systems Integrated Sensory Subsystems (ISS)
AFTI/F-16 EMI
. Flight Control Laws EMP
Flight Control Actuators
.. ABSTRACT (Continue on reverse side If necessary and identify by block number)

'This report describes new flight control technologies that were developed in part or fully under Navy R&D
funding for future advanced aircraft. The main thrust of these developmental programs was in the areas of
digital flight control systems, advanced control surface actuators, and integrated sensory subsystems.

DD JAIN72! 1473 EDITION OF I NOV 65 IS OBSOLETE

, ,
UNCLASSIFIED
SECURITY CL'ASSIFICATION OF 'IS PAGE (*%on Data Entered)
SECURITY CLASSIFICATION OF THIS PAGE (Wh~enDate Entered)

U SECURITY CLASSIFICATION 0F-rIS PAGE(Wion Date Entered)

 --

.7- NADC-82240-60

SUMMARY

High performance combat aircraft require Flight Control Systems (FCS) that permit the attainment
of mission objectives within the constraints of cost, weight, volume, reliability and maintainability.
The FCS must also be relatively invulnerable to enemy small projectile and radiation weaponry and
natural hazards such as lightning.

The complexity and variety of tasks that FCS's are called upon to perform has been steadily in-
creasing. However, the application of new technologies such as fly-by-wire, digital computation,
and integrated systems will permit the design and synthesis of FCS's capable of meeting these highly
demanding requirements.

This report presents the new technologies and concepts whose development has been advanced by
.Navy supported Exploratory and Advanced Development Programs.

-Aocession For

NTIS GRA&I
DTIC TAB
"'-'iU mnr
lno ilnced I]
J-,, tificat io.

Ditr"b-rtion/
Av_ iity Codes
Avail and/or
Dist Special

83 .
 NADC-82240-60

TABLE OF CONTENTS

Page

SU M M A R Y ............................................................... i/ii

TA BLE O F CONTENTS ..................................................... iii

LIST O F FIG URES ......................................................... viii

LIST O F TA BLES .......................................................... xiv

LIST OF ABBREVIATIONS, ACRONYMS, and SYMBOLS .......................... xv

INT RO DUCT IO N .......................................................... 1

DESCRIPTION OF THE INTEGRATED SENSORY SUBSYSTEM (ISS) CONCEPT ........ 3

ISS SENSOR SYSTEM CONFIGURATIONS .................................. 4
1. ISS Air Data System Sensors, Data Handling System,
and Redundancy Data Management System ............................... 4
2. ISS Inertial Data System Sensors, Data Handling System,
and Redundancy Data Management System ............................... 13
3. ISS/SAHRS Function for Sensing Attitudes, Heading, and
Inertial Velocity Components .......................................... 24
ISS REDUNDANT INPUT/OUTPUT (I/O) SYSTEM ............................. 28
ISS COMPUTATION NETWORKS WITHIN REDUNDANT
DIGITAL COMPUTER COMPLEXES ......................................... 33

THE ADVANCED FIGHTER TECHNOLOGY INTEGRATOR (AFTI/F-16)

DEVELOPMENT AND INTEGRATION PROGRAM ............................... 34
DESCRIPTION OF THE AFTI/F-16 DIGITAL FLIGHT CONTROL SYSTEM
(D FC S) ................................................................. 36
1. Primary Flight Control System (PFCS) .................................. 42
2. Secondary Flight Control System (SFCS) ................................ 45
AFTI/F-16 AIR DATA SYSTEM (ADS) ....................................... 46
1. A ir Data Probes .................................................... 47
2. Static Pressure Ports ................................................ 48
3. Sideslip Differential Pressure Sensor .................................... 48
4. Angle of Attack (AOA) Transmitters .................................... 48
4 5. Angle of Sideslip (AOS) Transmitters ................................... 48
6. Pneumatic Sensor Assembly (PSA) ..................................... 48
7. Central Air Data Computer (CADC) .................................... 49
8. Air Data System Redundancy Management ............................... 49
I1

. ii-
 NADC-82240-60

TABLE OF CONTENTS (Continued)

Page

THE ADVANCED FIGHTER TECHNOLOGY INTEGRATOR (AFTI/F-16)

DEVELOPMENT AND INTEGRATION PROGRAM (Continued)
FLIGHT CONTROL COCKPIT CONTROLS ................................... 50
1. Side-Stick Controller (SSC) Assembly .................................. 50
2. Rudder Pedal Assem bly .............................................. 53
3. Flight Control Panel ................................................. 53
4. Rotatable Throttle Controller Grip ..................................... 55
5. Interface With the Flight Control Computers .............................. 56
FLIGHT CONTROL INERTIAL SENSORS .................................... 57
1. Rate Gyro Assem blies ............................................... 57
2. Accelerometer Assembly .............................................. 58
3. Interface With the Flight Control Computers .............................. 58
4. Redundancy Management and Reconfiguration Techniques .................. 58
DIGITAL FLIGHT CONTROL COMPUTER (FCC) COMPLEX ..................... 61
1. Interface Between FCC Complex and the FCS ............................. 61
2. The Flight Control Computer (FCC) .................................... 62
3. The Actuator Interface Unit (AIU) ..................................... 66
FLIGHT CONTROL ACTUATORS .......................................... 69
1. Integrated Servoactuators (ISA) ......................................... 69
2. Power Drive Unit (PDU) Assembly ..................................... 72
REDUNDANCY MANAGEMENT OF THE DFCS ............................... 73
1. Input Voting/Monitoring Plane ......................................... 75
2. Output Voting/Monitoring Plane ....................................... 76
a. Failing Detection, Isolation, and Reconfiguration
After Single and Dual FCC Failures ................................. 77
3, Output ISA Voting/Monitoring Plane .................................... 77
FLIGHT CONTROL SYSTEM SELF-TEST .................................... 78
1. Inflight Integrity Management (IFIM) ................................... 78
2. Inflight Self-Test ................................................... 79
3. Preflight BIT ...................................................... 79
4. Maintenance BIT ................................................... 79
INDEPENDENT BACKUP UNIT (IBU) ........................................ 79
DFCS-ELECTRICAL SYSTEM INTERFACE ................................... 81
DFCS-HYDRAULIC SYSTEM INTERFACE ................................... 84
DFCS-AVIONICS SYSTEMS INTERFACES ................................... 84
PILOT'VEHICLE INTERFACE (PVI) ......................................... 88
1. Cockpit Controllers ................................................. 88
2. Control/Display System (CDS) ......................................... 88

iv
 NADC-82240-60

TABLE OF CONTENTS (Continued)

Page

THE ADVANCED FIGHTER TECHNOLOGY INTEGRATOR (AFTI/F-16)

DEVELOPMENT AND INTEGRATION PROGRAM (Continued)
THE ADVANCED FLIGHT CONTROL MODES OF THE AFTI/F-16 ................ 91
1. The Four Major Flight Control Modes ................................... 91
2. The Control Configured Vehicle (CCV) Modes ............................. 91
3. Control Law Design Methodology ...................................... 95
4. Decoupled Control Modes ............................................ 97
a. Longitudinal Decoupling ......................................... 104
b. Lateral-Directional Decoupling ..................................... 104

THE ADVANCED FLIGHT CONTROL ACTUATION SYSTEM (AFCAS)

DEVELOPMENT PROGRAM ................................................. 111
DESIGN PHILOSOPHY OF THE AFCAS CONCEPT ............................. 112
DESCRIPTION OF THE AFCAS CONCEPT .................................... 113
SERVO VALVE DESIGN REQUIREMENTS ................................... 115
AFCAS CONFIGURATIONS TESTED ........................................ 116
FLIGHT VERIFICATION OF THE AFCAS CONCEPT IN THE T-2C AIRCRAFT.......122
1. Technical Approach ................................................. 122
2. T-2C A irplane ..................................................... 125
3. AFCAS Flight Test Installation ........................................ 126
a. Mechanical System ............................................. 126
b. Hydraulic System ............................................... 128
c. Electrical System ............................................... 128
d. Instrum entation ................................................ 136
4. Flight Plan ........................................................ 136
5. Flight Test Results .................................................. 136
6. Sum m ary ........................................................ 137
FLIGHT VERIFICATION OF DIRECT DIGITAL DRIVE FOR AN AFCAS IN
THE T-2C A IRCRAFT .................................................... 137
1. Technical Approach ................................................. 137
3
2. Description of the AFCAS Direct Digital Drive (D )
Test Installation ..................................................... 138
a. Microcomputer Assembly ........................................ 140
b. Software Description ............................................ 140
c. Program Modules ............................................... 141
d. Flight Test Program Software ...................................... 142
e. Microcomputer Analog and Reduced Bit Resolution Program Software ...... 143
f. Laboratory Tests ............................................... 143

I4

v
" .-" . .. • . .. . -" . . . .: • '. . . . '" . _ , , . L
 NADC-82240-60

TABLE OF CONTENTS (Continued)

Page

3. Flight Test Program ................................................. 145

a. Flight Plan .................................................... 145
b. Flight Test Results .............................................. 145
c. Flight Program Sum mary ......................................... 145
ROCKWELL'S STUDIES OF PULSE MODULATED DIRECT DIGITAL
DRIV E CO NCEPTS ....................................................... 145
FLIGHT CONTROL COMPUTER COMPLEX-D 3 ACTUATOR INTERFACE .......... 147
1. Pulse Modulation Command ........................................... 147
2. D3 Redundancy Concept ............................................. 151
3. Fault Tolerant Redundancy .......................................... 151

THE HYDRA-OPTIC FLIGHT CONTROL ACTUATION SYSTEM (HOFCAS)

DEVELOPMENT PROGRAM ................................................. 154
TECHNICAL APPROACH ................................................. 154
DESCRIPTION OF THE HOFCAS ........................................... 155
1. Hydra-Powered Alternator (HPA) ...................................... 158
2. Electronic Drive Unit (EDU) .......................................... 159
3. Microcomputer Assembly and Power Supply .............................. 159
LABORATORY AND INTEGRATION TESTS .................................. 161
ANALYSIS OF TEST RESULTS ............................................. 162
1. Hydra-Powered Alternator (HPA) ...... ............................... 162
2. Fiber Optic Link ................................................... 163
3. Flight Test Configuration ............................................. 163
4. HO FCAS Performance ............................................... 163
RECOM MENDATIONS .................................................... 164
INTERFACE OF FLIGHT CONTROL COMPUTER COMPLEX,
D3 ACTUATION, AND HOFCAS ............................................ 165
FAULT-TOLERANT ACTUATION CONCEPT FOR A RESEARCH
TEST A IRC RA FT .......................................................... 169
BACKGROUND INFORMATION ............................................ 169
FLIGHT CONTROL ACTUATOR REQUIREMENTS ............................ 169
INTERFACE OF THE PRIMARY AND AFCS ACTUATORS ....................... 170
THE BELL HELICOPTER TEXTRON FOUR-VALVE, DIRECT DRIVE,
ACTUATION CONCEPT .................................................. 173
1. Summarized Description of Concept .................................... 173
2. FBW Control Paths .................................................. 173

vi
 NADC-82240-60

TABLE OF CONTENTS (Continued)

Page

3. Failure Management System .......................................... 175

*a. Failure Sensing Function ......................................... 175
b, Automatic Detection /Disengage Function ............................ 176
LABORATORY TESTS AND DEMONSTRATIONS .............................. 176

REFERENCES ............................................................ 179

ACKNOWLEDGEMENTS ................................................... 181

a vii
 NADC-8224060

LIST OF FIGURES

Figure Title Page

1 Schematic and Signal Flow Diagram of the ISS Concept ..................... 5

2 Schematic of the ISS Components and Associated Signal Flow Paths ............. 6

3 Multipurpose Air Data Probe .......................................... 7

4 Pressure Differential with c ........................................... 7

5 Computational Signal Flow in a Typical Analog Central Air Data

Com puter (CA DC) ................................... . ............ 9

6 Outline of Air Data Redundancy System (RDMS) Functions ........... 10

7 Block Diagram of the Data Handling System (DHS) ...................... 12

8 Block Diagram of the Redundancy Data Management System (RDMS) .......... 12

9 Rate Gyro Configurations for Dual Fail-Operational Capability ................ 15

10 Geometry of the Rate Gyro Sensing Axes for the ASSET Concept ............. 16

11 Geometry of a Conical Array of Six Rate Gyros ........................... 17

12 Geometry of a Concial Array of Six Linear Accelerometers ................... 18

13 ISS Inertial Sensor Data Handling System (DHS) ........................... 20

14 Rate Gyro Redundancy Data Management System (RDMS) ................... 22

15 Linear Accelerometer Redundancy Data Management System (RDMS) .......... 25

16 ISS/SAH RS Signal Flow Block Diagram .................................. 27

17 ISS Input/Output System for a Digital Flight Control System (DFCS) .......... 29
18 Air Data System/Flight Control Computer Complex Interface ................ 30

19 Interface of the Inertial Component Assemblies with the Flight Control

Computers Via Cross-Strapping ........................................ 31

20 Interface of the Inertial Component Assemblies with the Flight Control

Computers Via Inter Computer High Speed Data Link ....................... 32

21 Interface of the Inertial Component Assemblies 6nd ICA /p's with the
Flight Control Com puters ............................................ 33
* 22 Objectives of the AFTI/F-16 Program ................................... 35

viii
 NADC-82240-60

LIST OF FIGURES (Continued)

Figure Title Page

, 23 Digital Fly-By-Wire Flight Control System for the AFTI/F-16 ................. 37

24 Benefits of a Fly-By-Wire FCS with Digital Processing ....................... 38

25 Functional Schematic of the AFTI/F-16 DFCS ............................ 39

26 Air Data System Interface with the DFCS ................................ 44

27 Mechanization of the Speedbrake ...................................... 47

28 Installation and Assembly of the Side Stick Controller (SSC) for the F-16 ........ 50

29 AFTI/F-16 Side-Stick Controller (SSC) .................................. 51

30 Cross-Sectional View of Transducer Portion of the Side-Stick Controller ........ 52

31 Rudder Pedal Assem bly .............................................. 53

32 Flight Control Panel Installation ....................................... 54

33 Throttle Assem bly .................................................. 56

34 Installation of the Rate Gyro Assembjies ................................. 57

35 Installation of Accelerometer Assembly ................................. 59

36 Interface of the Flight Control Inertial Sensors with the Flight

Control Com puters ................................................. 60

37 Interface of the FCC Complex with the Dualized Avionics Multiplex Bus ....... 62

38 Layout/Outline Drawing of the Bendix BDX-930 FCC ....................... 63

39 Layout/Outline Drawing of the Actuator Interface Unit (AIU) ................. 67

40 Interface of the Actuator Interface Unit (AIU) within the DFCS .............. 68

41 Typical ISA Used for Powering the Primary and Vertical Canard Control
Surfaces . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

42 Schematic Diagram of the ISA ......................................... 70

43 Interface of an ISA with FCC Complex ................................. 72

44 Power Drive Unit (PDU) Assembly ..................................... 73

i

"i ix
 NADC-82240-60

LIST OF FIGURES (Continued)

Figure Title Page

45 Redundancy Data Management of the Hardware and Software

Functions of the AFTI/F-16 DFCS ..................................... 74

46 Block Diagram of the DFCS Indpendent Backup Unit (IBU) ................ 80

47 AFTI/F-16 Redundancy Concept ...................................... 82

48 AFTI/F-16 DFCS Electrical Power System ............................... 83

49 Hydraulic System Functional Block Diagram ............................. 85

50 DFCS and Interface with Avionics ...................................... 86

51 AFTI/F-16 Avionic Components ...................................... 87

52 AFTI/F-16 Avionics/DFCS Integration Technology Payoffs ................. 87

53 AFTI/F-16 Crew Station Layout ....................................... 89

54 Flight Control Options Displayed On MPD ............................... 90

55 Control Surfaces Used for Decoupled Flight Path Control .................... 92

56 The Normal Mode ........................... ....................... 93

57 Air-to-Air Gunnery (AAG) Mode ....................................... 93

58 Air-to-Surface Gunnery (ASG) Mode .................................... 94

59 Air-to-Surface Bombing (ASB) Mode .................................... 94

60 Pitch DFCS Multimode Control ........................................ 96

61 Lateral-Directional DFCS Multimode Control ............................. 96

62 Simplified Longitudinal Standard Normal Mode Block Diagram ................ 98

63 Simplified Lateral-Directional Standard Normal Mode Block Diagram ........... 98

64 Simplified Air Combat Gunnery Mode ................................... 100

65 Simplified Air-to-Surface Bombing Mode ................................. 101

66 Direct Lift (A n) Control Mode ......................................... 102

4 67 Direct Sideforce (Ay) Control Mode ..................................... 102

x
 NADC-82240-60

LIST OF FIGURES (Continued)

Figure Title Page

68 Pitch Pointing (a1) Control Mode ..................................... 102

69 Yaw Pointing (i1) Control Mode ...................................... 103

70 Vertical Translation (a2) Control Mode ................................ 103

71 Lateral Translation (f02) Control Mode ................................. 103

72 Effect of Lateral Translation Capability On Landing ....................... 105

73 Effect of Flat Turn Capability ........................................ 106

74 Effect of Fuselage Aiming On Air-to-Air Combat ......................... 107

75 Air Combat Pitch Pointing Mode ...................................... 108

76 Air-to-Surface Flat Turn Bombing Mode ................................. 109

77 Direct Drive Control Valve Package .................................... 114

78 Single System Artuator ............................................. 117

79 Dual System Parallel Actuator ........................................ 118

80 Dual System Tandem Actuator ....................................... 119

81 Force Motor ..................................................... 120

82 Block Diagram of Electronic Interface Unit and Dualized Control-By-Wire

A ctuator ......................................................... 12 1

83 Rudder Actuator Assembly .......................................... 123

84 T-2C Demonstrator Aircraft .......................................... 124

85 Schematic Diagram of Mechanical System ............................... 127

86 Original and Modified Hydraulic Systems ............................... 129

87 Schematic Diagram of Modified Hydraulic System ......................... 130

88 Block Diagram of System ............................................ 132

89 Simplified Diagram of Electrical Components ............................ 133

" 90 Simplified Diagram Showing System Redundancy ......................... 134

xi
i NADC-82240-60

LIST OF FIGURES (Continued)

Figure Title Page

91 Mathematical Model of the AFCAS .................................... 135

92 Simplified Block Diagram of T-2C Direct Digital Drive (D3 )

Test Installation ................................................... 139

93 Digital Fly-By-Wire Microcomputer Program Flow Chart .................... 141

94 Laboratory Test Setup Block Diagram .................................. 144

95 Digital Drive Unit Concept .......................................... 148

96 Digital Drive Concept Waveforms ..................................... 148

97 Closed Loop Response (Uncompensated) ................................ 149

98 Effect of Compensation Method on Closed Loop Response .................. 149

99 Pulse Modulated Waveform .......................................... 152

100 D3 Redundancy Concept ............................................ 152

101 HOFCAS Simplified Block Diagram .................................... 156

102 HOFCAS Operational Modes, Simplified Block Diagram .................... 157

103 HPA Turbine Wheel and Alternator .................................... 160

104 HPA Assem bly .................................................... 160

105 EDU With Fiber Optic Cables Attached ................................. 160

106 Microcomputer Unit With Fiber Optic Cables Attached ..................... 161

107 Torque Motor Characteristics ......................................... 164

108 AFCAS Frequency Response ......................................... 166

109 Digital Fly-By-Light Mode System Response, Maximum Capability,

HPA Power (1000 Hz and Aircraft Power (400 Hz) ....................... 166

110 Frequency Response, AFCAS Designed for D3 . . . . . . .. . . . . . . . . . . . . . . .. . . .

167

11l Analog Back-Up Mode System Response, Flight Gain Configuration,

A ircraft Pow er (400 Hz) ............................................. 167
112 Rotary Valve D3 Integrated Dual-Tandem Actuator ........................ 168

xii
 NADC-82240-60

LIST OF FIGURES (Continued)

Figure Title Page

113 HOFCAS System Redundancy Concept ................................. 168

114 Output Summing Attached Actuators Configuration........................171

115 Input Summing Attached Actuators Configuration ........................ 172

116 4-Valve FBW Actuation Concept ...................................... 174

117 Fault Sensing Concept .............................................. 177

ixi.

a
6 NADC-82240-60

LIST OF TABLES

Table Title Page

1 Air Data Parameter Requirements Summary .............................. 13

2 Adjustments Required to Account for Rate Gyro and Linear
Accelerom eter Dispersion ............................................ 19

3 Capabilities and Features Common to Both the Current

F-16 and the A FTI/F-16 ............................................ 41

4 Capabilities and Features Unique to the AFTI/F-16 ........................ 41

5 F-16 Flight Control System Functional Characteristics ...................... 43

6 Salient Features of the Bendix BDX-930 FCC ............................. 65

7 AFTI/F-16 DFCS Decoupled Longitudinal Control Modes ................... 99

8 AFTI/F-16 DFCS Decoupled Lateral-Directional Control Modes .............. 99

9 Electromechanical Force Motor Data ................................... 150

xiv
 NADC-82240-60

LIST OF ABBREVIATIONS, ACRONYMS, AND SYMBOLS

AAG Air-to-Air Gunnery

ABU Analog Back Up
AC Alternating Current
ACIA Asynchronous Interface Adapter
A/D Analog to Digital
ADS Air Data System
AFCAS Advanced Flight Control Actuation System
AFCS Automatic Flight Control System
AFTI/F-16 Advanced Fighter Technology Integrator/F-16
AIU Actuator Interface Unit
AN Normal Acceleration
7 AOA Angle of Attack
AOS Angle of Sideslip
ARI Aileron- Rudder-Interconnect
ASB Air-to-Surface Bombing
ASG Air-to-Surface Gunnery
ASSET Advanced Skewed Sensory Electronic Triad
AX Acceleration along the "X" Body Axis
Ay Acceleration along the "Y" Body Axis
AZ Acceleration along the "Z" Body Axis
BHT Bell Helicopter Textron
BIT Built-In-Test
OC Degrees Celsius
CAD Columbus Aircraft Division
CADC Central Air Data Computer
CAS Command Augmentation System
CCCM Cross Channel Comparison Monitoring
CCSL Coordinated Control-Surface Limiter
CCV Control Configured Vehicle
CDS Control Display System
C.G. Center of Gravity
ClU Control Interface Unit
cm Centimeter
CM Comparison Monitoring
CPU Central Processor Unit
CVCM Cross Voter Comparison Monitoring

xv
 NADC-82240-60

LIST OF ABBREVIATIONS, ACRONYMS, AND SYMBOLS (Continued)

D/A Digital to Analog

DAC Digital to Analog Converter
DC Direct Current
D3 Direct Digital Drive
deg Degree
DFBL Digital Fly-By-Light
DFCS Digital Flight Control System
SDHS Data Handling System
DLC Direct Lift Control
DOF Degree-of-Freedom
DP Digital Processor
DSF Direct Side Force
ECA Electronic Component Assembly
EDU Electronic Drive Unit
EHSV Electro-Hydraulic Servo Valve
EMI Electro-Magnetic Interference
EMP Electro-Magnetic Pulse
EPROM Erasable Programmable Read Only Memory
EPU Emergency Power Unit
0F Degrees Fahrenheit
FBW Fly-By-Wire
FCC Flight Control Computer
FCS Flight Control System
FICR Failure Isolation Computation Routine
FMDS Failure Management and Display System
F/O Fail-Operational
FTL Fault Tolerance Level
g Gravitational Constant
GAC Grumman Aerospace Corporation
GD General Dynamics
gm Gram
gpm Gallons per Minute
GSE Ground Support Equipment
H.M. Hinge Moment
HMS Helmet Mounted Sight
HOFCAS Hydra-Optic Flight Control Actuation System
- HPA Hydra-Powered Alternator

I xvi
 NADC-82240-60

LIST OF ABBREVIATIONS, ACRONYMS, AND SYMBOLS (Continued)

Hp Pressure Altitude Referenced to 29.92 in. Hg

Hpbc Pressure Altitude Referenced to Altimeter Barometric Setting
A:HP Pressure Altitude Rate
Hr Hour
H.T. Horizontal Tail
HUD Head-Up Display
Hz Hertz (cycles per second)
IBU Independent Back Up Unit
ICA Inertial Component Assembly
IDC Input Data Converter
IFBIT In-Flight Built-In Test
- , IFFC Integrated Fire and Flight Control
I FIM In-Flight Integrity Management
INS Inertial Navigation System
I/O Input/Output
IOC Input/Output Controller
L ISA Integrated Servoactuator
ISS Integrated Sensory Subsystem
k Kilo (103)
kg Kilogram
km Kilometer
kVA Kilo Volt Ampere
LCC Life Cycle Costs
L/D Lift to Drag Ratio
LEF Leading Edge Flap
LHS Lightweight Hydraulic System
L/m Liters per Minute
LQS Linear Quadratic Synthesis
LRU Line Replaceable Unit
LSP Low Speed Brake
LVDT Linear Vauable Differential Transformer
M Mach No.
. MHD Magnetic Heading Detector
MPa Megapascals
MPD Multi-Purpose Display
MPU Microprocessor Unit
im/s Meters per Second
msec Millisecond
MUX Multiplex Bus

U xvii
 NADC-82240-60

LIST OF ABBREVIATIONS, ACRONYMS, AND SYMBOLS (Continued)

N Newton (metric unit of force)

NAAD North American Aircraft Division
NADC Naval Air Development Center
ODC Output Data Processor
OFP Operational Flight Plan
p Roll Rate
POG Programmable Display Generator
PDU Power Drive Unit
PFCS Primary Flight Control System
PIA Peripheral Interface Adapter
PMG Permanent Magnet Generator
PROM Programmable Read Only Memory
PS Static Pressure
PSA Pneumatic Sensor Assembly
psi Pounds per Square Inch
PT Pitot Pressure
PVI Pilot/Vehicle Interface
PWM Pulse Width Modulation
pAOA
PO1 Sensing Port on Air Data Probe
P0 AOS Sensing Port on Air Data Probe
q Pitch Rate
qc Compressible Dynamic Pressure
r Yaw Rate or Distance from Accelerometers to Virtual NAV-Base
RAM Random Access Memory
RDMS Redundancy Data Management System
ROM Read Only Memory
RPM Revolutions per Minute
RSS Relaxed Static Stability
RVDT Rotary Voltage Differential Transformer
s Laplace Operator
SAHRS Strapdown Attitude and Heading Reference System
SAS Stability Augmentation System
SDOF Single Degree of Freedom
sec Second
SFCS Secondary Flight Control System
SMS Stores Management Set

xviii
 NADC-82240-60

LIST OF ABBREVIATIONS, ACRONYMS, AND SYMBOLS (Continued)

SSC Side-Stick Controller

SV Servo Valve
SVCR Sensor Voting Computational Routine
SVSR Sensor Voting Selection Routine
TEF Trailing Edge Flap
TFRR Transient Failure Removal Routine
TFAT True Air Temperature ('F)
TM Telemetry
TT Total Temperature
VA Volt Ampere
VC Calibrated Airspeed
VDC Volts Direct Current
VT True Airspeed
VTOL Vertical Take-Off Or Landing

Greek Symbols
U Angle-of-Attack
axq Gyro Orientation Angle
a1 Pitch Pointing Control Mode
a2 Vertical Translation Control Mode
(Angle-of-Sideslip
I3 Angle-of-Sideslip Rate
Oa Linear Accelerometer Central Half-angle
Oq Gyro Central Half-angle
01 Yaw Pointing Control Mode
32 Lateral Translation Control Mode
6 Deflection Angle
A Differential
0 Pitch Rate
11- -Microprocessor
a Air Density Ratio or One Standard Deviation
*Roll Attitude
(Roll Rate
(ba Linear Accelerometer Orientation Angle
'Heading Attitude
4Angular Rate

xix
 NADC-82240-60

INTRODUCTION

The Naval Air Development Center (NADC) has been and is currently sponsoring developmental
programs to develop and/or apply advanced technologies applicable to the Flight Control Systems
(FCS) of future tactical aircraft. The main thrust of these developmental programs is toward the
development of an advanced type of FCS - a next generation Digital Flight Control System
(DFCS) employing Fly-By-Wire (FBW) signal paths. The major goals of the developmental programs
are as follows:

* Improvement of mission performance through flight control technology.

e Increased flight safety and mission reliability via redundancy data management.
,.. Reduced vulnerability to combat damage.
. Simplicity and reliability through standardized and interchangeable components.
* Improved maintainability through Built-In-Test (BIT) and diagnostics.
a Reduced weight and volume requirements.
* Immunity to electro-magnetic hazards/threats.

The flight control developmental programs sponsored by NADC and their associated contractors
(as of report date) are listed as follows:

* Development of an Integrated Sensory Subsystem (155) - The Grumman Aerospace Corp.

e Development of a Digital Flight Control System for the Advanced Fighter Technology
Integration AFTI/F-16 Aircraft - General Dynamics Corp., Fort Worth Division (A joint
Navy, Air Force, and NASA program).
e Development of Two Advanced and Fault Tolerant Actuation Concepts for FBW Systems.
* Direct Digital Drive (D3 ) High Pressure (8000 psi) Actuation Concept - Rockwell
International Corp., Columbus Division.
' Four-Valve Tandem FBW Actuator - Bell Helicopter Textron, Fort Worth.
- Development of the Hydra-Optic Flight Control Actuation System (HOFCAS) Concept -
Rockwell International Corp., Columbus Division.

The ISS exploratory development program is now in its final phase and should be available for the
advanced development phase in the near future.
The first phase of the AFTI/F-16 program, dedicated to the development of a triplex DFCS is
still in progress and is scheduled to be completed in CY 1983. The first flight of the AFTI/F-16
took place in July 1982 and is now in its initial flight test program at the Edwards Air Force Base.
The six phases of the D 3 program have been completed with the actuation concept being success-
fully test flown in a T-2C aircraft. A follow on program now in progress, is directed to the develop-
ment of a compact FBW actuator with rotary type hydraulic servo valves, their electronic packages,
and feedback paths imbedded within the cylindrical portion of the actuator body.

L4

4
 NADC-82240-60

The first phase of the Four-Valve Tandem FBW Actuator program has been completed in July
1980. The concept has been successfully demonstrated via laboratory testing. The second phase
of the program was directed toward the development of a FBW actuator with manual revision
capability and was completed in November 1981.

The development of a laboratory version of the HOFCAS was completed in December 1980.
This report was prepared for the purpose of providing simplified descriptions of the concepts being
developed. A follow-on report will be issued to document future flight control developmental
programs and update existing programs.

,- -

f2
 NADC-82240-60

DESCRIPTION OF THE INTEGRATED SENSORY SUBSYSTEM (ISS) CONCEPT

The Naval Air Development Center defii,.d the requirement for an Integrated Sensory Subsystem
- . (ISS) after observing the numerous sensory subsystems employed on their F-14 aircraft. Several
subsystems on the aircraft had dedicated inertial and air data type sensors with the result being an
inefficient and wasteful proliferation of sensors. Under contracts to the Naval Air Development
Center (NADC) GAC evolved and developed the ISS concept which permits an optimal integration
of sensors with redundant digital computers or microprocessors so that the outputs of a basic set
of sensors can be shared and used by such aircraft subsystems as the following:

o Digital Flight Control System (DFCS).

o Cockpit instrumentation and displays.
o Engine inlet control.
o Engine thrust control.
o Wing sweep control.
o Inertial Navigation System (INS).
o Weapon delivery.

In addition, in order to meet very stringent reliability and survivability requirements future high
performance combat aircraft will incorporate redundant systems and components, further compli-
cating the sensor requirements.

A fully developed ISS will provide the following benefits:

o Increased flight safety and mission reliability for aircraft incorporated with Digital Fly-By-Wire
(DFBW) systems and Relaxed Static Stability (RSS) through the use of redundant sensors and
"* redundancy management.
* Decreased vulnerability to combat damage by the dispersal of components.
- Reduced parts count and inventory requirements.
- Reduced maintenance.
o Reduction of weight and volume.
e Increased operational readiness.

An ISS is composed of the following three main elements:

4 (1) A functionally reliable and combat damage survivable sensor set consisting of the following:

o Redundant low and high speed multi-purpose air data probes.

o Six hard mounted, skewed, and dispersed rate integrating gyros.
o Six hard mounted, skewed, and dispersed linear accelerometers.

-I
6I NADC-82240-60

(2) An Input/Output (I/O) System that links the data from the redundant air and inertial data
sensors to a redundant digital and/or distributed microprocessor complex for processing.
The signal paths of the I/O system may be in the form of hard wiring, multiplex bus, or fiber
optics, or any combination thereof.

(3) A computation network within a redundant digital computer complex consisting of sub-
routines that provide:

e Sensor data preprocessing to compensate for position errors and filtering of sensor outputs.
e Sensor redundancy data management for sensor signal selection, failure detection, and
failure isolation.
* State estimation and sensor data normalization algorithms to account for deterministic
errors associated with body bending coupling, dispersion of sensors, and random sensor
noise.
a Strapdown attitude and heading reference computations for the skewed and dispersed
inertial sensors.
* Air data computations for altitude, altitude rate, true airspeed, true Mach number, dynamic
pressure, angle of attack (a ), angle of sideslip (03), true static pressure, and total air
temperature.

The calculated parameters resident in the computer memories are transmitted to the using subsys-
tems on a time shared basis. Figure 1 is a Schematic and Signal Flow Diagram of the ISS Concept.
The vectors shown in a conical arrangement represent the sensing axes of the six skewed rate gyros
and the six skewed linear accelerometers. Inputs to the redundant flight control digital computer
complex that are from systems not a part of the ISS are marked with an asterisk. The required com-
putations for the ISS Concept, as illustrated in Figure 1, are performed in the redundant flight con-
trol digital computer complex on a time shared basis. All of the inputs to the computer complex
are used in the flight control and navigation tasks. However, the inputs from the INS and the flux
valves are also used to aid in the redundancy management of the ISS air data system.

Figure 2 is a schematic of the major components of the ISS and their associated signal flow paths.
Each of the three Inertial Sensor Assemblies contain two skewed linear accelerometers and two
skewed gyroscopes. The three multipurpose air data probes are able to sense pilot pressure (PT),
static pressure (PS), and angle-of-attack ( a ). The nose mounted air data probe is also able to sense
angle-of-sideslip (1). An additional method of sensing 1 is through sensing the difference in PS on
both sides of the fuselage. The two total temperature (TT) probes, one on each side of the fuselage,
are dualized to provide two fail-operational performance for this important air data parameter.

The ISS Inertial and Air Data Systems (ADS) provide two fail-operational performance. In the inertial
sensor sets this performance level is obtained through the use of skewed gyroscopes, and skewed ac-
celerometers (6 of each), and failure detection, isolation, and reconfiguration techniques. The ISS
ADS provides the same two fail-operational performance level through failure detection, isolation,
and analytical redundancy reconfiguration techniques. The methods of obtaining these two fail-
4I operational performance levels will be described in detail later on.
ISS SENSOR SYSTEM CONFIGURATIONS

1. ISS Air Data System Sensors, Data Handling System, and Redundancy Data Management System

The core of the ISS Air Data System (ADS) is in its set of redundant multipurpose air data probes.
Figure 3 shows a multipurpose air data probe configuration capable of sensing PS, PT, a , and 3.

Wl4
 NADC-82240-60

REDUNDANT I/O REDUNDANT

3
FLIGHT
•I CONTROL
DIGITAL
i ;i --
i COMPUTER
_ ] II l COMPLEX

"'- ~~~~~6
SKEWED/-_--_-.-..[_-L_

ACCELERO-
SRF NAV -M T R
METERS SUIRFACE*
FLUX AIDS POSITION / /
VALVES AND INS* /
COMMAND -INPUTS TO ISS
6
INPUT STHAT ARE NOT PART
OF THE ISS.

~~REDUNDANT
. AIR DATA
- ,PROBES/SENSORS

:':" t
TO FLIGHT
CONTROL & ENGINE ACTUATORS,
DISPLAY AND WEAPON SUBSYSTEM
Figure 1. Schematic and Signal Flow Diagram of the ISS Concept

However, because of its hemispherical nose shape, the pitot pressure (PT) indications are a function
of a and 3. With accurate position error corrections within the digital computer complex, true in-
dications of PT can be obtained.

Figure 4 shows the hemispherical sensor head of the multipurpose air data probe with two parts,
labeled PaI and P2. With the aircraft at a equal to zero, the two parts sense equal pressures.
At positive a, the pressure at the lower part (PaI) becomes greater than the pressure at the upper
port (Pa 2 ). The greater the value of a in reference to the relative air flow, the greater the pressure
differential between the two ports. a can be computed as a function of (PaI - Pa2) for either
positive or negative angles. However, this pressure differential increases with increasing dynamic
pressure. Therefore, a normalizing function must be introduced to account for changes in dynamic
pressure which is a function of altitude and airspeed. Normalization techniques for a and 3 mea-
surements are described in Reference 1.

Additional probe errors are caused by local pressure disturbances and pressure variations at each
probe location, which are functions of airspeed, Mach no.,a , 13,and the variations in flow geometry
caused by flaps, spoilers, landing gear, and control surfaces.

The pressure and differential pressure outputs of the air data probes are sent to pressure transducers,
as shown in Figure 2, via short pneumatic tubing runs to minimize the effect of transport lag. The

45
6.NADC-82240-60

UNIVERSAL TRANSDUCER
NON ISS INPUTS
MULTIPURPOSE FROM:
FROM:
MODULE ~PROBE * COCKPIT
CONTROLLERS
REDUN. * SURFACE
.DP AP DANT & POSITION
SURVIV- TRANSDUCERS
INTERFACE ELECTRONICS ABLE
WIRING. * THROTTLES

SBUS, OR * INS

TFIBER MAGNETIC
FLUX VALVES
S oINERTIAL 0 RF NAV AIDS

. ASSY o

:..',INERTIAL

- " PURPOSEO SENSOR

ASSY

MULTI- Lu INERTIAL z

- PURPOSE SENSOR "-

PROBE ASSY X
-- Z ]1 m DUAL ELEMENT

! u TOTAL

4TEMPERATURE PROBE

REDUNDANT &SURVIVABLE
"ICOMP UTATION SYSTEM

TO USING SUBSYSTEMS

E -. LOw SPEED PROBE

AND TRANSDUCER ASSEMBLY

Figure 2. Schematic of the ISS Components and Associated Signal FIcw Paths

6
 NADC-8224-60

P 2
PT 2

P P VIEW AA

,o..

LOCATION OF LOCAL STATIC PORTS

:-f" '" " "

7- - AIR
AI.
J ,F LOW

ALTERNATE BOOM L
MOUNTING STL FLOW ANGLE PORTS
LOCATED ON
HEMISPHERICAL HEAD
SWEPT STRUT (MAST)
FOR FUSELAGE MOUNTING
BASEPLATE ALIGNMENT PINS
ELECTRICAL HEATER CONNECTION
PRESSURE LINES

Figure 3. Multipurpose Air Data Probe

SENSING PORT

PNEUMATIC LINES P r2 \ ,

;.', FLOW ANGLE SENSOR

:4P SENIN POENINRHA

SFRONT VIEW OF

DIFFERENTIAL PRESSURE
READOUT

4 Figure 4. Pressure Differential with ct

47
 NADC-82240-60

pressure transducers are to be mounted in a Universal Transducer Module (still in the development
stage) which in addition to pressure transducers, will contain a microprocessor for preprocessing
data and provide data output control, a power supply, and I/O electronics. The pressure trans-
ducer outputs are usually non-linear with respect to ambient pressure and input pressure. In order
to correct for these non-linearities, the microprocessor will compute the required corrections based
upon a predetermined set of linearization coefficients contained within each pressure transducer.
After linearization, the corrected transducer data, under microprocessor control, is transmitted by
the I/O electronics to the redundant computer complex via redundant transmission lines as shown in
Figure 2. However, the linearized air data parameters transmitted to the redundant computer com-
plex contain probe errors due to local pressure disturbances and local pressure variations. The dif-
ferential pressures for a and j must also be corrected as functions of air speed and altitude.

In order to show the nature of the calculations required to remove these probe errors a signal flow
diagram for a typical analog Central Air Data Computer (CADC) is shown in Figure 5. The diagram
was obtained from Reference 2.

A Data Handling System (DHS) is integrated within the ISS Air Data System to perform the follow-
ing functions:

. Normalization of indicated air data parameters.

* Redundancy data management.
* Air data computations.

Normalization is a multi-parameter compensation scheme which converts the linearized pressure

transducer parameters from each probe location into useable air data parameters to perform the
Redundancy Data Management System (RDMS) functions and the subsequent air data calculations
to compute airspeed, altitude, altitude rate, Mach no., a , /3,and total temperature. The local flow
characteristics for the particular aircraft are required in order to accurately normalize the linearized
pressure data. The RDMS within the DHS performs failure monitoring, failure isolation, and signal
selection. In addition, through the use of analytical redundancy, a fourth channel of pressure ratio,
a , and 0 are derived from a set of complementary filters, using inertial navigation data and valid
air data parameters. Figure 6 outlines the methods employed by the RDMS to provide a two fail-
operational level of performance. The outline was obtained from Reference 2. Figure 7 is a block
diagram of the DHS. Figure 8 is a block diagram of the RDMS which is integrated within the DHS.
The failure monitoring that wi!l be performed by the RDMS consists of two types as shown in
Figure 8. The first type consists of a reasonableness test in which the latest calculated parameter is
"* compared with the previously calculated parameter. If the parameter exceeds the failure threshold
a transient failure is declared, and the associated data is not utilized in the subsequent calculations.
The failure threshold is a function of aircraft dynamics and short term noise characteristics of the
calculated parameter. For example, the threshold associated with the static pressure data will be a
function of the maximum change in altitude of the aircraft, plus the short term noise character-
istics of static pressure data that could occur between successive computation cycles.
0 The second type of failure monitoring performs comparisons between channels based on the most
current data. The associated failure monitor thresholds will be functions of the following:

. Probe-to-probe reproducibility.
- Accuracy and repeatability of position error calibration.
* *e Air data transducer accuracy.

8
 NADC-82240-60

PNEUMATIC TO ELECTRICAL COMPUTATION

TRANSDUCER SECTION SECTION INTERFACE SECTION

"" " E LECTRICAL

,"-"STATIC I p ALTITUDE ALTITUDE

PNEUMATIC PRESSURE ALTITUDE

INPUT. INDICATED PORT & MODULE INTERFACE

PS5 TRANSDUCER AP

~POSITION
STATIC PRESSURE
ERROR M

P S, Ps CORRECTOR

E LECTR ICAL

INPUT. INDICATED-*
TRACER
TOAND
TR A ISEEDTUE

I UI CETUCELECTRICAL
PRESSURE ,,
PNEUMATIC PROBE &
INPUT. INDICATED °i /-- J INTERFACE
PT, I

TOTAL
POSITIONPREESSURE
ERROR
i 'i ' . IC OR R ECTO R

MACH
P, COMPUTING MACH NUMBER
MODULE M INTE RFACE

AIRSPEED f INdiCATED

I COMPUTINGE
MODULEINRFC

aP PNUAIC

INPUT. INDICATED
. TEMPERATURE

TAnSignal FlowECTRICAL p c
(TASI

Og N t t
$IAIRSPEED

INTERFACE
T, MODULE

[TEMP COMP MOD I I ,TEFC

AGE TRANSDUCER ELECCOMPUTIN

VYING VYING SWEEP CMD I CM N

ANGLE TRANSDUCER 1ELECTRICAL INTERFACE

Figure 5. Computational Signal Flow in a Typical Analog Central Air Data Computer (CADC)

9
 ---.. v-. . . . r r7

NADC-82240-60

LM

SW 0 0 0
0N z z z

U.. c
0M GM V) cn

C, n .1-,

<I ~ -0 6L (f2.

o <00 0.0AD > u 0 < L,- 0 u..

Io z - V
ULnW < <U0wU

0. C)fl

- U UC
0. -0 <- < L , - C
00~~~
F- 2 C2 'ZF 2W2<>F lL
z < L < 0 < <
bl iCD AM 0 AM z AM C.)oDC

o-

C4 U,

CL 0

CC:2L << UZ
C2 2 0- a no
0
o I
n 0 i
2
U
r20
LU U LU A

CN C4

LU,

D 0 AMM

U. ni U

z ;w
-0 D (:).
<L
P #~. 1-0

U -U -J

I0

0 10
 NADC-82240-60

0w 0
0 C4M j ,
OI- : Z

QCNN
0 Z~~ C3~~
00 CL zcd
0 LL' 0
- nV0 0 I
U.j -0 a:wL> ,
1
(n Z in L 4 w
D~ wzjI > <u"
0
0 H ZUw J0 a -w
00=
0 L0
-ww1 09 -1
0H c L00,L F0><.U

V) z 0 D C

<1 CLLOXH >L)c

0
<IL%H(
2 m~ 0 0

H- 13 -L 0
W( Wz L-i W H-
U ***1n (C
3 WC 0 <0 U, -Z
-OHj 4 UJ z0 0%~u L
4Uw H ) uia
0 zU
03u 2 <~
CLC0 LC
0 0 ww 70 '
H
0
~ D 0

m 0 O H

n. 0 z (H z W W Z
H W~ ~~~ 00 W ~ ~~
Cc w
0 :

z n U. fA U.0
0 A. F- 01,
00
0C CC
- iL. 4. 0o CL C'4u
LL 4 o
21 uH w4a,

ZU u u cr
2 C
J ' V c

J
He * 0

4A 1
eQ NADC-82240-60

REDUNDANT
LINEARIZED/
CHARACTERIZED
AIR DATA SIGNALS SENSOR REDUNDANCY
DATA MANAGEMENT SYSTEM
____________________(RDMS) "BEST ESTIMATE" DATA

SENSOR NORMALIZATION a FAILURE DETECTION

* POSITION ERROR * FAILURE ISOLATION

COMPENSATION e SIGNAL SELECTION AR

* ANGLE OF ATTACK * ANALYTIC REDUNDANCY CODATA

COMPENSATION COMPUTATIONS COMPUTATIONS

* ANGLE OF SIDESLIP 0 "BEST ESTIMATE"

COMPENSATION COMPUTATION

__ OUTPUT

* INERTIAL VELOCITY DATA

* ATTITUDE HDG DATA

Figure 7. Block Diagram of the Data Handling System (DHS)

NORMALIZED CHI 0 go TRANSIENT PERMANENT FAILURE

AIR (CH2 FAILURE MONITOR MONITOR
DATA * COMPARISON MONITOR
PARAMETERS CH3 REASONABLENESS PS/P t Tt
-- - CHECK Of

ANALYTIC SINL"ETAIR
REDUNDANCY ION ESTIMATE" DATA
CALCULATIONS CALCULATION CALCULATIONS

It
INERTIAL NAVIGATION
DATA

Figure 8. Block Diagram of the Redundancy Data Management System (RDMS)

12
 NADC-82240-60

* System dynamic response errors.

* Errors associated with analytical redundancy computations.

Should a parameter exceed the failure threshold for a specified length of time, a permanent failure
isdeclared.

The signal selection routines in the DHS control the sensory data which will be utilized to determine
the "best estimate" for each air data parameter. Should a transient or permanent failure occur, the
associated sensory data is not utilized in the "best estimate" computations.

The final element in the DHS is the air data computations which utilize the "best estimate" sensory
data from which true airspeed, Mach no., altitude, t, 03,and etc. are calculated and then made avail-
able to the using subsystems via the I/O System of the ISS.

Table 1shows how the air data parameters are shared among the using subsystems.

2. ISS Inertial Data System Sensors, Data Handling System, and Redundancy Data Management
System

The ISS Inertial Data System uses six skewed rate gyros and six skewed linear accelerometers in order
to obtain a two fail-operational performance level while minimizing the number of sensors and
sensor types required. The six linear accelerometers have the same linear acceleration sensing range
and the six gyros have the same angular rate sensing range. This feature minimizes the number of
sensor types required. The refinement of the skewed sensor concept was advanced by the Grumman
Aerospace Corporation during its development of the Advanced Skewed Sensory Electronic Triad
(ASSET) System for Flight Control (References 4 and 5). The ASSET concept was successfully
flight tested in an EA-6B aircraft.

The ASSET program was limited to the development of skewed rate gyro configurations. No effort
was expended in the development of skewed linear accelerometer configurations. The sensing axis
of each of the six rate gyros were skewed in a manner to permit sensing of components of pitch,
roll, and yaw rates about the aircraft's body axes. This feature permitted the development of rate
gyro configurations that provided dual fail-operational capability with only six rate gyros. In the

Table 1. Air Data Parameter Requirements Summary

'%-I
.
I-
u in I-
4 LU0

PARAMETER "w "w U. L.

0 0
- I. C. _
SUBSYSTEM > Z ,

COCKPIT DISPLAYS X X X X X X

OUTER LOOP FL IGHT CONTROLS X X X" X X X

INNER LOOP FLIGHT CONTROLS X x X X

VARIABLE GEOAE1 RY INCLUDING AIR X X . X

INLET & V/STOL ENGINE ORIENTATION

NAVIGATION WEAPONS SYSTEM x X X X

'AS BACKUP TO INERTIAL VELOCITY AT LOW SPEEDS.

,13
1
 NADC-82240-60

conventional approach of employing four rate gyros per body axis, mounted orthogonally, twelve
rate gyros are required to obtain dual fail-operational capability (via use of majority voting tech-
niques). Thus the ASSET approach permits the elimination of six rate gyros. Figure 9 illustrates
the salient features of both the conventional and the ASSET concepts.

Figure 10 shows the geometry of the sensing or input axes of the six rate gyros which are oriented
to position them on the surface of a cone, with the cone axis oriented along the aircraft's Z body
axis. The central half-angle is optimized to permit the use of rate gyros with identical sensing range
capability despite the fact that the maximum rates about each of the aircraft's orthogonal body
axes differ. Other criteria for cone orientation and cone central half-angle selection are to:

9 Minimize Analog-to-Digital (A/D) resolution requirements.

e Minimize the average total error amplification due to skewing.
0 Retain the quantization necessary to satisfy FCS and Strapdown Attitude and Heading
Reference System (SAHRS) requirements.

With careful selection of the cone central half-angle, based on the knowledge of the maximum rate
requirements of the aircraft, it is possible to maintain the accuracy of the cone configuration equiv-
alent to that of the conventional, orthogonally mounted, 3-axis rate gyro package. In Figure 10 the
apex of the cone is pointed forward. It may also be pointed rearward as shown in Figure 1.

The rate sensed by each gyro is a function of the following:

e Gyro orientation angle

* Gyro central half-angle
e Rates about the aircraft body axes
e Rates induced by the fuselage bending modes and the vibrations and bending modes of the
mounting bulkhead

The following equation, in matrix form, describes the actual signal from each rate gyro:

W1 sin 0q 0 cos 3q

w'-2 -sin 3q
0 cos Oq wozT
sin q cos aq -sin 3q sin Oq cos q Equation 1
".)3

w4 -sin Aq cos oaq sin Oq sin oaq cos Pq

w5 sin Pq cos O 1 sin 3q sin aq cos Pq WXT

L 6 sin 3q coscaq -sin1 3q sin 01q cosf 3q

where:

•.q = to the gyro orientation angle

O,
q = to the gyro central half-angle

*i 14

" -. " . . -
 NADC-82240-60

t=

CaC

.0

0.

C 0

CmC

CD c

0 U L0

< LL

LUI

>

4 15
 NADC-82240-60

".3- A 30o

.'-.
6 0' PITCH
4 AIRCRAFT
S AW BULKHEAD

6-
Z Y AW

5CENTRAL HALF ANGLE

• - INPUT AXES

2
-
~TWO-PAC SENSOR
K

•BULKHEAD

SECT A-A

Figure 10. Geometry of Rate Gyro Sensing Axes for the ASSET Concept

ci = to the rate sensed by the ith gyro (i = 1 to 6)

WZT = to the total rate about Z axis
=-t

w'YT == to
woXT the total rate about Y axis
to the total rate about X axis

(See Figure 11 for the geometry of the skew angles)

wz ZT,yT and' wXT, are a function of aircraft body axes rates, fuselage bending, and bulkhead
bending and vibrations. The errors introduced by the structural bendings and vibrations are removed
by the ISS Inertial Sensor Data Handling System and the true values of the aircraft rigid body axes
rates cwz, wy, and cj x are made available to the FCS and for attitude and heading computations.
These tasks are performed in the digital computer complex with the true values of the aircraft's
body rates being updated at the rate of about 50 iterations per second.

The ISS inertial sensor set utilizes six skewed linear accelerometers of which the input or sensing
axes form a cone surface oriented along the aircraft's Z body axis as shown in Figure 1. The criteria
for cone orientation and cone half-angle selection are the same as those for the skewed rate gyros.

6i
16
 NADC-82240-60

41 6

Figure 11. Geometry of a Conical Array of Six Rate Gyros

The linear acceleration sensed by each accelerometer is a function of the following:

e Accelerometer orientation angle.

e Accelerometer central half angle.
* Acc lerations along the body axes.
. Length of moment arm from "rcraft C.G. to the accelerometer.
* Length of moment arm from center of rotation of the bulkhead to the accelerometer.
a Aircraft angular rates and angular accelerations about the body axes.
e Flexing and bending mode characteristics of the mounting bulkhead and fuselage.

The following equation, in matrix form, describes the actual signal from each linear accelerometer:

al -osjPa 0 sinOa

a2 -cO,,a
-cos P3a
O
0 -sin 03a
ZT]
az
a3 -cos 0a -cos 0 a sin 0a sin 0 a sin Oa a
YT Equation 2
a4 cos 9a -cos 0 a sin/Oa sin 0 a sin Oa
a5 -cos 0a cos 0 a sin Oa sin a sin Oa -axT

a6 cos a cos a sin Oa sin a sin Oa

I.7
i NADC-82240-60

where:

Oa = to the accelerometer orientation angle

= to the accelerometer central half-angle

ai = to the linear acceleration sensed by the ith accelerometer (i = 1 to 6)

aT
(Iz = to the total acceleration along the Z axis

ayT = to the total acceleration along the Y axis

= to the total acceleration along the X axis

(See Figure 12 for the geometry of the skew angles)

azT, ayT, and axT are functions of accelerations along the aircraft body axes, aircraft angular rates
and angular accelerations about the body axes, and flexing and bending mode characteristics of the
mounting bulkhead and fuselage. The errors introduced in these terms are removed by the DHS
and the true values of ax , ay, and a z are made available to the FCS, SAHRS, and other using sub-
systems at the rate of about 50 iterations per second.

The rate gyros and the linear accelerometers of an ISS are dispersed in order to enhance its surviv-
ability in combat operations. Rate gyro and linear acceleration information used for self-contained
failure detection and isolation, flight control feedback and for attitude reference purposes must be
adjusted to account for rigid body aircraft motion (about the center of gravity) and flexible fuselage

Y. .

4 .

Figure 12. Geometry of a Conical Array of Six Linear Accelerometers

S18
 NADC-82240-60

and bulkhead motions. Centripetal and tangential acceleration "errors" that arise from rigid and
flexible body motion must be "normalized" or removed from the outputs of hard mounted linear
accelerometers (located off of the C.G.) before the information is used for failure detection and
isolation, flight control functions or as inputs to the SAHRS.

With flexible inputs not removed from the rate gyro outputs prior to entering the SAHRS routines,
the indicated attitude will reflect flexible body as well as rigid body motion.

To minimize errors in estimating the centripetal r) and tangential (r J ) accelerations for

SAHRS data, the concept of a virtual NAV-BASE and digital bending filter incorporated within the
DHS, was developed. The virtual NAV-BASE eliminates the requirements for accurately locating
the linear accelerometers with respect to the C.G. of the aircraft. The digital bending filter allows
differing bending inputs, sensed by the dispersed linear accelerometers, to be precisely removed with
minimum effect in system bandwidth. For flight control usage, these acceleration data, referenced
to a virtual NAV-BASE, are then "translated" to the C.G., or to any other location in the aircraft
that satisfies control law requirements (i.e., under the cockpit for ride control).

Table 2 (obtained from Reference 1) lists the adjustments required to account for rate gyro and
linear accelerometer dispersion.

Figure 13 shows a block diagram of the ISS Inertial Sensor Data Handling System (DHS) which is
imbedded as software in the digital computer complex.

Inputs to the DHS are provided by the six rate gyros and the six linear accelerometers. The signal
flows from the rate gyros, their normalization, redundancy management, and subsequent distribu-
tion will be described first.

Table 2. Adjustments Required to Account for Rate Gyro and Linear Accelerometer Dispersion

NORMALIZATION FOR REMOVAL OF FLEXIBLE REMOVAL OF FLEXIBLE

FAILUR DETECTION AND RIGID BODY INPUTS AND RIGID BODY IN-
AND ISOLATION FOR FLIGHT CONTROL PUTS FOR SAHRS
(1) 12) (3

GYRO NOT REQUIRED IF NOTCH FILTERS AFTER NOT REQUIRED

GYROS ARE ON IN VOTING
FLEXIBLE BULKHEAD*

ACCELEROMETERS RIGID BODY ANGULAR RATE (w) AND ESrIMATE

OF ANGULAR ACCELERATION I .
NECESSARY FOR _', AND rU TERMS**

FLEXIBLE
BODY NOTCH FILTERS FOR EACH SENSOR
PRIOR TO VOTING

*INFLEXIBLE AT BODY BENDING FREQUENCIES. HIGHER FREQUENCIES HANDLED BY ANALOG PRE,-IL-

TER IN 1,0
r IS DISTANCE FROM ACCELEROME TERS TO VIR TUAL NAV BASE FOR COLUMNS I AND 3 AND DIS
TANCE TO CENTER OF GRAVITY FOR COLUMN 2

19
 NADC-82240-60

Nz 4

<44 ~
SNaLu~lJ 91038cH~1d~Zx

z -z

~z
C~.C>

<. z

4~t Z~ - ~Z4. >

0 3Z

IC 7-

z
 NADC-82240-60

The inputs to the skewed rate gyros are the rigid body axes rates (p, q, r) and the undesired fuselage
, and bulkhead bending rates. Equation 1 is the rate gyro skew matrix. The six simultaneous linear
equations are solved at the iteration rate of approximately 50 times per second to obtain updated
values of c~xT, WyT, and WZT. Extrapolated fuselage and bulkhead bending rate estimates from
the State Estimator are subtracted or filtered from cwxT, Cy T , and WZT and the resultant values
are used to aid in the rate gyro normalization process. The angular rates sensed by each rate gyros
input axis (wl, w2, W'3, w4 , w 5 , wo6 ) are also sent to the Rate Gyro Normalization Routine for
filtering. The filtered values are used as an aid in the failure detection processes, and calculation of
p, q, and r, and are inputed to the Rate Gyro RDMS (See Figure 14). Imbedded within the Rate
Gyro RDMS are the following routines:

a Transient Failure Removal Routine (TFRR)

* Sensor Voting Selection Routine (SVSR)
* Failure Isolation Computation Routine (FICR)

The Transient Failure Removal Routine (TFRR) accepts four types of inputs. These are as follows:

* (1) Rate gyro output signals.

(2) Estimated rate gyro outputs.

(3) Failure Isolation Computation Routine (FICR) failure signal.

(4) In-flight BIT (IFBIT) failure signal.

Each rate gyro (after filtering) furnishes an output signal corresponding to the angular rate applied
about its input axis. The maximum rates expected from each rate gyro depends on the cone half-
angle and on the maximum roll, pitch, and yaw angular rate capabilities of the aircraft. The estimated
outputs, one for each rate gyro, are available from FICR computations carried out during the prev-
ious iteration. Whenever a rate gyro output is unacceptable for a sufficient period to establish it as
permanently failed the FICR makes the information available to the TFRR so that the failed rate
gyro is not included in any future calculations.

During every iteration each rate gyro output signal is subtracted from the estimated output. The
difference is compared with the threshold corresponding to each rate gyro. If this difference is less
than the threshold the output signal is changing at an acceptable rate and the rate gyro is considered
- - to have no temporary failure. If the difference is greater than the threshold the rate gyro is classi-
fied as temporarily failed and a signal is made available to notify the Sensor Voting Selection Routine
(SVSR) and the F ICR. The signal also identifies which rate sensor has failed.

The SVCR produces satisfactory estimates of the angula, ,ites about the roll, pitch, and yaw axes of
the aircraft. The voting subroutines (i.e., VOTE 6, VOTE 5, VOTE 4, and VOTE 3) which compute
the angular rates are selected on the basis of the failure state.

When there are no failures, the six rate gyro outputs exceed the number required to obtain satisfac-
*tory values of p, q, and r. To minimize computational resource requirements VOTE 6 has been de-
signed to select 4 rate gyros, using parity equations, so that the error in the estimate of aircraft rates
is minimized. To eliminate the first rate gyro these equations compare each rate gyro output with
the estimated value of its output, obtained from the other three rate gyros. The rate gyro output
which deviates most from its estimated value is not used in the calculation of p, q, and r. The addi-
tional rate gyro is removed by selecting a favorable configuration that minimizes the error in the
rate output.

21
 NADC-82240-60

INPUTS FROM RATE GYRO NORMALIZATION

ROUTINE (FILTERED w 1 , 'J 2 , w 3 , W4 , uJ5 , 6 )

SENSOR VOTING IFBIT FAILURE SIGNAL

COMPUTATIONAL
ROUTINE (SVCR)
TRANSIENT FAILURE TEMPORARY>
REMOVAL ROUTINE FAILURE
(T R )DISCRETES (6). CL

TO STATUS DISPLAY

VOTING SELECTION -""

VOTE VOTE VOTE VOTE

- "I6 5 4 3

ORTHOGONAL RATES (p,q, & r)

TO USING SYSTEMS

F FAILURE ISOLATION
COMPUTATIONAL FICA ROUTINE
I ROUTINE (FICR) CALCULATE SELECTED ON BASIS
I ESTIMATED SENSOR OF PERMANENT
OUTPUTS FAILURE STATES

UI NE

I I
p. . FICR 6 F R5FIR 4

PERMANENT
FAILURE DISCRETES

" Figure 14. Rate Gyro Redundancy Data Management System (RDMS)

* 22

- - -
 NADC-82240-60

When one sensor has been classified as a failure, VOTE 5 selects the best four of the remaining five
by the use of parity equations. With two failures no choice of rate gyros that would decrease roll,
pitch, and yaw error is possible and, therefore, the four remaining are used to compute the angular
rates in VOTE 4.

The occurrence of three failures leaves only three rate gyro outputs available to calculate p, q, and r
in VOTE 3. Lnder this condition a warning signal is generated to inform the pilot of the potential
danger.

Descriptions of the computational processes of the four VOTE subroutines are contained in Appendix
2 of Reference 4.

The Failure Isolation Computational Routine (FICR) detects and identifies rate gyros which have
permanently failed. Two ways of examining for permanent failures are used. First, if any of the
rate gyro estimated outputs exceed 5 degrees per second, the error of those rate gyros are examined
using a Cross Voter Comparison Monitoring (CVCM) technique while the other rate gyros are not
tested. If all of the rate gyro estimated outputs are less than 5 degrees per second the errors from
all of the rate gyros are estimated using a Cross Channel Comparison Monitoring (CCCM) technique.

Using the CVCM technique, the values of the p, q, and r rates produced by the SVCR are used to
generate estimates of the six rate gyro outputs. The individual estimates of the outputs are then
divided by the actual rate gyro outputs. If the absolute value of (dividend - 1.0) is the largest of all
being examined and greater than a threshold for a period of 0.2 seconds, the rate gyro is classified
as a permanent failure.

With the CCCM technique, used when the rate gyro outputs are less than 5 degrees per second, the
error of a rate gyro is estimated by means of one of a set of fifteen equations (called "parity equa-
tions") which are defined and described in Reference 4. Briefly the fifteen parity equations describe
the relationships between all possible groups of four rate gyros selected from the six in the skew
arrangement.

When the parity equations are all equal to zero the rare gyro outputs are error free. However, an
equation not equal to zero indicates an error and a possible permanent failure. If the error persists
continuously for at least 1 second, the rate gyro producing the error is classified as permanently
failed.

With no permanent failures subroutine FICR 6 is used. With one permanent failure subroutine 5 is
used. When none or one previous permanent failure exists, the next permanent failure can be ident-
ified by the parity equations. With two permanent failures (FICR 4), the parity equations indicate
that one of the four remaining rate gyros is a permanent failure but it cannot identify the one that
has failed.
The FICR subroutines are defined and described in detail in Reference 4.

i* Existing outside the functional boundaries of the RDMS, the IFBIT detects catastrophic (permanent)
failures of the rate gyros. The IFBIT output furnishes information that a rate gyro has failed and
its identity.

, The inputs to the skewed linear accelerometers are the point mass accelerations, and the undesired
rotational, fuselage and bulkhead bending accelerations. Equation 2 is the linear accelerometer skew
"* matrix. The six simultaneous linear equations are solved at the iteration rate of approximately 50

23
w . . . .
 NADC-82240-60

times per second to obtain updated values of axT, ayT, and azT. Extrapolated fuselage and bulk-
head bending accelerations and rotational accelerations from the State Estimator, and the p,q, and
r angular rates from the Rate Gyro RDMS are used to remove the errors in axT, ayT, and azT and
the resultant values are used to aid in the linear accelerometer normalization process. The linear
accelerations (a1 , a a 3 , a4 , a5 , a6), sensed by each linear accelerometer, are sent to the Linear
Accelerometer Normalization Routine for filtering. The filtered values are used as an aid in the
failure detection and isolation process, and the calculation of ax ,ay, az, and are inputed to the
Linear Accelerometer RDMS (see Figure 15). The calculation processes, logic, and signal flow are
the same as those for the Rate Gyro RDMS.

The State Estimator of the ISS Inertial Sensor DHS is used to calculate an estimate of the aircraft's
fuselage and bulkhead bending rates and accelerations as well as the kinematic accelerations at the
dispersed sensor locations. The estimates of bending and kinematic effects are then subtracted from
the raw sensor data to permit effective failure detection and isolation and accurate calculation of
the aircraft's orthogonal angular rates and linear accelerations.

. State-of-the-art digital design techniques were applied by the Grumman Aerospace Corporation to
develop the State Estimator. Reference 6 describes the development and functions of the State
.Estimator.

" 3. ISS/SAHRS Function for Sensing Attitudes, Heading, and Inertial Velocity Components

A Strapdown and Heading Reference System (SAHRS) is integrated within the ISS for the calcula-
tion of the following parameters using corrected values of p, q, and r and ax, ay, and az and inputs
from non ISS sensors:
* Roll altitude ( )
* Pitch altitude ( 0
* Heading altitude ('I')
9 North component of velocity
e East component of velocity
e Vertical component of velocity

Corrected values of the p,q, and r angular rates, obtained from the ISS Inertial Sensor DHS are inte-
grated once and processed to obtain the Euler Angles (b, 0, and *. The corrected values of ax ,
ay, and az orthogonal linear accelerations, also obtained from the DHS, are integrated once to ob-
tain the inertial velocity components along the aircraft body axes and thse values together with the
Euler Angles are used to obtain the North, East, and Vertical inertial velocity components. The inte-
grated values of these velocity components are used to calculate position changes for navigational
purposes.

The outputs of the ISS/SAHRS are used by the following aircraft systems:

* Flight Control System (FCS)

e Automatic Flight Control System (AFCS)
e Cockpit displays
* Weapons delivery

* 24
 NADC-82240-60

INPUTS FROM LINEAR ACCELEROMETER NORMALIZATION

ROUTINE (FILTERED a 1 , a2 , a3 , a4 , a5 , a6 )

ISENSOR VOTING -J
COMPUTATIONAL IFBIT FAILURESIGNAL
IROUTINE (SVCRI

TRANSIENT FAILURE -
REMOVAL ROUTINE TEMPORARY
(TFRR) DISCRETES
FAILURE (61. 0,,
~TO STATUS

VOTING SELECTION
ROUTINE

rVO
E vo-r]v, I Ior
VOTE
6
VTE
5
VOTE
4
VOTE
3

ORTHOGONAL ACCELERATIONS
(ax . ay, az ) TO USING SYSTEMS

FAILURE ISOLATION FICR ROUTINES SELECTEDl

COMPUTATIONAL ON BASIS OF PERMANENT
ROUTINE (FICR) CALCULATE FAILURE STATES
ESTIMATED SENSOR
OUTPUTS

FICR SELECTION
ROUTINE
..
........

PERMANENT
FAILURE DISCRETES

Figure 15. Linear Accelerometer Redundancy Data Management System (RDMS)

25
 NADC-82240-60

Figure 16 shows the ISS/SAHRS Signal Flow Block Diagram. The skewed rate gyro and linear
accelerometer data are resolved by the ISS Inertial Sensor DHS, to three outputs along vehicle axes.
Spurious inputs from aircraft rigid body and flexible body motions, coupled with the dispersion of
skewed vector sensors, are removed by the DHS.

After initialization of the inertial frame to vertical and north the Attitude Transformation Update
and Orthonormalization Routine operates on the rate gyro outputs to maintain the knowledge of
the angular relationship between the aircraft body and inertial frame. This angular relationship is
used to transform the linear accelerometer data obtained along the body axis to the vertical. The
information is then mixed with "aiding" data such as air data velocity, to "erect" the system through
a long-time-constant filter to the vertical. Therefore, the long-term accuracy of the vertical indica-
tion depends on the accuracy of the reference velocity. The reference frame is "slaved" to north,
defined by the magnetic heading information, again through a long-time-constant filter.

Relative velocity and relative position (range and range rate) can be obtained from this system when
the system is aided with relative position data from a landing guidance sensor such as those described
in Reference 7.

Alignment on the ground and in flight will be performed in two steps:

e Coarse Alignment - where one or two iterations of the magnetic heading and accelerometer
data are used to generate a crude estimate of attitude and heading (body axes orientation with
respect to vertical and north). After this is completed, the system reverts to Fine Alignment.
e Fine Alignment - is composed of a 40-sec time constant, second-order vertical erection loop
that uses reference velocity, and a 10-sec time constant, second-order heading filter that uti-
lizes magnetic heading information.

Attitude transformation presents one of the most demanding requirements on the throughput of a
digital computation system. Studies utilizing the information in Reference 8 have shown that a 50
msec update rate using a second-order quaternion algorithm will be satisfactory for SAHRS per-
formance. Typically, at a 300 deg/sec roll rate input the computational error is approximately 0.04
degrees per iteration. Subsection 3.4 of Reference 2 shows that this SAHRS 50 msec update rate
will satisfy ISS computational requirements.

The results of the SAHRS performance analysis and an equipment survey showed that in-production
Single-Degree-Of-Freedom (SDOF) rate integrating gyros in the 12,000 to 30,000 gm-cm/sec angular
momentum class and in-production linear accelerometers with self-contained loop electronics will
satisfy the performance requirements of an ISS.

Typical rate gyro performance requirements for an ISS/SAHRS are listed as follows:

e Random drift 0.20 deg./hr. (0 a

* Start/start bias repeatability 0.20 deg./hr. (la
4 * Gsensitivity drift repeatability 0.20 deg./hr./g (lOa
* Anisoelasticity 0.50 deg./hr./g 2
9 Compensated temperature sensitivity 0.006 deg./hr./°C
e Wheel run-up time 3 seconds
• * Scale factor linearity 200 PPM
(parts per million)

4 26
 NADC-82240-60

z
)- >

2x

• &0. *>.

- aa m
zz >

02 = 2 2 w

x 0
<" Li-.,,,5

il-u- -

-: --., -

,I,

II

L~ll 27
e NADC-82240-60

Overall SAHRS performance and accuracy requirements are listed as follows:

" Magnetic Heading 0.5 degree (lo)

* Pitch 0.15degree (lo)
* Roll 0.15 degree (lo)
- Body rates 0.5%
, Linear acceleration 0.005 g

(Note: The above performance and accuracy values are attained by the Sperry SRS-1 100 SAHRS)

The performance requirements for the ISS/SAHRS inertial sensors are dictated by the accuracies
required for inputs to the FCS of a VTOL type aircraft designed to land on small ship platforms
during adverse weather and sea state conditions.

ISS REDUNDANT INPUT/OUTPUT (I/O) SYSTEM

The redundant and dispersed Input-Output (I/O) system is the element of the ISS concept that
links the data from the redundant sensor set to the using systems via a redundant digital computer
complex. To reduce vulnerability to combat damage the I/O system signal paths are redundant and
dispersed. Figure 17 is a simplified signal flow and block diagram of an ISS I/O system for a DFCS.
It represents only one of the many methods in which an I/O system may be configured. Flight
safety critical inputs to the digital Flight Control Computer (FCC) complex are transmitted via
redundant and dedicated signal paths. Output signals from the FCC's to flight safety critical signals
are transmitted in the same manner. Signals from the FCC's to systems which can be in a failed
state without degrading the controllability of the aircraft are transmitted via the Dualized Avionics
Multiplex Bus.

Note that the outputs of the Air Data Sensors, which are in the digital format, are inputed into the
FCC's via dedicated triplex multiplex bus signal paths. Figure 18 shows this interface in greater
detail. Each of the three Universal Transducer Modules shown contain a microprocessor for pre-
processing the air sensor data and to provide data output control, a power supply, and the I/O elec-
tronics necessary to interface with the redundant multiplex buses. Since the air data values are in
digital form in the microprocessors their transmission to the FCC's can be most efficiently accom-
plished via multiplex buses.

The other flight safety critical inputs to the FCC's shown in Figure 17 are in the analog format and
are converted to the digital format in the FCC's.

For all inputs the functions of failure detection, failure isolation and signal selection are performed
in the FCC's. Signal outputs from the FCC's to the flight safety critical systems (after undergoing)
D/A conversion) are in the analog format.
Methods of inputting the critical ISS inertial sensor data to the FCC's will now be described. The
most reliable method is that of "cross-strapping" the output of each inertial sensor to all three FCC's
as shown in Figure 19. The inertial sensors are housed in three Inertial Component Assemblies
(CA's). Each ICA contains two skewed single degree-of-freedom (SDOF) rate gyros and two skewed
linear accelerometers. The ICA's are widely dispersed and mounted on one fuselage bulkhead to
reduce the vulnerability of the ISS inertial sensors set to combat damage. Loss of an ICA will not
prevent the FCC's to output corrected values of p, q, r, ax , ay, and a z to the using systems. The

28
 NADC-8224O&60
wi -j

zz
~Jw Wj<

0~ -J<
0 <) zz

-i CL
-j

< xx
z z> < Z w O0
<~ ch
<- cj LW
<U m

w w _j L

F1 D z 2
u a >
m
0< 0
z
<D

z -J
'nn

ICZ -iZ(

N U,

0I x 0 0

<JO~ J O. j 00
uU ~ z. CL~

<- oz - L6 - = MM 0

LIC.

- ' L c cn Z

-2-, o o-Lo C

u U J -cf

;c- Z w- I Z
I Ou.
z
LLI C LOL
_~
<I <.0
0.0i uDC
'n .'

<llN 1V3iI A13V .H

IL < p< u-i c ui Lcc V29
 NADC-82240-60

UNIVERSAL TRANSDUCER flMULTIPURPOSE

MODULE PROBE

PS PT A.a APa

L1NTERFACE ELECTRONICJ
REDUNDANT
MULTIPLEX DATA
BUSES

MULTI--
PURPOSE <U

z 0 -10 DUAL ELEMENT

m
- r TOTAL
CL___(_ TEMPERATURE PROBE

I NTERCOMPUTER

TO USING SYSTEMS

* Figure 18. Air Data System/Flight Control Computer Complex Interface

0 30
 -171

NADC-82240-60

C,,L

- CL

0.

ca~

u 0 0 U

0I 00
0J z zL
a4
cn cn (

U u C,44-

310
 NADC-82240-60

losses of two like sensors (two rate gyros and/or two linear accelerometers cai be readily compen-
sated for by the RDMS of the ISS Inertial Sensor DHS.

The output signals of the inertial sensors are in the analog format and the required A/D conversions
are performed in the FCC's. Note that each FCC has twelve dedicated input ports to receive the
inertial sensor signals. For each input port there is a dedicated A/D converter. It is readily apparent
that the cross-strapping method is costly in terms of system Life Cycle Costs (LCC), hardware parts
count, and weight.

Cross-strapping is used in the quadruplex DFCS of the F-18. To attain two fail-operational per-
formance the DFCS utilizes four FCC's, twelve rate gyros, and twelve linear accelerometers (four
sensors per axis). The axes of the F-18 inertial sensors are orthogonal. Each FCC has twenty-four
dedicated input ports to receive the analog signals of the twelve rate gyros and twelve linear accel-
erometers. In contrast an ISS concept using cross-strapping requires half as many input ports (12
vs. 24) to obtain two fail-operational performance.

A less costly method of inputing inertia sensor signals is shown in Figure 20. The output of each
inertial sensor (in the analog format) is sent to only one of the FCC's and then transmitted to the
other two FCC's via the inter computer high speed data link. Since each FCC receives data from all
inertial sensors all FCC's are able to perform the ISS Inertial Sensor DHS function. Note that each
FCC now requires only four dedicated input ports and A/D converters. This method of inputing
inertial sensor data to the FCC's is employed in the triplex DFCS of the AFT1/F-16 aircraft.
Studies conducted by the General Dynamics Corporation have shown that this method is less costly
than cross-strapping with only a small decrease in reliability.
" ,~NTER COMPUTER HIGH _Ji
"'" SPEED DATA LINK

I A SDOF GYRO

1 LINEAR ACCEL.
LINEAR ACCEL.

SDOF GYRO

ICA SDOF GYRO WOUTPUT'S

FC SO GOF
FCC'S TO
2 LINEAR AccEL.
USING
NEAR ACCEL.YSTEMS

SDOF GYRO

ICA SOOF GYRO 111110

LINEAR ACCEL.

Figure 20. Interface of the Inertial Component Assemblies with the Flight Control Computers
Via Inter Computer High Speed Data Link

4 l32
 NADC-82240-60

In some future DFCS configurations it might be desirable not to burden the FCC's with the ISS
Inertial Sensor DHS function. This burden could be placed upon a network of distributed micro-
processors ( Pp's) configured as shown in Figure 21. Each inertial sensor output signal is sent to
only one input port of a pp and then transmitted to the other two pp's via an inter computer
high speed data link. The outputs of the three ICA pp's are corrected values of p, q, r, aX, ay,
and aZ which are inputted, to the three FCC's.

ISS COMPUTATION NETWORKS WITHIN REDUNDANT DIGITAL COMPUTER COMPLEXES

The use of the ISS concept in aircraft with DFCS's does not impose the requirement for the develop-
ment and procurement of dedicated or special purpose digital computers. This course of action
would be most undesirable as it would greatly increase the Life Cycle Costs (LCC) of operational
aircraft integrated with an ISS. The required computations for an ISS may be performed in digital
computer architectures listed as follows:

(a) Use of available on-board redundant FCC's on a time shared basis.

(b) Use of a computational architecture containing both redundant distributed microprocessors

and redundant FCC's on a time shared basis. The microprocessors, dedicated for ISS func-
tions, would reduce the computational load of the FCC's.
INTER COMPUTER HIGH SPEED
DATA LINKS

SDOF GYRO

ICA SDOF GYRO ICA FCC

1 LINEAR ACCEL. I = 01
LINEAR ACCEL.

ICAOF OUTPUT'S
SDOF GYRO
I SDOF GYROCApF FCC'S TO

•2 LINEAR ACCEL. - =2 USING

=- . SYSTEMS
- LINEAR ACCEL. S

SDOF GYRO

ICA SDOF GYRO ICA Ap

3 LINEAR ACCEL. - =3

LINEAR ACCEL.

*CORRECTED VALUES OF x, y'W' ax, ay ANDaz

Figure 21. Interface of the Inertial Component Assemblies and ICA pp's with the Flight
Control Computers

4 33
* NADC-82240-60

THE ADVANCED FIGHTER TECHNOLOGY INTEGRATOR (AFTI/F-16)

DEVELOPMENT AND INTEGRATION PROGRAM

The purpose of the Advanced Fighter Technology Integrator (AFTI/F-16) Program is to design,
develop, and integrate into a modified F-16 aircraft, a set of advanced technologies capable of pro-
viding improved weapon system effectiveness and survivability in air-to-air and air-to-surface
The set of technologies to be designed, developed, integrated, and verified by flight tests in an
AFTI/F-16 aircraft are listed as follows:

* Advanced Digital Flight Control System (DFCS)

* Advanced Integrated Fire and Flight Control (IF FC) System
- Advanced Armament/Weapon Delivery System
a Aerodynamic Optimization
* Advanced Crew Stations Systems and Concepts

Figure 22 summarizes the objectives of the AFTI/F-16 Program.

The AFTI/F-16 Program is being sponsored and funded by the Air Force, Navy, and NASA. The
prime contractor is the General Dynamics (GD) Corp./ Fort Worth Division. The Navy is providing
part of the funding and program direction for only one of the set of technologies being developed;
the advanced Digital Flight Control System (DFCS). Only the design and development of this tech-
nology will be described in this report. However aircraft subsystems interfacingwith the DFCS will
also be described. The subcontractor for the design, development, and fabrication of the main
components of the DFCS is the Flight Systems Division of the Bendix Corporation located at
Teterboro, New Jersey.

The general objectives of the AFTI/F-16 DFCS Advanced Development Program are as follows:

0 Demonstrate the operation of a second generation DFCS.

0 Evaluate the operational capabilities of a triplex DFCS.
0 Demonstrate the benefits of multi-mode flight control capability.
0 Determine the configuration of an independent flight control backup system.
* Evaluate pilot acceptance of Direct Lift Control (DLC), Direct Side-Force Control (DSF)
and fuselage pointing capabilities.
. Demonstrate increased mission effectiveness and reduced pilot workload.
* * Develop adata base and experience for future DFCS applications.

Some of the more specific objectives are as follows:

0 Define computer characteristics

- Architecture
* - Throughput requirements
- Memory requirements

*I 34
 NADC-82240-60

LU Z

LU

Ca Lu

L LU U--MCmC
C=~ -J ~ui
LU _u

cc LU CM

CEE
00

auusi
G3U-\\\ U.

0 LU\'
LU

-J

CL L

W C.)

LU c

CJ L)U

u. -a LU

LU 0a 0 .
Waa9 L)U 0LUjCC
-00 LUZ J
_U LUOO

35
 NADC-82240-60

" Evaluate redundancy data management for dual fail-operational (F/O) capability.
* Evaluate confidence level of in-line testing.
* Define independent backup flight control switching intelligence.
- Define 1553 Multiplex Bus interfaces.
0 Confirm multi-mode selection
- Normal
- Air-to-air
* - Air-to-ground strafing
- Air-to-ground bombing.

DESCRIPTION OF THE AFTI/F-16 DIGITALFLIGHT CONTROL SYSTEM (DFCS)

* The AFTI/F-16 Digital Flight Control System (DFCS) is a full Fly-By-Wire (FBW) system with no
provision for mechanical back-up. Advanced digital computational technology, electronic mechani-
zation, and control theory are used to eliminate mechanical type input signal paths in all axes. From
Figure 23 it is evident that no mechanical linkages or control cables are used between the cockpit
controllers (side-stick, rudder pedals, flight control panel, etc), the control surface and speed brake
actuators, and the leading-edge flop power drive unit. The AFTI/F-16 Digital FCS represents a
major technological advance over the current analog FBW FCS used in the operational F-16s. A
DFCS permits the pilot to choose from a "menu" of multiple, selectable tasks/functions in order to
optimize the flight control task as a function of flight condition, aircraft configuration, and mission
segment. This fact is illustrated in the block diagram of Figure 24.
Basically the AFTI/F-16 DFCS is configured as a triplex-redundant system with three independent
* - - electronic branches. The prime electronic assemblies are the three identical digital Flight Control
Computers (FCC's) which perform the functions listed in the schematic of Figure 25. The triplex
aircraft motion sensors (rate gyros and accelerometers) provide feedback stabilization and aircraft
state information to the FCC's. Quadruplex redundant pilot sidestick and rudder pedal force sensors
accept pilot command inputs and transmit them in the form of analog voltage signals to the triplex
FCC complex for processing. The reason for using quad-redundant force sensors will be explained
later.

The DFCS is protected through automatic failure detection via voting, in-line monitoring, and
analytical redundancy techniques, failure isolation, and subsequent system reconfiguration. The
triplex DFCS is able to provide two fail-operational performance (within a 95% probability) follow-
ing two FCC failures.

Figure 25 illustrates schematically the various inputs to the triplex DFCS complex, the values of
which are used by the stored control laws to compute control surface deflection commands which
are transmitted to the seven Integrated Servoactuators (ISA) via triplex electrical signal paths.

The AFTI/F-16 aircraft is a highly modified version of an F-16A from which the analog FBW type
FCS has been removed and replaced with a triplex DFCS. In the design of the F-16 series aircraft
advanced flight control technologies were allowed to impact the design process at the beginning.
One advanced flight control concept, in particular, that of Reload Static Stability (RSS) had a very
large impact on the F-16 design in regard to its resultant aerodynamic configuration, size, weight,
range, performance, and thrust to weight ratio. It was the use of RSS that resulted in the selection

36
NADC822 4OL
60

CO
L-

LL

E.

Qf 3

a u 0

-C

\LLL

Cd,

L.

WX ail

IE-

37,
 4 NADC-82240-60

Lu

oJ 0

U. cc

LU,

Lu 0 U,

Z;L 6 9-
LU =U
z .
04-= 9- L ui
C.3LU U. cz 0 L.
.. 9-0 0 LU 0

LU Lu ()

I- L.

LU -Ol -LU U.
- 09- .jCJ C

U.. -0w* 0

CL
LL CCI

1-0 0)
_ M
0J
U3 k
S L Cm__ LL

* ~ e. -

LI x

U.
4 ~ GLLC
0.
- ca
CZI

zu cm
.j LU

LUYa U.

* 38
 NADC-82240-60

4 U.

D L+

o0 0 0 0 >J

< z'j ui L w w wJ
0t =) cc U)
a0
O 0 w0 oW uj ww
ow i- U w O tb~
~~~~~
~ . r 4 4 z4

xa w LL/ > . w W. U U,

UO U-
0T 0
cca zz
>.
w u U-

cc-u z
0 0: -C

U z 0 U~ 4-<
Ei 2 -i ,

CL L 0 0 zo

0 0 * 0000CC0t- C

u -) o c L UZU
-0. -J

u) LU U4 w

z U-
-- j

0n 0r cn cn >

o~ LUL.~ J)
-'2 0 Cdi o
D

0~~co0 -~ 4
U a.) Ct . N

0 >
~
cr ~ U-
wJ ww
cc 2r L 0

0.UO UJ U >0w z w c
a z W4
> u _j Z C.I -
-cc
- Lu '
65 c -j44

4u - xL ) 39J u
 NADC-82240-60

of an analog type FBW FCS. Incorporation of the RSS concept has been justified by its contribu-
tion to the performance, maneuverability, and range of the F-16. In subsonic flight the horizontal
tail (H.T.) is lifting and in supersonic flight the H.T. experiences a down load. RSS permits this
down load to be reduced. Reduced H.T. trailing edge up deflection provides the following benefits:

" Increased aerodynamic lift

0 Reduced trim dre.

These benefits in turn result in:

* Reduced wing area requirement which in turn allows a reduction in structural weight and
- skin friction drag.
0 Reduced fuel requirements for a given range, or increased range for a given amount of fuel.
Three basic systems comprise the FBW type FCC's of the F-16 series aircraft and are listed as
follows:

1. Primary Flight Control System (PFCS)

Provides three-axis flight path control through the deflection of the primary control
surfaces.

2. Secondary Flight Control System (SFCS)

Provides high lift, aerodynamic braking, and improved maneuvering performance through
the deflection of lift and drag modulation devices.

3. Air Data System (ADS)

Provides aerodynamic intelligence for the FCS and other using systems. Sensed aero-
dynamic parameters are inserted in the flight control law equations for computation by
the FCC's.

The optimal integration of these three systems results in FCS's that possess a variety of advanced
capabilities and unique features. Those common to both the analog and digital FCS's will be listed
initially in Table 3. The additional capabilities and unique features of the AFTI/F-16 DFCS will
be listed subsequently in Table 4.
The basic FCS functional characteristics of the operational F-16's with analog FBW systems are
presented in Table 5 (obtained from Reference 9) and are catagorized into the following areas:

* Primary Control Surfaces

4 * Secondary Control Surfaces
0 Trim
. Pilot's Primary Controls.

The data in Table 5 is also applicable to the AFTI/F-16 DFCS and provides insight as to the pilot
controller inputs used to command the required combinations of control surface deflections to
create pitch, roll, and yaw moments and to trim the aircraft. However, some of the quantitiative

40
 NADC-82240-60

TABLE 3. CAPABILITIES AND FEATURES COMMON TO BOTH THE

CURRENT F16 AND THE AFTI/F-16

." Full FBW FCS provides maximum flexibility for optimizing flying qualities and the
flight control task over the entire flight envelop.
- Use of the RSS concept with its many benefits.
• Three-axis Command Augmentation System (CAS) and Stability Augmentation
System (SAS) provides precise control and excellent handling qualities.
* Automatic angle-of-attack (AOA) and r )rmal load factor limiting allow the pilot
to use maximum capability of the airplane without inadvertent loss of control.
* Aileron-rudder interconnect provides improved high AOA handling qualities.
* Combined programming of the leading and trailing edge flaps provides optimized air-
foil camber for high L/D during cruise and maneuvering flight.
. Effective leading edge and trailing edge flaps provide good take-off and landing
performance.
* Effective speed brake provides good speed control in maneuvering flight.
- Redundancy of components and signal paths contributes to high probability of
mission completion and increased flight safety.
• Pilot's side-stick controller enhances precise control at high load factors.
* Built-in test capability ensures FCS flight readiness with minimal downtime for
maintenance actions.

TABLE 4. CAPABILITIES AND FEATURES UNIQUE TO

THE AFTI/F-16
* Six independent degrees of freedom flight control capability over the entire flight
envelop.
" The following advanced longitudinal control modes:
- Direct lift (direct control of An)
- Fuselage pointing (al mode)

- Vertical translation (cx2 mode)

* The following advanced directional control modes:
- Direct sideforce (direct control of Ay)
- Fuselage pointing (t 1 mode)

- Lateral Translation (3 2 mode)

pI4

4i 41
 NADC-82240-60

TABLE 4. CAPABILITIES AND FEATURES UNIQUE TO

THE AFTI/F-16 - Continued

0 Maneuver enhancement/gust alleviation to provide the following benefits:

- Automatic direct lift through force stick error signal

- Quickened pitch response

- More precise target tracking
- Reduced aircraft response to random turbulence and gusts
-- Elimination of "g" lag
0 Digital type FCC's permit optimal integration of flight, fire, and propulsion control
technology advancements.
. Use of in-flight integrity management to improve flight safety.
- * Automatic weapon line pointing.
* Reduced redundancy levels to reduce costs, spares, weight, space, and power require-
ments and retain typical FCS high reliabilities.
* Integrated flight controls and Flight Management Display System ,FMDS) to reduce
pilot workload and simplify cockpit layout.

values in Table 5 will not be exactly the same as for the AFTI/F-16, The table provides no data on
the canards of the AFTI/F-16. The use of the canards to enhance multi-mode capabilities will be
discussed later,

1. Primary Flight Control System (PFCS)

The PFSC interfaces directly with the Secondary Flight Control System (SFCS) and the Air Data
System (ADS). The interface with the ADS is shown in Figure 26.

Longitudinal or pitch control is achieved through fore-and-aft forces applied to the Side Stick
Controller (SSC) in addition to the SSC pitch commands the following signals also influence pitch
control:

* Angle-of-attach (c)
- * Dynamic pressure
* Ratio of dynamic pressure to static pressure (qc/Ps)
0 * Pitch rate
* Differential commands to the horizontal tails for augmenting roll control
. Pitch trim
a Normal acceleration
* Autopilot inputs

* 42
 NADC-82240-60

TABLE 5. F-16 FLIGHT CONTROL SYSTEM FUNCTIONAL CHARACTERISTICS

PRIMARY CONTROL SURFACES

AXIS LONGITUDINAL LATERAL DIRECTIONAL

Type fSufaceHoriz Tail

Type of Surface Horizontal Tail Flaperon Asymmetrical Rudder

Deflection (Max) 2±5 Deg (Min.) -*30Deg

(I to Hinge Line) Deg +20 to -23 Deg -10 Deg (Max.)
±25

.', Not Lad)

SurMa,
(Max,e ReLit 60 DeglSec 52 DegiSec ±60 Deg/Sec 120 Deg/Sec @0 HM

Stall Hinge Moment 15,850 Ft-Lb 14,400 Ft-Lb 4,700 Ft-Lb @ 12 Dog
(Both Hyd Systems) 14,400 Ft-Lb @0 Deg @20 Deg @ 0 Deg
i ~~~Stl nge
yd 7,20m
@0 DeFt0 e-L Og 235 tLb@1 o
Stall Hinge Moment 7,200 Ft-Lb 7,925 Ft-Lb 7,200 Ft-Lb 2,350 Ft-Lb @12 Deg
(One Hyd System) @0 Deg @20 Deg @0 Deg

SECONDARY CONTROL SURFACES

TYPE OF LEADING EDGE FLAPERON SPEED BRAKE

SURFACE FLAP
Deflection (Max) -2 to+20 FlapDeg
to-23
(Streamwise) -2teamose++25to-23Deg0 (T.E. Flap) to 60 Deg

Surface Rate Limit 30 Deg/Sec

(Max, No Load) 30 Deg/Sec @0 H.M. 5 Deg/Sec (flap function) (Per Surface)

Stall Hinge Moment 175,000 In-Lb @0 Deg 15,850 Ft-Lb @20 Deg Single System Only
(Both Hyd Systems)
Stall Hinge Moment 87,500 In-Lb @0 Deg 7,925 Ft-Lb @20 Deg 6,065 Ft-Lb/Surface
(One Hyd System)

Command Servo Electrical FeedbackNoeoe

Dual Position & Rate None None

TRIM
AXIS LONGITUDINAL LATERAL DIRECTIONAL

Type of Surface Horizontal Tail Flaperon and Ratio

Horiz Rudder
Tail Variable

Auht
Ahia ±2.4 G's (Surface
Function Deflection
of Variable Gain)
t.90 O's/Sec from t4.5 Deg/Sec/Sec from Set by Rate Pilot
Trim Rate Pilot's Controller Pilot's Controller Turns Knob on Panel

PILOT'S PRIMARY CONTROLS

4 AXIS LONGITUDINAL LATERAL DIRECTIONAL

Type Pilot's Sidestick Controller Pilot's Sidestick Controller Rudder Pedals

Displacement Minimum Minimum -0.5 In.

Breakout 1.75 Lb ±1.0 Lb -15.0 Lb

Max Command Max Roll Command

Maximum Force (+10.86 G'sat 40.0 Lb Aft) Max Surface at 110 Lb
(-4 G's at 18.44 Lb Fwd) (308 eg/Sec at17.0 Lb)

.4 43
 NADC-82240-60

LL

z 4UIIIHS
lIIIII 4-

oZ C.,

u- \-a L
o m c
0 -
z 9L .

hCL.

-cL a.

* N I44
iNADC-82240-60
I

The pilot applies pitch trim by displacing the four-way trim button on the grip of the side stick.
Fore and aft displacements of the trim button drive an electrical gear-head motor that is located
in the manual trim panel and connected to a stack of three potentiometers and a thumb-operated
pitch trim wheel. Triplex trim signals are generated, summed with other pitch command signals,
gain-adjusted, and sent to both the left and right horizontal tail integrated servoactuators (ISA).

Lateral or roll control is achieved through lateral forces applied to the SSC. The roll axis employs
both CAS and SAS as an aid to achieving precise flight path control. Roll CAS is a roll-rate-referenced
command system capable of minimizing the roll response per unit of lateral side stick force. Roll
command at any instant is a function of the following:

e Lateral side-stick force

* Roll rate
* Roll trim
* Autopilot commands

Both flaperon and differential horizontal tail deflections are used for roll control to assure that
adequate roll control power is available throughout the flight envelope.

Roll trim is achieved in a manner similar to pitch trim. Roll trim is accomplished by lateral dis-
placement of the four-way trim button on the grip of the side stick controller. These lateral trim
*commands drive an electric gear-head motor in the manual trim panel. Triplex trim signals are
generated, summed with other signals and gain-adjusted to provide a variable trim distribution to
the flaperons and the horizontals tails in accordance with the horizontal tail-flaperons ratio
schedule.

Directional or yaw control is achieved by the use of the rudder pedals. The yaw axis employs
basic stability augmentation for yaw damping and automatic turn coordination through the use
of lateral acceleration referenced command augmentation. A gain adjusted aileron-rudder inter-
connect (AR I) and a roll rate-to-rudder cross feed minimize roll-yaw coupling, extend the AOA
for departure boundary, improve tracking, increase spin resistance, and provide improved turn
coordination. Rudder commands at any instant are a function of the following parameters:

Rudder pedal force

e Yaw trim
* Yaw rate
* Lateral acceleration

The pilot trims the aircraft in yaw by rotating the yaw trim knob, on the manual trim panel, which
is connected to a stack of three potentiometers which generate triplex yaw trim signals. These
signals are summed with other signals in the FCC's and sent to the rudder ISA.

2. The Secondary Flight Control System (SFCS)

The Secondary Flight Control System (SFCS) consists of the following elements:

* Full-span leading-edge flap (LEF).

* Flaperons functioning as trailing edge flaps (TEF).

4 45
 NADC-82240-60

e Speed brakes located on each side of the aft fuselage.

e Canards functioning as speed brakes by deflecting the leading edges of each inward (toe-in).

The LEF is scheduled as a function of Mach number and AOA with the scheduling being based on
performance improvement and buffet reduction rather than on stability and control requirements.
The LEF provides the following benefits:

e Increased lift for take off and landing.

e Automatic and near optimal wing camber as a function of flight condition thereby providing
improved turning capability and increased effectiveness of the vertical fin at high AOA.

TEF flap and roll commands are summed in the FCC's to produce a desired flaperon deflection.
In the F-16 FBW systems no additional components are required to accomplish the TEF function
except for the electronic components required to accept the flap command in the FCC's. The
absence of a flap command to the FCC's results in a zero flap command to the flaperons.

The speed brake consists of two pairs of clamshell surfaces located adjacent to the engine nozzle
and inboard of the horizontal tail surfaces in the trailing edge of the left- and right-hand horizontal
tail shelf structure. See Figure 27 obtained from Reference 9. Each speed brake consists of the
fol;,wing:

* A double acting hydraulic actuator.

* An upper door with linkage.
o A lower door with linkage.
a Three hinged seals that are activated by the motion of the speed brake.

Figure 27 shows in detail the mechanization of the speed brakes, location, and maximum available
deflection.

AFTI/F-16 AIR DATA SYSTEM (ADS)

The configuration of the AFTI/F-16 Air Data System is shown in Figure 26. This configuration
is basically the same as that of the F-16 with the exception of the addition of two sideslip probes,
which are identical to the AOA probes, and repositioning of the right-side air data probe to eli-
minate canard interference. The Electronic Component Assembly (ECA) used in the F-16 ADS
is not required in the AFTI/F-16 ADS because its functions can be performed in the digital FCC's.

The AFTI/F-16 ADS is comprised of the following units:

. Nose mounted air data probe.

* Fuselage side-mounted air data probe.
" Two flush fuselage-mounted static pressure ports.
* Two radome mounted AOA transmitters.
- Nacelle mounted total temperature probe.
* Pneumatic Sensor Assembly (PSA)
4

* 46
. NADC-82240-60

UT PER
LINK

" / 60-DEC UPPER &

LOWER DOOR FULL
EXTENS ION

"PPER CLOSURE SEAL UPPER

!'. " ?,/zDOOR

ACTUATOR

DOOR

F.S. 502.25
LOWER
LOWER LINK
CLOSURE SEAL

Figure 27. Mechanization of the Speedbrake

* Central Air Data Computer (CADC).

* Sideslip differential pressure sensor.
* Two Angle-of-Sideslip (AOS) transmitters.

Three independent sources of total pressure (PT),static pressure (Ps), AOA, and AOS are sensed
and converted to electrical signals. Middle-value selection of the air data parameters are employed
to provide triplex identical signals to be inputted to the FCC's for control law computation, gain
ga scheduling, and LEF computation.

1. Air Data Probes-

The nose mounted in data probe provides dual sources of static 'Ps1 and PS2 and total
* ~PT, and PT2) pressures. Aerodynamic compensation is provided to minimize the static
pressure source position error. The multi purpose fuselage air data probe (also referred

o6 47
 NADC-82240-60

to as the side-mounted probe) provides a single source of static and total pressure and
AOA. Aerodynamic is provided
compensation probe to minimize the static pressure source
rposition error. This multi-purpose provides five pneumatic outputs (PT3, PS3 ,
P011, Pa 2 , and P3) as inputs to the Pneumatic Sensor Assembly (PSA) where pressure
transducers corvert pneumatic signals to electrical analog signals. The method of sensing
AOA is the same as that described for the multipurpose probe shown in Figure 3. The
AOS ports are pneumatically connected together to provide a pseudo-static pressure
(P ) reference.

A nacelle mounted total temperature probe provides the Central Air Data Computer
(CADC) with a total temperature (TT) signal (analog). This signal is required by the CADC
for true airspeed and air density ratio computations. The probe is installed on the left
side of the engine nacelle so that the intake of the total temperature probe is located
outside of any boundary layer and not in the wake of any upstream portion of the aircraft.

2. Static Pressure Ports

*Two flush-mounted static pressure ports provide inputs to a Sideslip Differential Pressure
Sensor for AOS measurement. This measurement is used to compensate the third AOA
source (Side-Mounted Air Data Probe) error. The flush ports are located diametrically
opposite to each other and forward of the cockpit.

3. Sideslip Differential Pressure Sensor

This sensor is installed remotely to the PSA to provide an AOS function to the digital
Flight Control Computers (FCC). The sensor output is a function of AOS and dynamic
pressure.

4. Angle-of-Attack (AOA) Transmitters

Two conical air-flow-detector type AOA transmitters are mounted diametrically opposite
on each side of the radome. Each transmitter contains three identical rotary voltage
differential transformer (RVDT) outputs, one output to each of the FCC's (see Figure 26).
These transmitters convert conical probe rota ion (a function of airflow striking probe
slots) to electrical analog signals.

5. Angle-of-Sideslip (AOS) Transmitters

These units are AOA transmitters mounted on the underside of the fuselage in a manner
to sense AOS. Their triplex outputs are sent to the FCC's (see Figure 26).

6. Pneumatic Sensor Assembly (PSA)

The PSA contains eight pressure sensors that convert pneumatic inputs from the nose
4" and fuselage mounted air data probes into electrical signals. The differential pressure
type AOA signal is normalized to dynamic pressure at the hemispherical head of the
side mounted air data probe. This dynamic pressure is provided by subtracting the
average AOS pressures (psuedo-static pressure) from the total pressure at the hemispherical
head. Normalization (division) is achieved by use of a multiplier as a feedback element
in an operational amplifier c;rcuit. The PSA supplies triplex static and dynamic pressure
* signals to the FCC's.

*l 48
 NADC-82240-60

7. Central Air Data Computer (CADC)

The CADC is a digital computer which accepts inputs from the nose-mounted air data
probe, total temperature probe, and the two AOA transmitters. The CADC consists
of the following six basic functional sections:

1. Power Supply
2. Pressure Sensors (2)
3. Input Data Converter (IDC)
4. Digital Processor (DP)
5. Output Data Converter (ODC)
6. Chassis Assembly

The values of PT, and PS, are inputted into the Input Data Converter (IDC) via two
pressure sensors and the barometric, AOA and total temperature signals are inputted
directly to the multiplexer and A/D converter section of the IDC. The IDC converts
the data from these inputs to digital form and transmits them to the d;gital processor
(DP) section for processing. The input data is corrected for source errors in the com-
puter. The required CADC outputs are calculatcd, formatted, and transferred to the
output data converter (ODC) section. The ODC converts the data to the appropriate
signal form (digital, analog, synchro, discrete) and distributes the following air data
parameters to the using systems:

- Pressure altitude (Hp) (referenced to 29.92 in Hg)

* Pressure altitude (HpBC) (referenced to altimeter barometric setting)
- Pressure altitude rate (IP)
• Pressure ratio (Ps/PsL)
. True temperature (TFAT)
- Calibrated airspeed (VC)
- True airspeed (VT)
* True angle-of-attack (OL)
* Air density ratio (a)
8. Air Data System Redundancy Management

All air data parameters essential for flight control law computation are inputted to the
FCC's to a triplex level of redundancy. Redundancy management (failure detection,
isolation, and reconfiguration) of these inputs are performed by the FCC's. Initial
failures are detected by middle value voting techniques and the failed input is isolated
thereby permitting single-failure operational capability. After the occurrence of a
second like failure it is not possible to isolate the failed input by using voting techniques
and the data from two sensors is ignored. The FCC's will then employ a backup set of
flight control laws that provide alternative control laws that provide at least Level 2
flying qualities with a goal of level 1.

-4 49
 NADC-82240-60

After the failure of a second AOA sensor the AOA feedback will be removed and not used
as an aid in the computation of control laws. In the cases of dual failures of the other air
data sensors fixed gains will be used in control law computation.

FLIGHT CONTROL COCKPIT CONTROLS

The AFTI/F-16 flight control cockpit controllers consist of a right side mounted Side-Stick Con-
troller (SSC) Assembly, a Rudder Pedal Assembly, an FCS Control Panel, and a Rotatable Throttle
Controller Grip for pitch pointing control.

1. Side-Stick Controller (SSC) Assembly

The SSC is equipped with physically and electrically isolated transducers so that applied
forces (resulting in only small total displacements) to the stick grip will generate quadrex
electrical command signals. These signals are summed with other signals, gain-adjusted in
the FCC's, and transmitted to the primary control surface integrated servo-actuators.
Linear variable differential transformer (LVDT) type transducers are used in these appli-
cations.

The pilot achieves longitudinal pitch control through fore-and-aft forces applied to the
minimum-deflection type force-sensing controller. In a similar manner, the pilot achieves
lateral or roll control through lateral forces applied to the SSC. Figure 28 shows the
installation and assembly of the SSC for the F-16 and was obtained from Reference 9.

TOP
VIEW ___

44

1 GUN/CAMERA Trigger (2 position) 7

2. NWS/A/R DISC/MSL STEP Button
3. DESIG;RET SRCH Switch (2 way, Momentary)
4. TRIM Button (4 way, Momentary)
5. WPN REL Button
6. Side Stick Grip
d 7. Paddle Switch

Figure 28. Installation and Assembly of the Side-Stick Controller (SSC) for the F-16

450
q NADC-82240-60

The AFTI/F-16 SSC grip is the same as the F-16 grip except that two switch housings have
been added, one 4-way slide switch is located where the F-16 grip has a 2-way slide switch,
and since the autopilot disconnect is not required, the space is used for the Integrated
1 Flight and Fire Control (I FFC) switch.

"* The switch housings were added to provide room for three new redundant switches. The
Independent Backup Unit (IBU) and Control Configured Vehicle (CCV) switches are
located in the top housing. A flip-over paddle was added to the top housing to provide
switch actuation for the CCV switch. The IFFC switch is located in the lower housing
on the aft side of the stick and is operated through the paddle located at the lower forward
side of the stick.

The new 4-way slide switch was added to provide the Helmet Mounted Sight (HMS) switch-
ing feature. The switch is located in the same position as the 2-way DESIG-RET SRCH is
located on the F-16 grip. The HMS position is on the inboard side. A space position is
on the outboard side.

The trigger switch is the same as for the F-16 grip except the name has been changed from
CAMERA/GUN to RCD-LASER/GUN. The descriptions of the AFTI/F-16 grip was obtained
from reference 10. Figure 29 shows the AFTI/F-16 Side-Stick Controller (SSC).

The AFTI/F-16 SSC will be essentially two-fail operative. Quadrex force transducers in
both the pitch and roll axes are used as an aid in detecting and isolating single and dual
failed force transducers. Figure 30 is a cross sectional view of the transducer portion of
the SSC showing the locations of the pitch and roll LVDT's, four per control axis.
i IBU
;- TRIM

'-'.WPN REL

DES/RETSRCH/HMS

RCD-LASER/GUN

CCV

NWS
- MSL STEP
A/R DISC

Figure 29. AFTI/F-16 Side Stick Controller (SSC)

51
 rrrr - -7

I NADC-82240-60

KNURLED NUT ELECTRICAL CONNECTOR

TUBE DIAPHRAGM

ELECTRICAL HARNESS

PROVISIONS FOR
TUBE ASSEMBLY \\MECII BREAKOUT

ROLL LVDT 1 I
(4) TYP CIRCUIT BOARD

ACCESS FOR PITCH /'

LVDT ADJUSTMENT

* ~ROLL FLEXURE BEAMS ) V

PITCH LVDT
_ (4) TYP
STOPS (4) >

FLEXURE BEAMS
HOUSINGPITCH

4 Figure 30. Cross-Sectional View of Transducer Portion of the Side-Stick Controller

52
 NADC-82240-60

. 2. Rudder Pedal Assembly

The rudder pedal assembly for the AFTI/F-16 is the same as for the F-16. This assembly
consists of a minimum-deflection force-sensing pair of conventional rudder pedals. The
pilot achieves directional control through translation of either pedal (approximately
+0.5 inch) to generate quadrex electrical signals using an LVDT type transducer. These
signals are summed with other gain-adjusted signals in the FCC's and transmitted to the
rudder ISA. Rotation of either pedal generates dual electrical brake command signals
from two LVDT type transducers. The rudder pedal assembly is also used for nose wheel
steering.

The rudder pedals are used for yaw pointing of the aircraft when the FCS is configured
to operate in one of the pilot selected advanced control modes. Figure 31 illustrates the
configuration of the rudder pedal assembly.

The AFTI/F-16 Rudder Pedal Assembly is also two-fail operative. Quadrex force sensors
are used as an aid in isolating single and dual failed force transducers.

3. Flight Control Panel

The Flight Control Panel Assembly is located on the pilot's left hand side as shown in
Figure 32. It provides a redundant or backup set of trim and reset switches. The
components on the panel are:

Figure 31. Rudder Pedal Assembly

53

- -- -- - - - - -
 NADC-82240-60

A"Mzu ot MeJ eg~cool

N" a
'"mFof
w4101
Low"" Trgl Switchspr no ad
141
LODToPa tver ButinCo
nelA ntroe

K~t" THat'eSSwttch Snring-Loade

White~~T
Button.n MomentarCononletisDiabes
Rests
ontolll ligt N--Tri S itc on t lri
IntgrtedSevoctuto Paelan Sienol
FAlursWihaRstal otolraeAtw

Fiur 32llghootrlPneontalto

TR4
 NADC-82240-60

- Redundant Trim Switches (3 position toggle)

1. L YAW
R YAW
2. L WING DN
R WING DN
3. NOSE UP
NOSE DN

- Fault Reset Switches (pushbutton)

1. Electronic
2. Servo

* Redundant Stick Trim Switch

(2 position toggle: OFF-ON)

When the Stick Trim Switch is positioned to "ON", trim functions can be controlled from
the stick trim button on the 6SC or from the Flight Control Panel. The "OFF" position
of the Stick Tr;m Switch deactivates the SSC trim button so that trim can only be con-
trolled through the Flight Control Panel trim switches. The backup trims are required
*. because the SSC trims are nonredundant. The panel trim funct ons are implemented as
triply redundant discrete inputs to the FCC's where the trim integration function resides
in software. The reset buttons are included on the panel to permit "blind" failure resets
for those rare instances when both Multi-Purpose Displays (MPD) are not operational.
The panel resets operate essentially the same as the MPD resets in the way they reset a
first or second like failure.

- The term second like failure refers to two independent failures which have been detected
by the same monitor and does not necessarily imply that the two failures were physically
identical. For example, given that the first failure was branch A pitch rate gyro, then a
second like failure would include a pitch rate input demodulator in branch B or C, an
A/D converter failure in B or C, an 800 Hz power supply in B or C as well as an identical
pitch rate gyro failure in branches B or C.

There are no limitations as to which failures may be reset or the number of times a reset
request may be repeated. The FCS Operational Flight Plan (OFP) presents a second
-failure from being reset back to the no failed state and creating the chance for an apparent
simultaneous dual like failure situation with potential loss of control. The price paid for
this added protection is that the first failure must be latched in permanently whenever a
second like failure occurs.

* 4. Rotatable Throttle Controller Grip

Rotation of the Rotatable Throttle Controller Grip provides the controlling inputs of the
following CCV modes upon pilot selection:
. Pitch Pointing
"* Vertical Translation
0 Direct Lift

* 55
KNADC-82240-60

The entire Throttle Assembly as shown in Figure 33, is located in the cockpit above the left console
and is operated by conventional fore and aft motion. The throttle grip rides on a spline shaft on
recirculating ball bearing with a total travel of 8.12 inches. Each end of the spline shaft is mounted
to aircraft structure with small brackets. The output of the throttle linkage is connected to an en-
closed push-pull control to transmit motion to the engine mounted fuel control.

5. Interface With the Flight Control Computers (FCC)

The redundant analog and discrete inputs from the cockpit controllers are supplied to each
FCC with the required cross-strap for redundant data selection implemented by a serial
digital data exchange between the FCC's. Analog/Digital (A/D) conversion of the inputted
cockpit controller signals takes place within the FCC's.

I
8.12

Figure 33. Throttle Assembly

. P5.
 ' NADC-82240-60
b -

FLIGHT CONTROL INERTIAL SENSORS

The AFTI/F-16 Primary FCS inertial sensors are contained in Rate Gyro Assembly and an Acceler-
ometer Assembly which are identical to those used in the production versions of the F-16. These are
quadrex units designed for use in the quadrex analog type FBW system for the F-16. For use in the
AFTI/F-16 only 3 of each of the 4 like sensors will be interfaced with the triplex DFCS.

Dual like inertial sensor failures will not produce a loss-of-control condition. The second like failure
will be detected by cross channel monitoring, and a "control law reconfiguration" scheme will be
used to maintain safe flight.

1. Rate Gyro Assemblies

Three Rate Gyro Assemblies (one each for pitch, roll, and yaw) consist of four rate gyros
each to provide redundant signals as a function of body rates about their respective axes.
For the AFTI/F-16 one rate gyro in each assembly will not be interfaced with the triplex
DFCS. Figure 34 shows the installation of the Rate Gyro Assemblies. The Rate Gyro
Assemblies are identical for each of the three locations and do not require separate stock-
ing for three different applications. Spring-balanced, floated rate gyros are used; each rate
gyro contains a permanent magnet torquer for self testing and a monitor of spin-motor
lead-phase voltage characteristics. Each rate gyro subassembly contains a phase-splitting
capacitor potted in the head pin end), a cable, and external connector.

// /

• Y' RAM- ; , (

POLL
RAT
v t

II

Figure 34. Installation oi Lhe Rate Gyro Assemblies

57
[, NADC-82240-60

2. Accelerometer Assembly

The Accelerometer Assembly consists of eight identical torque-balanced accelerometers

fastened to a common mounting base and enclosed with an aluminum dust cover. Each
accelerometer is entirely self-contained, analog closed-loop, sensing transducer having
all solid-state silicon electronics. Four of the accelerometers are aligned within the
assembly to sense acceleration along the normal axis and four for sensing lateral accel-
eration. For application to the AFTI/F-16 one of the lateral and of the normal accel-
erometers will not be interfaced with the triplex DFCS. The installation of the
Accelerometer assembly in the aircraft is shown in Figure 35. The location of the
installation is consistent with requirements to yield minimum body-bending effects and
maximum aerodynamic stability. The physical center of the Accelerometer Assembly
is located forward of the aircraft C.G. as near as practicable to the center of
percussion.

For self-test purposes, an external electrical command is summed linearly in the torquing
coil of the accelerometer. The accelerometer mounting flange, located near the C.G.
of the assembly, contains four threaded mounting holes. Each of the four external
connectors is equipped with two split cables and end connectors. One split cable is
connected to a normal accelerometer; the other, to a lateral accelerometer.

3. Interface with the Flight Control Computers

•
*.i Figure 36 shows the method of interfacing or inputting the analog outputs of the flight
control inertial sensors into the triplex FCC complex. The output of each sensor is
directly inputted into one of the FCC's. This input is transmitted to the other two
FCC's via the intercomputer data link. All FCC's know the current inputted value
from each sensor and each FCC is capable of performing the failure detection, isolation,
and reconfiguration task.

4. Redundancy Management and Reconfiguration Technique

• Failure detection and isolation of the inputted inertial sensor data is performed at the
Input Voting/Monitoring Plane (see Figure 36) which is in the software programmed
into the triplex FCC complex. Redundant inertial sensor inputs are supplied to each
FCC with the redundant data selection implemented by a serial digital data exchange
between the FCC's. First and second like sensor failures are detected by cross-channel
monitoring techniques. After a sensor failure the average values of the remaining two
sensors are used by the FCC's. After a second like sensor failure no effort is made to
determine which of the two remaining sensors is failed and the remaining inputs of the
particular inertial quantity being sensed is ignored. The flight control laws which
normally utilized the values of the sensed inertial parameter are reconfigured.

The AFTI/F-16 is statically unstable subsonic flight and must employ artificial
*means to restore pitch static stability. Control laws employed to artificially restore
static stability require aircraft pitch rate as an input. To avoid aircraft pitch divergence
and subsequent loss of control after the loss of two pitch rate sensors a backup technique
of estimating pitch rate (q) is utilized. These estimated values are inputted into a set of
backup control laws. Estimation of pitch rate is based on AOA and horizontal tail
position.

l 58
 NADC-82240-60

ACCELEROMETER
,-. ~~AS
SE MB L Y -- --

e , .ELECTRICAL

- CONNECTOR (4)

Figure 35. Installation of Accelerometer Assembly

In case of dual loss of like inertial sensors that are only used to enhance performance the
reconfigured control law is the non-reconfigured control law with the affected feedback
removed. The intent is to provide alternative control laws that provide at least Level 2
flying qualities with a goal of Level 1. Inertial sensor reconfiguration strategy is summa-
rized below

-| 59
 NAD C-82240-60
z
<c/

zE
ui TT
I-6

44
wm
U.LL-LL
a-

wU _ _ (D_

000

0-0

ci L) 4-,
cij cic
LL

zz

< W-

0- 0 <I V

0 cc
CL CO

Lu >:

0 <
-j-c

0 >)

cc-c

60-
 NADC-82240-60

SENSORS RECONFIGURATION STRATEGY

AFTER SECOND FAIL MISCOMPARE

, Pitch Rate Gyro * Estimation of q is based on AOA and hori-

zontal tail position
0 Normal Accelerometer * Reconfiguration to pitch rate command
system

• Roll Rate Gyro . Remove feedback and adjust SSC gain

* Lateral Accelerometer * Remove feedback

* Yaw Rate Gyro * Remove Ay and 4 feedbacks, use only

the product of pa feedback to rudder

DIGITAL FLIGHT CONTROL COMPUTER (FCC) COMPLEX

The digital FCC Complex consists of three Flight Control Computers (FCC), an Actuator Inter-
face Unit (AlU) and four Inverter Control Assemblies (ICA)which provide AC power for the
sensors. The FCC's and AIU were designed and fabricated by the Bendix Corporation. The FCC
Complex provides the performance capability required of the FCS for the six DOF AFTI/F-16
airrcraft during both ground and airborne operations. The FCC's are advanred third generation
Bendix Model BDX-930 processors incorporating innovative redundancy management concepts.
The triplex FCC Complex is fully operational after a first FCC failure. The probability of main-
taining Operational State 1 performance after a second FCC failure is greater than 95 percent.
The three FCC's are physically and functionally identical and contain identical software.

The FCC's provide the computational power required to implement the AFTI/F-16 airframe
performance requirements. Signals from airframe sensors and indicators, and electrical power
are supplied to the FCC's to permit the required Stability Augmentation System/Command
Augmentation System (SAS/CAS) computations. Also supplied to the FCC's are inputs from
the pilot's controllers and, through the airplane multiplex bus, interactive mode selection and
failure annunciation and recovery. The AFTI/F-16 is controlled by the FCC's and the AIU
through appropriate commands to the flight control surfaces.

1. Interface Between FCC Complex and the FCS

The salient interface signals between the FCC Complex and the FCS are shown in
Figure 25, most of which being of the analog type. Input filtering is provided on
all analog inputs for high frequency noise suppression, and the I/O is protected
from EMI and lightning induced electrical transients.
4
The FCC Complex supplies certain instrumentation output signals, such as angle
of attack and sideslip, pitch, roll, and yaw rate, stick and rudder pedal forces,
and FCC and AIU temperature. All of these signals are protectively buffered so
that a failure external to the FCC Complex will not cause an internal malfunction.

'II 61
 NADC-82240-60

The FCC Complex interfaces with the avionics system via a serial-digital, dual-
redundant, time-division multiplex data bus as shown in Figure 37. The FCC
Complex responds as a remote terminal to commands received on this bus.
The FCC Complex is designed to provide a bus complement of 32 subaddresses
of 32 words each so that up to 1000 words of bus traffic is permitted. A
failure of an FCC will cause its bus transmitter to stop, and this action will
force the bus controller to transfer communication to the second bus and
another FCC. The bus interface is designed so that no failure external to the
FCC Complex will cause an internal malfunction.

2. The Flight Control Computer (FCC)

A layout/outline drawing of the Bendix BDX-930 FCC is shown in Figure 38.

The FCC contains a computer unit that contains a 16-bit paralleled twos-comple-
ment binary arithmetic unit and sufficient memory to perform the AFTI/F-16
FCS functions. The BDX-930 FCC is faster, smaller, and consumes less power
than its predecessors. It employs a pipeline organization feature to provide con-
current fetch, decode, and execute operations in order to increase the number of
instructions executed per unit of time. All major FCC elements are provided with
hardware error detection features to aid in the failure isolation and fault recovery
process.

AVIONICS MULTIPLEX BUS B

FLIGHT FLIGHT
CONTROL CONTROL
COMPUTER COMPUTER

SERIAL DATA
EXCHANGE

~FLIGHT

CONTROL
COMPUTER

Figure 37. Interface of the FCC Complex with the Dualized Avionics Multiplex Bus

62
 NADC-82240-60

EliZE
I 1' 191
at~

sC_

630
 NADC-82240-60

The Central Processor Unit (CPU) architecture employs an asynchronous memory

interface that allows the CPU (I/O) controller, avionics MUX bus, and inter FCC
data link from loading down the memory access bus.

The processor can perform single precision, double precision, and saturated
arithmetic operations. Saturated arithmetic capability minimizes fixed-point
-. arithmetic overflow problems. All memory is equipped with parity, and program
. memory is accomplished in EPROMS to preclude unintentional alteration. The
processor contains 22 registers which are usable by the programmer.
The FCC contains an Input/Output (I/0) controller that controls the I/O operation
of analog I/O, discrete I/O, and inter FCC serial data transmission. Branch failure
partioning will permit a CPU failure to cause the I/O controller to obtain analog
and discrete inputs and to transmit these data to the other FCC's so that input
redundancy is maintained. The analog I/O function is implemented with 12-bit
analog-to-digital (A/D) and digital-to-analog (D/A) converters.
The FCC contains a serial data transmitter and two receivers, which are used to
perform the inter-FCC data exchange. The bus format is one way communica-
tion with a Manchester bi-phase level code and is accomplished at a 1-MHz bit
rate. The transmitters and receivers contain hardware to verify the validity of
all messages and to aid in the failure isolation process. The bus structure will
support 128 messages with up to 7 data words each. Each receiver bus access
is limited to 1 K of RAM to prevent address errors from propagating into the
processor work area and destroying in-line data computation.
The avionics MUX bus interface employs a single transmitter receiver, which is
interfaced directly with RAM, with access limited in hardware to 1K words.
This limited access feature also prevents MUX interface failures from propagating
into the FCC. The interface contains a tag timer that can be reset by the bus
controller and read by the FCC to obtain a time correlated data set.

The FCC also contains considerable hardware logic, which is used in the fault
isolation, identification, and self-testing process. This logic includes hardware
that permits two good FCC's to "shut down" a third by mutual consent of the
two good FCC's without dependence upon any action of the third "failed"
- .FCC's processor or software (this is first failure logic). The first failure of an FCC
causes the issuance of a discrete, which enables (unmasks) a self-test interrupt
handling routine. When a second FCC failure is detected, the FCC that detects
the failure issues a "start self-test" interrupt, which causes the FCC's to execute
the self test routines in an attempt to isolate the problem. The first FCC to
declare a self-test failure is "shut down", and the system continues to operate
on the last good FCC (this is second failure logic). Logic is wired so the last
remaining processor will not be disconnected even if it also "fails" self test.
6During the early part of the flight test program, the DFCS will include a simple
triplex analog Independent Backup Unit (IBU) control system that will auto-
matically be engaged if both of the remaining FCC's either pass or fail self test.
This feature is authorized only if an IBU is installed, otherwise, the failure logic
described above will apply.

Ug 64
 NADC-82240-60

Each FCC contains a branch of analog electronics for the IBU function and is used
as a dissimilar backup to the digital computation function, and is used primarily to
protect against generic software faults in the system. This feature was added because
the purpose of the AFTI/F-16 program is to provide a test bed for new flight control
concepts and their integration into a total weapons system. Since it is a test bed, it
will undergo many software modifications. The IBU can be engaged manually with
a switch on the SSC. The IBU is also automatically engaged whenever a second FCC
failure can not be isolated by self test or when all three FCC's have been declared
failed.

The FCC's possess a Built-In-Test (BIT) capability for preflight and maintenance
testing. The preflight BIT is capable of isolating failures to a FCS line replaceable
unit (LRU) to a 95% confidence level. No special Ground Service Equipment (GSE)
is required and the need for pilot participation is minimized. Results of tests are
shown on the Multi-Purpose Displays (MPD). BIT pre-flight testing functions
cannot be engaged in flight. The BIT function tests all software, sensors, and
actuators. After the satisfactory completion of preflight BIT the cockpit con-
trollers and switches are tested as the pilot goes through the normal "sweeps"
through these devices while on the ground.

Automatic maintenance BIT functions are identical to those of the preflight BIT.
Test results are displayed on the MPD's which also provide a built in "multimeter"
function. Fifteen thousand test points are automatically checked during the
maintenance BIT procedure (5000 for each FCC). Manual test procedures can
be configured on previous failure histories. The BIT maintenance testing func-
tion cannot be engaged in flight. Table 6 lists the salient features of the FCC's.

TABLE 6. SALIENT FEATURES OF THE BENDIX BDX-930 FCC

* PROCESSOR
* 16-Bit Microprocessor
0 Pipeline Architecture
* 22 Usable Registers
* Single Precision, Double Precision
and Saturate Arithmetic
* 162 Microsec Benchmark Program
0 Expanded Instruction Set
* MEMORY
0 Asynchronous Memory Control
* 7k RAM Scratchpad
* 32k EPROM Program
* 64 Words EAROM - Failure Record
* ATTRACTIVE FEATURES
* Analog IBU - Independent of Digital Processor
* Self Contained DC Power Supplies
* Two Real Time Clocks
" 12-Bit A/D and D/A Converters

* 65
 NADC-82240-60

3. The Actuator Interface Unit

The Actuator Interface Unit (AIU) shown in Figure 39, houses those components
of the FCS that are not triply redundant. These components include the dual re-
dundant Leading-Edge Flap (LEF) interface, the fourth servo-amplifier for each
ISA, and several simplex-to-triplex Junction Box type of functions.

The caution, and warning, IBU engage, and the stall warning light drivers are
designed so that two discrete outputs are required for illumination. This feature
makes it extremely remote that the lamps will be inadvertently illuminated by a
failure.

The LEF circuitry contains a switching matrix that is controlled by the FCC's to
determine which FCC ;s driving the LEF. The electronics also include summing
amplifiers that close the rate and position loops around the LEF drive motors.
The rate and position information is also transmitted to each FCC for monitoring
purposes. Switches that deactivate either or both LEF drivemotors are also
contained in the AIU.

A switching matrix driven by discrete outputs from the three FCC's is used to
select which FCC drives the fourth servo amplifier for each ISA and to determine
which back-up ISA valve coil is driven by this amplifier. The valve current wrap-
around is supplied to all FCC's to permit monitoring.

The Junction Box functions of the AIU include the conversion of ISA pressure
switch positions and ISA positions from simplex to triplex form. The "pseudo"
position of each ISA is brought into the AIU where the signal is demodulated into
a DC signal and then sent to each FCC for ISA monitoring purposes. Figure 40
shows the integration of the AIU within the DFCS.

-6 66
V7~ ~ NADC-82240-60

PE

I C

4-

10 .0

s4-

leo

67~14
 NADC-82240-60

K0

CC-
00

U..

poc4-
0z

Z a,

4-
a: t'0

IZ

4j

68O
 i .., .
: . .-. , , . :. . -
-~ -
.- i. l . -.. - -
. . : 7

,4 NADC-82240-60

FLIGHT CONTROL ACTUATORS

This section describes the Integrated Servoactuators (ISA's) that drive the primary flight control
surfaces and the Power Drive Unit (PDU) assembly that actuates the Leading Edge Flaps (LEF).
The speedbrake actuators, part of the secondary FCS, do not interface with any other portion of
the primary or secondary FCS and therefore will not be described.

1. Integrated Servoactuators (ISA)

Each of the seven ISA's (Figure 41) accept electrical commands from the FCC's in three electro-
hydraulic servovalves (EHSV). These commands are converted into a power-ram position, which
then positions the respective flight control surface. Actual ram position is fed back to the EHSV's
mechanically to close the command loop. This mechanical feedback concept provides for improved
surface position resolution and allows optimum ISA packaging.

A schematic diagram of the ISA is shown in Figure 42 to illustrate the following functional
characteristics:

. A unique mechanical position anJ actuator rate feedback scheme combines the feedbacks into
a single input to EHS J.
e Three EHSV's are provided for redundancy purposes. EHSV SV1 and SV2 normally share
control of actuator position, while the third EHSV, SV3, is held in standby. Note that flows
from SV1 and SV2 flow to the larger piston, and its cha, ber, on the main control valve. The
flows from SV1 and SV2 are summed in the chamber.

-r

Figure 41. Typical ISA Used for Powering the Primary and Vertical Canard Control Surfaces

S69
 NADC-82240-60

ccc

0 IfIc

4Y L l

r - - - - - - - - - - - - --

00L

Ldd

cm0

uC

Go~

70.
 NADC-82240-60

* EHSV failure detection is accomplished by comparing E HSV first-stage pressures.

- Self-contained hydromechanical failure detection and correction logic is incorporated for first
failures of the EHSV's or for the hydraulic system. A first failure of SV1 or SV2 will cause
transfer of control to the standby EHSV SV3. A first failure of SV3 will lock the ISA on SV1
and SV2 control.
. Hydraulic system failure correction is given precedence over all EHSV failures. EHSV's SV1
and SV2 operate on one hydraulic system, and SV3 operates on the other hydraulic system.
One hydraulic system (System B) is connected to pressure P1 and R1 and the other (System A)
is connected to pressures P2 and R2 .
* Fail-safe capability is incorporated to allow the ISA to mechanically center upon receipt of an
electrical command to the fail-safe solenoids from an external electronic model and monitor
unit.
In the mechanization of its hydraulic-mechanical redundancy management, the ISA utilizes self-
contained failure logic to provide no less than one-failure-operative performance. With the aid of
FCC monitoring, the ISA provides two-failure-operative performance if one of the two failures is
electrical. By arming the outer loop monitor, the ISA provides fail-safe performance for multiple
failures. The voting procedure is described below:

1. Hydraulic System Failure Detection. Hydraulic system voting logic is built into the ISA.
SThis logic takes precedence over EHSV voting logic and will cause the ISA to select a good
hydraulic system.

2. ISA EHSV Failure Detection. Each ISA uses three EHSV's. Normally, two of these EHSV's
(SV1 and SV2) share control of actuator output position. If a failure occurs in either of
these EHSV's the third EHSV (SV3) will assume control of actuator output position.

3. Flight Control Computer EHSV Failure Detection. Each EHSV has two windings, each of
which has total EHSV command capability. These windings are driven by the FCC in an
active standby manner so that a servoamplifier first failure will be detected and isolated and
a spare servo,-nmplifier in the AIU will be activated. This procedure will prevent the needless
activation of the ISA self-contained voting logic and the loss of a level of redundancy in the
EHSV control network. Figure 43 shows the interface of an ISA with the FCC complex
and how a failed servoamplifier is bypassed and replaced with a spare servoamplifier in the
AIU.

4. ISA Position Versus Computer Model Position. Each FCC models the ISA and compares
the model output with the ISA position. If the two differ by a prescribed amount after a
time delay, then an ISA fail discrete signal is outputted to the AIU. This modeling and mon-
itoring is accomplished in software.

5. Mechanical Centering and Locking of a Failed ISA. Additional failure protection capability
is designed into the ISA so that it can be mechanically centered by bypass commands from
4il the FCC's. These commands cause all EHSV's to be bypassed and allow a self-contained
spring to mechanically command the ISA to a predetermined neutral surface position.
Centering can only occur if (1) the pilot has previously armed the outer loop monitor, and
(2) the outer loop monitor has determined that the surface position is in error. This action
prevents spurious failure. The ISA should be centered after failures of either SV1 or SV2,
and failure of SV3. A combined failure of SV1 and SV2 is tolerable provided that SV3 and
4l Hydraulic System A are operational.

4 71
 NADC-82240-60

"SENSORS FCC' SERVO AIMPS. E-SVs$

i.",AND (FOR ONE ISA)

PNSCONTROLLERS

Al

Fiur FCC a SAt F Cm

STFCASA S N S
I

SA C
"i' / FAILED "A

2 ji #SIGNAL
1 -- PATH " "ONE OF 7 SPARE
SA D
SA'$ IN AIU

2.Pwe rieUit(DU ssml

UAIU

.• . Figure 43. Interface of an ISA with FCC Complex

.-.. 2. Power Drive Unit (PDU) Assembly

The Power Drive Unit (PDU) assembly shown in Figure 44 is comprised of the basic hydromechanical
drive unit plus the following components:

* A hydraulic flow-control valve and manifold to drive both hydraulic motors.

S*e Flow limiters to limit speeds for low loads on the LEF.
. A redundant electromechanical actuator to provide a positioning command to the hydro-
mechanical drive unit.
a Position summing linkage.
9 Operational controls.

Two in-line hydraulic motors (one driven by each hydraulic system) power the LEF. These motors
are located on opposite sides of the PDU to provide maximum separation of the two independent
hydraulic systems. Two hydraulic motors, ganged together by a common shaft and pinion, drive a
bull gear. The bull gear, in turn, drives two output shafts; one on the left side, another on the right
side. The shaft on the left side contains gearing that drives a feedback arm through a 60-degree arc
of the electromechanical actuator's rotating crank.

The shaft on the right side contains an overtravel mechanism that allows one degree of additional
surface motion at each end of the operating stroke of the panel. After contacting the stop, the sys-
tem will come to a halt within two output shaft revolutions of the PDU. This safety feature provides
internal stops that prevent the LEF from being driven into hard structure in the event of a malfunction

72
 NADC-82240-60

RET "B"

F.i<FSS "B"
RET "A"

PRESS "A 1 ",C T I ON

CO .,DNSERVO - -

,, 7
-.- - r? .ACK
CASE DRAINS -- /.L',",ACE
POWER DRIVE 'NIT
1T RETRACT

SEAL DRAIN4S DIRECTION~

Figure 44. Power Drive Unit (PDU) Assembly

in either the control or drive portion of the PDU. The positive stops consist of a threaded nut,
which moves inboard and outboard as the shaft rotates, and brake discs held with bellville-type
washers. Should the PDU tend to overtravel, the traveling nut engages the spring washers and com-
presses them, producing a clamping action that stops shaft rotation and stalls the motor.

REDUNDANCY MANAGEMENT OF THE DFCS

The reliability and fail-onerational requirements for the AFTI/F-16 DFCS are listed as follows:

e Fail Operatiom., iequirements

e First Failure - Undergraded full operational performance
* Second Failure - Safe flight capability
- Fully operational (State 1) performance 95% probable

e Reliability Requirements
to Loss of Control - 1 in 107 flight hours
to Mission Abort - 1 in105

To fulfill these requirements the DFCS must possess redundancy management techniques capable of
providing optimum failure survivability via detection and isolation of failed components to the LRU
level and reconfiguring the remaining unfailed components to provide the maximum level of aircraft
safety and the highest probability of mission completion.

- To meet the fail-operational and reliability requirements the DFCS has three voting/monitoring
planes as shown in Figure 45. Two of these planes are in software and are at the sensor/controller
interface and the output flight control surface command interface. The hardware voting plane is
located internally to the ISA's and can be used to isolate failures associated with the FCC output
circuitr and ISA EHSV coils as well as internal ISA failures.

73
 NADC-82240-60

VOTER/MONITOR PLA NE VOTER/MON/TOP PLANE

FTAEROTWR
SO ZONE1 ZONE2 ZONE3 ZONE4

r7
CONTROLLERS FCs!SA 's

AIU TR4IITRPLAN

Figure 45. Redu,-ancy Data Management of Hardware and Software Functions of the
AFTI/F-16 DFCS
The general guidelines inevolving redundancy management techniques are listed as follows:

* No complete dependence for a FCC to recognize that ithas failed after performing a self-test
* routine.
* Minimize "nuisance" alarms by
- Placing maximum reliance on comparison monitoring
- Provide safeguards against transitory fai:ures
- Use self-test methods only when required
e Implement a safe, dependable failure detection and identification scheme.
o Provide a failure survival capability as good as the present F-16 analog FBW system.

The redundancy management technique are significantly different at each of the three voting/moni-
toring planes and will therefore be discussed separately.

74
RD-A128 624 NEU FLIGHT CONTROL TECHNOLOGIES FOR FUTURE NAVAL 1/3
AIRCRRFT(U) NAVAL AIR DEVELOPMENT CENTER WARMINSTER PA
AIRCRAFT AND CREW SYSTEMS TECHNOLOGY DIRECTORATE

EhIhhlllhlliE
UNCLASSIFIED W W KANIUKA SEP 82 NADC-82248-60 F/G 1/3 NL

EIIIIIEIIIIIIE
EhIh~llhhlI
sEonmoEll-mE./
ElllEEEEEE~hEE
EllIEllllEE~hE
I/IE-//I/I//I
 Lo 18 . 12.5

=Lo

.258

MICEROCOPY
RESOLUTION TEST CHART
NATIONALBUREAU 19
OF STANDARDS63-A
 NADC-82240-60

1. Input Voting/Monitoring Plane

The primary purpose of the input voting/monitoring plane is to detect and isolate failures associated
with the sensors, controllers, and input circuitry. Redundant analog and discrete inputs are supplied
to each FCC with the required cross-strap for redundant data selection implemented by a serial data
exchange between FCC's.
During normal (no failure) operation input data are acquired from pilot inputs, motion sensors, and
ISA position signals, at a rate of 256 samples/second. The remaining FCS cyclic input data are ac-
quired at a rate of 64 samples/second. These data are sequentially received by the Input/Output
Controller (IOC) of an FCC, deposited in an area of scratchpad Random Access Memory (RAM)
dedicated to the IOC, and simultaneously transmitted to the other two FCC's. The IOC's are micro-
sequenced controllers designed to operate independently of the FCC Central Processing Unit (CPU).

Whenever an FCC is declared failed, the redundancy management hardware issues discretes, indicat-
ing the failure, to the IOC of the failed FCC as well as to other branches. The discretes cause the
IOC of the failed FCC to execute its backup (free run) file and continue to supply input data to the
other FCC independently of any command from the failed FCC.

Each FCC performs input voting and monitoring on the redundant input data deposited in its
scratchpad memory by its IOC as well as by the IOC's of the other two FCC's. Two selector/
monitor algorithms are used to perform the input voter/monitor function. One for discrete input
and one for analog input data.

Discrete inputs are monitored before voting and one monitoring cycle is permitted (persistence)
before a disagreement is declared a failure. Discrete inputs are monitored and voted in digital packed
word form and each digital word may contain as many as sixteen discrete inputs. Most discretes are
monitored at a rate of 64 samples/second but discretes affected by mechanical contact bounce are
monitored at a rate of 13 samples/second.

The selector algorithm for discrete inputs performs a majority vote if the three discretes agree or if
* one of the three disagrees (a possible first failure condition). If a first failure is declared and the two
remaining discretes disagree, (a possible second failure condition), the selector algorithm selects the
last value used prior to the disagreement. If the disagreement cannot be resolved, a failure is de-
clared and the most safe state for the discrete is assumed and selected.

All analog input signals are monitored and selected. AOA and AOS signals are normalized and com-
pensated for errors prior to utilization.

A good channel average algorithm operates on all analog inputs, (except ADA and p; eumatic Beta
which use a modified median select algorithm) and selects the average value when signals from either
*- two or three good channels are present. If any signal does not track the other two sensor signals
within a band of tolerances, that sensor signal is not included in the average. If a sensor signal does
track for a number of consecutive frames, the failure management function is called to remove it
from the input sensor set. After a first analog input failure, the failure management function is
called to resolve any disagreement between the last two good channels (a possible second failure con-
dition). If the second failure can be isolated, the remaining good channel signal is selected. If the
second failure cannot be isolated, a reconfigured set of control laws is provided.

Reconfigured control laws are available for all an3log inputs except controller commands. When
reconfiguration is required for the pilot sidestick and rudder pedal inputs, the signal from the
appropriate fourth force transducer on these input devices is used.

6 75
 NADC-82240-60

The averaging process associated with the good channel average algorithm eliminates sensor state
differences. All inputs which either change rapidly or are significant contributors to gain and phase
margins are acquired at a rate of 256 samples/second to minimize data latency.

2. Output Voting/Monitoring Plane

The output voting/monitoring is accomplished by the ISA monitor and the LEF monitor. These
.• monitors employ a combination of FCC software and hardware to manage the output and system
interfaces.

The ISA monitor is designed on the assumption that all critical FCC and output electronic failures
can be detected by compiling the coil currents (obtained via analog wraparounds) which drive the
three EHSV's in each of the seven ISA's. Because of the number of ISA's and the complexity of the
monitoring scheme the ISA monitoring function is broken into a subframe ISA monitor which
executes every 8 milliseconds and a Frame Cyclic ISA monitor which executes once per frame (16
milliseconds).
* The subframe ISA monitor is given the task of comparing the coil currents of the primary ISA coils.
With no failures in the system FCC A drives the primary coil of SV1, FCC B drives the primary coil
of SV2, and FCC C drives the primary coil of SV3 with the average displacement of SV1 and SV2
actually driving the surface via the primary hydraulic system (see Figure 43). An erroneous com-
mand from an FCC or a failure of the output electronics (D/A, sample and hold, servoamplifier, or
primary coil) will cause a difference to exist between the three primary coil currents. Each FCC
(if not the cause for the difference) is capable of isolating the failure to a particular branch. If an
FCC determines that its coil current is erroneous the output electronics are disconnected and a
backup coil current is supplied using the fourth servoamplifier driven from a different branch. If
this procedure is accomplished successfully the other two FCC's allow the partially failed branch to
remain on line. If the branch with the failure does not disconnect the primary coil, the other two
branches will totally disconnect the failed branch. The task of isolating this failure to either a com-
puted output failure or an output electronics failure is left to the ISA frame cyclic monitor for
first failures.

The Subframe ISA monitor becomes quite complicated after a first failure has been detected and
isolated. The resulting system actions are dependent on the cause of the first failure (output elec-
tronics or erroneous FCC output), the cause of the second failure, whether or not the first or second
failure affected one or all coil currents, and which branches sustained the failures.

The Frame Cyclic ISA monitor is allocated the tasks of monitoring the backup (or 4th) ISA servo-
amplifiers, handling the automatic reengagement of primary servoamplifiers which have been taken
off line due to a transient condition, monitoring the ISA centering and pressure system discretes,
monitoring the computed ISA commands after a first output electronics failure, and controlling the
activation and computations associated with the ISA model which can be engaged after an ISA
mechanical or hydraulic system failure.
The LEF monitor performs cross-channel monitoring on the LEF commands, position transducer
signals, and LEF motur rates. The LEF monitor also contains a digital model of the LEF actuation
system. The digital model is modified to reflect full or half speed drive as a function of the LEF
failure state.

76

I
 NADC-82240-60

a. Failure Detection, Isolation, and Reconfiguration After Single and Dual FCC Failures

As stated previously each FCC receives the ISA commands of the other FCC's via the intercomputer
data link. Each FCC, by comparison monitoring of ISA commands decides if another FCC was
failed in any (or all) of its ISA commands. For the single failed FCC situation we will assume that
FCC A experienced a failure. FCC A is expected to detect its own failures by comparing actual ISA
position with that of the model ISA's, and shut itself down if these values disagree. FCC B and
FCC C monitor FCC A's ability to monitor itself.
If FCC's B and C detect a failure of one (or more) of FCC A's ISA commands, they wait to see if
FCC A identifies the failure and takes appropriate action.

If FCC's B and C receive intelligence via the inter FCC data link that FCC A has taken the proper
action, then FCC's B and C take no action.

If however, FCC's B and C do not receive intelligence that FCC A has taken the appropriate action
(in connection with an FCC A failure they have detected) then they assume that FCC A has failed
and should be switched out or isolated from the system. In this case, FCC's B and C send a discrete
to FCC A to indicate a failure. If both FCC B and C send such a discrete, then discrete failure logic
in FCC A switches FCC A out of the system. The failure discrete which FCC's B and C send to
FCC A is not associated with a particular ISA, but rather implies a failure of FCC A as a whole.

The sequence of events which take place after a second FCC failure (FCC B or FCC C) will now be
discusesd.

If both FCC's (B and C) detect a failure in a particular ISA command, then both FCC's revert to
previous value of the appropriate ISA command. Both FCC's will initiate self-test procedures. When
the "failed FCC" is identified it is switched out of control of the affected ISA.

If only one of the two remaining FCC's detects a failure, then both FCC's revert to previous values
for all ISA commands. Both FCC's initiate the self-testing procedures. When "failed FCC" is ident-
ified, it is switched out of control of all ISA's.

If both of the last two FCC's fail self-test, then the first to respond controls the FCC switching. The
self-test sequence is arranged in priority order.
If neither of the last two FCC's fail self-test, bt one FCC does not complete it (no pass or fail),
then it is designated the failed FCC.

If neither of the last two FCC's fail self-test, and no other information is available for the decision,
then one is arbitrarily designated the failed FCC based on a predetermined hierarchy.

If a second FCC failure is not isolated then the IBU is engaged. The IBU is also automatically engaged
whenever all three FCC's indicate that they have failed.

3. Output ISA Voting/Monitoring Plane.

The primary purpose of the ISA voting/monitoring plane is to detect and isolate mechanical and
hydraulic failures internal to the ISA. Since the ISA's are driven by a dual-redundant hydraulic
system, they are designed to provide a fail-operate capability. In order to provide this capability, a
first failure upstream of the ISA must be isolated before the ISA voter/monitor will detect this

77
 , - . .- • + + , ..

I NADC-82240-60

same failure. The ISA monitor will declare a failure in approximately 20 to 40 milliseconds; there-
fore, the upstream monitors must remove a failure before this time limit. Because of this time limit,
the upstream voter monitors will remove the failure on the first detection and will initiate the
proper corrective action but not declare a failure for several cycles. This latency is desired because
it reduces the tendency of the DFCS to declare hard failures during transient or nuisance failure
conditions.

FLIGHT CONTROL SYSTEM SELF-TEST

Flight Control System (FCS) self-test capability can be viewed as consisting of four tiers of tests of
increasing complexity and fault isolation coverage. These tests include (in order of increasing com-
plexity) the following:

e Inflight Integrity Management (IFIM)

e Inflight Self-Test
9 Preflight BIT
* Maintenance BIT

1. Inflight Integrity Management (IFIM)

IFIM is active whenever power is supplied to the FCS and is used to protect against only those
failures which, if they occur in flight, would have a direct effect on FCS performance and/or cause
a reduction in the redundancy level of the system.

The primary I FIM mechanisms are in the three voting/monitoring planes previously described. When
failures are detected at these planes by the cross-channel monitors and persist, the pilot is informed
of the failure by means of caution or warning lights and by a message on the Multi-Purpose Display
(MPD). By keying up the FCS fault-report page on the display, the pilot will be informed of the
exact failure which has occurred and can attempt a reset. A reset command will attempt to clear the
failure which is on the MPD and has the effect of restoring the redundancy by one level. Thus a
reset of a second-like failure only resets the second failure and cannot foil the voting algorithm. If
a reset is successful, the appropriate message will disappear from the MPD and the lights will be
extinguished.

IFIM also includes several hardware monitors which are also continuously active. These monitors
perform the following functions:

* Monitoring of electrical power which will produce a high priority vectored interrupt in case of
power failure.
* Watchdog timer function which disengages FCC outputs in case of major FCC failures.
o Monitoring of the word count and fresh data flags on the Avionics and inter-FCC MUX bus
receivers.
1 Monitoring of fresh data flags on A/D converter outputs.

The purpose of these hardware monitors is to declare a single hardware failure instead of a multitude
of input or system failures as would be the case if only the selector/monitors were employed. The
information from these hardware monitors in conjunction with the failure information supplied by
the selector/monitors is used by a software routine called the "failure manager" to determine what
failure message is supplied to the MPD's.

78
 NADC-82240-60

2. Inflight Self-Test

Inflight self-tests include hardware tests which are always active, software cyclic tests, and tests
which are event-driven (usually a failure). A failure detected by these tests is always recorded by
the FCC and transmitted via the MUX bus to the instrumentation system but does not necessarily
result in a failure indication to the pilot.

Hardware self-tests, which are always active, include memory parity on all FCC memory and parity
checks on all MUX bus receivers. A memory parity failure causes a processor interrupt and executes
a software routine which records the failed Icoation for later retrieval. Parity errors on transmissions
are used to alert the FCC not to depend on the data received during that cycle.
CPU self-test, RAM, and ROM tests are event-driven and are executed in the air only after a second-
like failure is detected at the output monitor plane or on the ground by a higher level test.

3. Preflight BIT

Preflight BIT includes all I FIM and inf light self-tests plus additional FCC tests and testing of LRU's
external to the FCC. Preflight BIT is initiated, controlled, and monitored via the MPD's. The tests
are performed by the FCC resident Operational Flight Plan (OFP) and can only be initiated on the
ground (interlocked by appropriate discrete inputs).

When power is applied to the FCS, all FCC's are in the IBU mode and will start executing flight con-
trol laws associated with the NORMAL multimode when brought on line, with IFIM and inflight
self-tests operational, and with the corresponding NORMAL mode display. The pilot will then check
his failure lights and fault-report display and then initiate preflight BIT.
The preflight BIT display lists all manual tests which are still pending (not yet performed) and fail-
ure codes for all tests which were not passed.

4. Maintenance BIT

Maintenance BIT adds non-structured tests to those performed by preflight BIT in order to check
out the FCS after maintenance or modification or as an aid in failure isolation to an LRU. All
maintenance BIT tests are designed to use an assumed healthy FCC to find faults in exterior inter-
faces or equipment.

Maintenance BIT essentially turns the MPD's into remote computer terminals which allow the op-
erator to look at the contents of any memory location in all three FCC's simultaneously and then
use an MPD as a multimeter. For example, the operator could key in the memory location cor-
responding to left AOA, have a technician position the transducer to a predetermined position, and
then determine if all electrical inputs were received by each FCC properly with the correct sign and
within magnitude tolerances. By proper test selection, the operator can check out any system mod-
ification or isolate failures to an LRU.
4
INDEPENDENT BACK-UP UNIT (IBU)

To provide an independent backup capability, a very simple analog Independent Backup Unit (IBU)
was developed. The IBU provides stability augmentation and control in pitch and direct flight con-
trol surface control in roll and yaw. A block diagram of the IBU is shown in Figure 46.

79
K-
 NADC-82240-60

0 c

- c zr T
0

F-- w~ 0

xOQ

~
Io zz

0' z~ -

I _ _ _ _

10.

80.
 • r, .NADC-82240-60

The design philosophy for the IBU is to provide a simple controller and ensure safe flight. Per-
formance requirements of the IBU are to meet Level 3 requriements of MI L-F-8785 (ASG) for
cruise and descent and Level 2 for landing. Control in the longitudinal axis is provided by a pitch-
rate command system. The pitch command from the stick is shaped and then lagged by a prefilter.
The pitch rate error signal is then gained and fed through a proportional-plus-integral network. Dif-
ferent lead/lag compensators for the gear-up and gear-down configurations augment stability margin
and improve handling qualities over the entire flight envelope. The forward loop integrator is not
driven when the IBU is not engaged. The integrator is driven when the IBU is engaged and there is
no weight on wheels. In order to minimize engagement transients, the integrator is also driven when
the digital system is operating, but a self-test is being carried out If the IBU is engaged, the inte-
grator is discharged when there is weight on the main landing gear. Direct aileron control is used in
the lateral axis. The command signal from the rudder pedal is simply passed through a shaping
gradient function. Fixed biases are provided to drive the trailing-edge flaps. With the landing gear
up, the flaps are driven to 0 deg. The leading-edge flaps are not driven by the IBU, but are simply
locked once the IBU is engaged. Analysis has shown that the IBU satisfies Level 3 flying qualities
for cruise and descent and Level 2 for landing. Stability margins in the pitch axis are adequate over
the flight envelope. Pilot-in-the-loop simulation has shown that the IBU provides capability for a
safe return to base and landing for emergency situations.

DFCS-ELECTRICAL SYSTEM INTERFACE

The FCS will receive electrical power from two essential 28 VDC buses, backed up by four batteries,
a 5kVA emergency generator, and a 500 VA permanent magnet generator. The electrical system
will provide uninterruptable, regulated power regardless of transient voltages or fault conditions
elsewhere in the airplane system. This input power will be provided to each branch of the FCS
where it will be finally converted and regulated to produce the 800 Hz AC and the DC voltages to
power the DFCS. Figure 47 illustrates the AFTI/F-16 DFCS redundancy concept. Column 3 of this
figure illustrates how the primary 40 kVA generator is backed up.

A schematic diagram of the AFTI/F-16 Electrical Power System is shown in Figure 48.

Note that each of the two essential 28 VDC busses receives its power from a dedicated 115/200
VAC-to-28-VDC converter. Each converter may receive input power from either the 40-kVA
primary generator of the Emergency Power Unit (EPU) 5-kVA generator. The EPU is driven by
engine high-pressure bleed air in the event of improper generator voltage or frequency. Should both
the 40 kVA primary generator and the 5 kVA emergency generator fail to supply power to each one
of the 28 VDC converters, then the Permanent Magnet Generator (PMG) portion of the EPU's
emergency generator supplies 18-volt, 3-phase, 1,200 Hz, power to a third converter dedicated
wholly for FCS power supply usage. Furthermore, this converter supplies power to the EPU's
electronic controller so that continued operation of the EPU is assured (a "boat-strap" type of
arrangment).

Multiple power sources preclude loss of power to the battery buses because of a malfunction in the
input power system. The 24-volt batteries (18 cell, Ni-Cad type) prevents power interruptions dur-
*ing power source switching operations.

The electrical power supply system does not provide any "turn-off" capability to the pilot. How-
ever, provisions are included for shutting off electrical power to the FCS while the airplane is on the
ground. Ground "shut-off" provisions are desirable from the standpoint of equipment life and re-
liability so that the FCS need not be energized during routine maintenance operations.

81
4 NADC-82240-60

UL

01+ >
01
ccu

zo
u C
0u > C

z >

>, Lu -

LL

C.) R -j

z
0 - 0,n
U) zcc U
I- M.- CC
LL. CLL.
4
>U cc. -- cc

LL.L.
U.a L0L

9L.

z 04<)

cc
-az -i
u

4 82
 NADC-82240-60

UL

C-

~~-Z

Nj4

I U)

0' z

83 -
 NADC-82240-60

DFCS-HYDRAULIC SYSTEM INTERFACE

Flight control surface actuation is dependent entirely upon the successful operation of the hydrau-
lic system. The AFTI/F-16 hydraulic system consists of two separate and independent systems
that supply hydraulic power for operation of the primary flight control surfaces, the vertical ca-
nards, LEF, speed brake, landing gear, nose wheel steering, wheel brakes, fuel flow proportioner,
and other utility functions.

A block diagram of the hydraulic system is shown in Figure 49. The major elements of the hydrau-
lic system consist of the following elements:

0 Two main hydraulic pumps

* An emergency hydraulic pump
0 Hydraulic reservoirs
0 Seven flight control ISA's.

DFSC-AVIONICS SYSTEM INTERFACES

The DFCS and certain elements of the avionics system (see Figure 50) will be operated in aggregate
to provide the integrated-DFCS-demonstrator functional capability. As shown in Figure 50 the
DFCS is comprised of the FCC complex, inertial sensors, cockpit controllers, the ISA's and LEF
actuators, the AIU, and the FCS control panel. The avionics system elements used to provide the
principal Pilot/Vehicle Interface (PVI) for DFCS mode control and status annunication include the
following:

- * Two interactive Multipurpose Display (MPD's).

a Two Programmable Display Generators (PDG's).
0 A dual redundant Stores Management Set (SMS) Control Interface Unit (CI U).
* A redundant multiplex (MUX) data bus.
" Primary and backup MUX data bus controllers.

The primary avionics MUX bus controller will be the Fire Control Computer (FCC), and the back-
up bus control will be provided by the SMS.

Figure 51 shows some of the avionic type components that the DFCS will interface with via the
avionics and display multiplex buses. The safety of the AFTI/F-16 aircraft and the fail-safe integ-
rity of the FCS will not be affected by failures of the dual avionics bus or any of the interfaced
equipment.

The DFCS/avionics functional interface provides the following benefits:

* Data entry capability.

- Display of mode requested/mode status.
0 Status and fault reporting.
0 Built-in test capability.

Figure 52 illustrates some of the payoffs obtained by optimal integrations of the avionics systems
and the DFCS.

84
 NA DC- 82240-60

AIRFRAME-MOUNTED ACCESSORY DRIVE GEARBOX

42517 GPM.I~
D T_
IPOWPOER

HYD
UMP
EMERGENCY
UNIT
EMEW,
r -

I 5E i i '
t~2 ' 3PM PS

7- I- pm,~

L -- --

IiIf-
-- I-- HYDRESERVOIR SYTM-
YYORAULC RESERVOIR - -

RN HORIZONdTALTAIL F. GNSNPREDO

SPTEEDRAEERVACTUATOR

INERAE FILTR GRUND SCONEC

ACUMLAO e SHTOFFVA.V

INTGRTE PRESSUREAO
SWTC RESSR NIAO

I I LOW
L IMITER

Figure
Hydralic
4. Systm FunctOna lokDiga

THOTEBOT4TFE TRE
r - - - - - -85
 NADC-82240-60

I z' F
'0
-~ ~

Za

a
100-oa

.
ar- -

JE
~~~. -

y61 C

-E.' 86I~ ~.
~ !
- I 1 a
 NADC-82240-60

0 -MULTIPURPOSE

SHEADUPPROGRAMMABLE

IMPROVED STORES
MANAGEMENT SET (SMS)
CENTRAL INTERFACE
UNIT (CIU) IFFC PROCESSOR

PROVISONING SENSOR/TRACKER POO

0 HELMET MOUNTED SIGHT

HUD ELECTRONICS
UNIT (EU)

Figure 51. AFTI/F-16 Avionic Components

AIRFRAME POSITION AND RATES

TARGETPOSITITON
V

DIECO EQUIVALENT FUNCTION WITH LESS REAL

ESTATE

C O IMPROVED FUNCTIONAL RELIABILITY

4 L 1 BUTTON COCKPIT RECONFIGURATION

'. * ESTABLISH COMMAND PATH FOR CLOSED

LOOP OPERATION
** AFTI-16 REFLECTS NEXT GENERATION
°- - 11 I ; - ..i . . I " . . . . .I. ..
0LEVEL OF INTEGRATION

Figure 52. AFTI/F-16 Avionics/DFCS integration Technology Payoffs

44 87
 NADC-82240-60

PILOT/VEHICLE INTERFACE (PVI)

The AFTI/F-16 crew station is shown in Figure 53. This arrangement was established to provide an
effective Pilot/Vehicle Interface (PVI) co -istent with the multimode task - tailored FCS design
* philosophy. Cockpit Controlled Control/Display System implementation is based on mission task
. requirements.

1. Cockpit Controllers

The operation of the cockpit controllers will now be reviewed. In the pitch axis the SSC
always commands normal acceleration so that command ambiguity is removed. The de-
coupled direct lift SSC commands generate coupled normal acceleration response after
the flaps reach maximum deflection. This provides the pilot with full authority control
of normal acceleration. Manual vernier control of pitch pointing and vertical translation
is accomplished using the left hand controller (Linear Throttle), a unique twist grip con-
troller physically integrated with the throttle grip assembly.

The SSC lateral force commands roll rate at all times. For the standard operation, the
rudder pedals command the following:

e Conventional rudder deflection with stability augmentation in the normal mode.

* Direct side force (flat turn) in all of the combat modes.

In the decoupled operations, the rudder pedals command decoupled motions that are
in concert with the pitch axis control and are task oriented.

2. Control/Display System (CDS)

The AFTI/F-16 DFCS is designed to provide integrated control for changing mission seg-
ments with a single stroke of one of several keys integrated within the Cockpit/Display
System (CDS). The CDS is designed to accomplish the PVI function in a manner which
permits effective mode selection. The CDS provides the capability for integrating multi-
ple equipment control functions and displays on a single panel. In addition, the CDS
has the capability to automatically configure the avionics and DFCS upon selection of a
particular mission phase.

The CDS consists of a Head-Up Display (HUD), two interactive Multipurpose Display
(MPD), and dedicated cockpit switches (see Figure 53). The MPD's provide control/
display/status functions for the avionics and the DFCS.

The CDS provides redundant DFCS/ avionics interfaces in the DFCS safety-critical paths
to ensure flight safety. These interfaces are single-failure tolerant so that no one failure
in the MPD's/DFCS interface will decrease the number of functions controlled through
the MPD's. Display symbology and formats provide the information required by the
pilot to effectively perform mission-related tasks.

Figure 54 shows the flight control options displayed on the left MPD.

88
 NADC-82240-60

zz

U.-

00 0

00

L) U)
-0 zz
00
LL

00
cc It
0 2

U) 0

R- U.

0 42

L-J -j

0~ 89
 NADC-82240-60

, .:.FAULT NST BIT AU T

I'
FC _ ( IFFC

V COUPLED/DECOUPLED

DEMOD - DRAG MODULATION

FLIGHT 0 ENABLE/DISABLE
CONTROL
MODE ST
ASSOCIATED 0___
WITH OPTIONS A- DECOUPLED

ON G PNT -OPTIONS
21i'NORM
DLFT -0

FCS ViO

Figure 54. Flight Control Options Displayed On MPD.

From the flight control standpoint a single stroke of one of the keys shown will perform the
following functions:

* Enables the MPD to present the pertinent mode information.

e Commands the FCC's to execute the control laws optimized for the mission segment.
e Enables the preselected manual 6 DOF modes on the left hand controller (throttle grip) and
the rudder pedals.
* Presents the proper symbology on the Radar/EO and HUD displays.

The pilot can then engage or disengage the selected decoupled mode. The 6 DOF modes imple-
mented may be modified at any time by pressing the appropriate switches on the MPD. Pressing
the "PRSET" key will cause the display shown in Figure 54 to be presented.

By use of the MPD, the pilot can select, by flight control mode, any of the following options:

CONTROLLER OPTIONS
9 Side Stick Controller (SSC) e Flight Path Maneuver Enhancement
r Pitch Rate Maneuver Enhancement

* LH Controller (on throttle grip) * Pitch Pointing

* Vertical Translation
* Direct Normal Force (Direct Lift)

- Rudder Pedals * Direct Sideforce (flat turn)

e Yaw Pointing
9 Lateral Translation
* Blended Pointing

90
 NADC-82240-60

THE ADVANCED FLIGHT CONTROL MODES OF THE AFTI/F-16

The AFTI/F-16 is intended to provide superior flying qualities by incorporation of the ability to
tailor aircraft response to a particular mission segment and by the utilization of vertical canards
and maneuvering flaps to further enhance mission effectiveness. Conventional non-multimode
*"designs generally result in a compromise between smooth ride and good handling qualities. The
AFTI/F-16 FCS will be designed to void this compromise by emphasizing tracking and handling
qualities during attack mission phases and ride qualities elsewhere.

1. The Four Major Flight Control Modes -

The AFTI/F-16 DFCS demonstration vehicle will be provided with four major flight control modes
*•and decoupled CCV modes utilizing redundant control surfaces in each control axis, (Figure 55),
which can be selected in any pitch and yaw combination to tailor aircraft to the mission. The major
modes are as follows:

* Normal Mode - The normal mode is used throughout the applicable flight envelope for takeoff,
cruise, and landing and for the performance of secondary mission tasks, such as air refueling
and formation flying. The mode is designed to provide smooth ride, gust alleviation, and re-
duction of pilot workload during secondary mission segments.

e Air-to-Air Gunnery (AAG) Mode - The air-to-air gunnery mode is used throughout the air
combat flight envelope to provide rapid maneuvering during target intercept and precise track-
ing. Control law design strategy for this mode is based on optimizing tracking characteristics.
The pilot is then able to quickly null target tracking error while he obtains improved handling
qualities by using the trailing-edge flap as a longitudinal control surface. The resultant opti-
mized response is then applicable to a wide variety of air-to-air target acquisition and-tracking
tasks.
* Air-to-Surface Gunnery (ASG) Mode - The air-to-surface gunnery mode will provide rapid and
precise pointing for increased accuracy and survivability when ground targets are being strafed.
Design criteria used to null target tracking error are the same as those developed for the air-to-
air mode, with additional emphasis placed on improving gust alleviation response.

o Air-to-Surface Bombing (ASB) Mode - The bombing flight path mode will provide precise con-
trol of the aircraft flight path and will improve gust alleviation to facilitate bombing accuracy
and to enable the employment of effective control strategies in order to increase aircraft
survivability.
Figures 56 thru 59 inclusive illustrate how these modes are executed by the pilot.

2. The Control Configured Vehicle (CCV) Modes.

Adjunct CCV modes preselected through the display subsystem and featuring control inputs through
41 the twist-grip throttle and the rudder pedals will provide added capabilities. These modes include
vertical and horizontal translation, pointing, and direct force. Any longitudinal and lateral combina-
tion of adjunct modes may be selected translation may be used with the mode. For example, vertical
and horizontal translation may be used with the normal mode for formation flying; longitudinal
pointing and lateral direct force may be used to augment the air-to-surface gunnery mode. Domin-
nant features of the standard and decoupled DFCS multimodes are summarized separately for pitch
and lateral-directional control in Figures 60 and 61.

491
 NADC-82240-60

0i
"Ai

CLC

o 0
U.1

CL..

UU-
hi-z

uiL

uj C6 -n

==U
CA 0

-IU.

w-o
*u
IC

z9
 NADC-82240-60

S*O COORPNATE FLIGULT

IS FI FU L/ O FMDE OOU LED T IO S

* N RM L ODECR
DOWNC-CONTLROLUO

PARAMECTTOERSRMEER

PILOT
ONOALINA SLOSES O HT
LIUN
TAOLORED
MGMOOT WITH PAAG
CONTROLUS * ANSATONSECTO
OFNDECOUPLED OPTION

CONTLIGTPTOTO

4 93
4t NADC-82240-60

AIR. TO- SURFACE GUN MODE

TARGET ACQUISITION -6
9 PILOT FINDS
TRUCKCONVOY- ALIGN FOR GUN FRING
0 HANDS ON A U
SELECTION OF e PILOT USES DECOUPLED
DECOUPLED OPTION OPTION FOR RAPID
AND ACCURATE
HEAD.UP-DISPLAY STRAFING

C 31 30'
N MODE
SELECTED
KNTOLFT HIGH G TURN TO
* PREPROGRAMMED PNT DEFEAT TARGET
SELECTION
e ONE BUTTON /DEFENSES

2
"POP-UP" FOR TARGET -
ACQUISITION S
GUN FIRED

KTAS J,

_- ,GROUNDTRACK

Figure 58. Air-to-Surface Gunnery (ASG) Mode.

AIR- TO - SURFACE BOMB MODE

TARGET ACQUISITION
•PILOT FINDS

HANDS O"
FUEL DEPOT
HANDS
IIEI
ONRELEASE
EJ WEAPON
ALIGN FOR

SELECTION OF ' PILOT USES OECOUPLED

DECOUPLED OPTION OPTION FOR RAPID

HEAD.UP-DISPLAY AND ACCURATE

"" ALIGNMENT

MODESELECTED 00000
" PREPROGRAMMED A C3
* ONE BUTTON C3 L HIGH G TURN TO
SELECTION C3 DEFEAT TARGET
DEFENSES

Ii iACQUISITION
.POP-UP" F
ARGEZT

-' K T A

"" - --" GROUND TRACK

Figure 59. Air-to-Surface Bombing (ASB) Mode.

94
 NADC-82240-60

3. Control Law Design Methodology

The AFTI/F-16 multimode flight control laws are structured to provide both standard conventional
control and decoupled control. The standard control characteristics are tailored for maneuvering
flight and target acquisition, and the decoupled control characteristics are tailored to vernier track-
ing and small-amplitude-weapon delivery-error removal. High performance, unique maneuverability,
rapid pitch and roll response, neutral speed stability, turn coordination, departure prevention, high
angle-of-attack (AOA) performance, and high AOA command limiting are major design considera-
tions in the development of AFTI/F-16 multimode control laws. Flying quality requirements, such
as MIL-F-8785B and suggested revisions thereto, have been developed for use in the design of con-
ventional aircraft and will be satisfied by the AFTI vehicle in the conventional modes. Since, by
use of these criteria, decifiencies have been discovered in accurately predicting pilot ratings and
tracking characteristics for the high-order system associated with highly augmented aircraft, addi-
tional criteria - such as C*, D*, Neal-Smith, and the newly developed step-target tracking techniques
- were used to compare designs. Six-degree-of-freedom simulation results will provide the final test
and acceptance criteria for all modes before flight testing.

This multimode design flexibility allows the DFCS designer to establish gains, feedbacks, and com-
pensation parameters not only as a function of flight condition, but also as a function of the mode
selected. The system performance can then be tailored to match the desired characteristics of a
specific task at hand, e.g., air-to-air combat, air-to-surface bombing, etc. Besides the design flexi-
bility obtained with a multimode system, the AFTI/F-16 vehicle utilized the trailing-edge flap as an
additional control surface (along with the horizontal tail) in the longitudinal axis to enhance its
performance capability. The flap used in conjunction with the horizontal tail can produce changes
in the lift vector without the corresponding rotation of the aircraft, characteristic of systems using
only the horizontal tail-for control. The motions of both flap and tail can be coordinated by proper
control system design to provide maneuver enhancement, decoupled motions, and gust alleviation
to the aircraft.

The four basic task-tailored modes significantly affect the inner-loop design of the flight control
system. It should be evident from examining Figures 60 and 61 that the modes can also be cate-
gorized by the influence on performance of aircraft flight path, as in the Normal and Bombing
Modes, or aircraft attitude as in the Air Combat and Strafe modes. This influence is manifested in
the flight control system design by controlling the normal acceleration and pitch rate responses of
the aircraft, respectively.

The analytical method used to derive the longitudinal axis feedback gains for each of the AFTI/F-1 6
standard modes is based on optimal control theory and is referred to as linear quadratic synthesis
(LQS). In this procedure, weighting parameters in a quadratic cost function are selected to yield
systems with desirable performance characteristics. Weighting parameters reflect consideration of
gust response, tracking performance, pahse and gain margin, short period frequency and damping,
transient response characteristics and other relevant performance indices. Simplified longitudinal
and lateral-directional block diagrams of the Normal Mode flight control system are shown in
Figures 62 and 63, respectively. In addition to the feedback gain variables, further response tailor-
ing is provided to the system through the command path parameters, pilot gain (Kp) and pilot time
constant (1/a). Optimum L/D flap scheduling is also provided.

Most of the inner loop gains are scheduled with air data derived quantities to provide good response
over the entire flight envelope. In addition, some filter time constants and break frequencies are
also scheduled. Structural limiting is also provided to prevent overstressing the airplane during
maneuvers. One of several ways this is done is by limiting flap travel as function of air data, Mach,
altitude, and load factor.

4| 95
 NADC-82240-60

COORDINATED MANEUVERING CONTROL PATH CONTROL

FLT PATH
CONTROL 0 PRECISE ATTITUDE SATTITUDE g'GUST
CONTROL GUST ALLEVIATION * ALLEVIATION

DEOPOSTOLEOULDDCULE EOPE

C_OROL AITUD ATTITUDE 1LT

PATH
CRNLTOONTROL ONTROLG CONETRLF

SRESPONSIVE *PRECISE
MANEUVERING DECOUPLED
CONTROL FLT PATH
CONTROL

Figure 60. Pitch DFCS' Multimode Control

SCOMMAND COMMAND COMMAND *COMMANO

irCOMMAND FLAT TURN FLAT TURN FLAT TURN

*CONVENTIONAL 'PRECISE 0PRECISE 0 PRECISE
*MANEUVERING MANEUVERING MANEUVERING

NOMLMI CMAT STRAFE

COMMAND COMMAND COMMAND COMIMAND

dTRANSLATION POINTING POINTING FLAT TURN

*PRECISE 0PRECISE 0PRECISE ePRECISE

POSITION DECOUPLED DECOUPLED FLT PATH
CONTROL ATTITUDE ATTITUDE CONTROL
CONTROL CONTROL

Figure 61. Lateral-Directional OFOS Multimode Control

4 96
 NADC-82240-60

A conventional approach to directional augmentation was selected for the system concept. Lateral
acceleration feedback provides coordination; washed-out yaw-rate feedback enhances dutch-roll
damping, and an aileron-rudder-interconnect (ARI) and a roll rate-AOA (pa) interconnect improves
turn coordination. The principle alternative to this design philosophy is the beta-dot (sideslip rate)
feedback concept employed in the decoupled system. The beta-dot signal is synthesized using non-
redundant Euler attitude and velocity data. A technique has been developed to give fail-operate
performance in the decoupled control mode in the event of the loss of these data. However, higher
reliability is desired for the primary takeoff-and-directional control system so a separate, more con-
ventional, directional control system has been developed for the Standard Normal Mode. In sum-
mary, the approach that was taken in the flight control system design of the Standard Normal Mode
was to develop a system that produced smooth, uniform, responsive handling qualities and provided
gust alleviation of normal acceleration, while simultaneously meeting the stability and root location
specification, and not violating any physical constraints. Using this design procedure, gain schedules
were established over a suitable range of flight conditions.

As in the Normal Mode, certain tradeoffs arise in the development of the flight control laws in the
Air Combat Mode. However, since tracking characteristics and aircraft attitude control are of pri-
mary interest in this mode, a different set of performance measures are emphasized in evaluating
system designs. To aid in achieving the desired level of performance in the mode, additional com-
pensation was added to the command path of the flight control system. The structure of this com-
pensation is depicted in Figure 64. Because of task similarities, the Air-to-Surface gunnery mode is
virtually identical to the Air-to-Air gunnery mode in the pitch axis. It should be noted that the
gains shown in the block diagram will also have values different than in the Normal Mode. The
methodology described above was also used in developing the Air-to-Surface bombing mode which
emphasizes precise flight path control and is illustrated in Figure 65.

4. Decoupled Control Modes -

The decoupled control laws provide independent control of all six degrees-of-freedom of motion.
They are accessible from all major modes. Certain default or preselected decoupled motions have
been defined for the major modes but can be overriden by pilot input through the MPD's. AFTI/
F-16 decoupled maneuvers are illustrated in Figures 66 thru 71 inclusive.

The control laws for implementing these decoupled modes were synthesized using classical fixed-
flight condition feedback control methods. Within this classical approach, there were several ways
to decouple pitch rotation from normal acceleration. One such way is to analytically derive the
precise compensators, as a function of aircraft aerodynamic parameters, required to cancel the basic
aircraft dynamics while at the same time substituting a set of desired dynamics for the total closed-
loop system. Results of analysis showed that responses obtained from a system decoupled in this
manner were unacceptably sensitive to errors in the designer's knowledge of the fundamental air-
craft aerodynamic parameters.
Another approach, and the one selected for implementation, is the use of feedbacks of the variables
to be decoupled to produce high-gained error signals to drive the control surfaces. This approach
d does not rely on unreasonably accurate estimates of the aerodynamic parameters to accomplish an
acceptable degree of decoupling. When the Normal or Bombing Modes are selected, the system con-
figuration will emphasize flight-path control whereas the gunnery modes will emphasize attitude
control. An important consideration in the final selection of the decoupled variables is the fact that
the IFFC algorithms produce normal and lateral acceleration commands and pitch rate, yaw rate and
roll rate commands. Thus, An and q and Ay, 0, and p were selected to be decoupled for the longi-
tudinal and lateral modes, respectively. Coordinated control-surface limiting (CCSL) was developed

4 97
 NADC-82240-60

Trailing-Edge
Flap Command
8F

Normal Knn- +

Acceleration a
Anc n

Normal
Acceleration. An
Pitch - +- + +

Angle 8 el Elevator
of Deflection
--c Command
~ 8t
Figure 62. Simplified Longitudinal Standard Normal Mode Block Diagram

Differential

P Elevator
-K Command

++

I
S +35 Command

___
Canard
57.3 Command

Pedal
Force

Figure 63. Simplified Lateral-Directional Standard Normal Control Mode Block Diagram.

98
 .. -'_ - . . -. -

NADC-82240-60

as a means to accomplish smooth transient-free control-surface coordination and limiting. Proper

decoupled mode operation is always dependent on balanced forces and moments. This requirement
dictates that when a control-surface limit occurs, further decoupling inputs to other control surfaces
must be inhibited and the system status maintained at that point until surface limits are no longer
exceeded. This CCSL concept has been implrmented, as required, in the longitudinal and lateral
axes to achieve control-surface limiting while retaining the maximum available decoupled response.
The engagement of the decoupled modes with correct initial conditions is accomplished in an initial-
ization routine that automatically balances the system with the existing set of inputs and outputs.

Tables 7 and 8 list the Longitudinal and the Lateral-Direction Control Modes respectively. The
controllers and control surfaces used are also listed.

Table 7
AFTI/F-16 DFCS DECOUPLED
LONGITUDINAL CONTROL MODES

0 Modes
1. AN - Direct Lift
2. ci - Fuselage Pointing
3. a2- Vertical Translation

• Controller
* Twistable Throttle Grip

0 Control Surfaces
* Flaperons Balanced with Horizontal Tail

Table 8
AFTI/F-16 DFCS DECOUPLED LATERAL-
DIRECTIONAL CONTROL MODES

* Modes
1. Ay - Direct Sideforce
2. 0 - Fuselage Pointing
3. 02 - Lateral Translation

0 Controller
* Rudder Pedals

4 0 Control Surfaces
* Vertical Canards Balanced with Rudder

99
 NADC-82240-60

r I -

us 0 'I
jt

C.,
-C ,-
m= mu
A2

UU-

51.
0f 0

ELLI

cc OWiU.

1d00
4 NADC-82240-60

99 t

LU ca

03 3GUL H31
F0

U,.

CI

I~a _________ __________ ________cc

Ula. e ~ - az

'I-NOIIWU313~~V
- NI 33C.'U
cc
us0
2 =
a. o - I-

ci = a

Sb =-
61M
I/u-
C0L0

100
 NADC-82240-60

* VERTICAL FLIGHT PATH CONTROL

4 i CONSTANT ADA
" 3"g" AUTHORITY
* INCREASED "g" CAPABILITY

* USEFUL FOR PITCH ATTITUDE

CORRECTIONS
s MINIMUM ALTITUDE LOSS DURING
DIVE RECOVERY %,
* MOST IMPRESSIVE OF THE
LONGITUDINAL MODES
-
Figure 66. Direct Lift (AN) Control Mode.

* DIRECTIONAL FLIGHT PATH CONTROL

- WINGS-LEVEL FLAT TURN
- 0.Sg SIDEFORCE
r No SIDESUP

" USED FOR HEADING CORRECTIONS

AND LATE LINEUP CORRECTIONS " N *
ELIMINATING SIGHT PENDULUM EFFECT
* EASY FOR PILOT TO ADAPT

Figure 67. Direct Sideforce (Ay) Control Mode.

a PITCH ATTITUDE CONTROL

e CONSTANT FLIGHT PATH
a 30 TO 40 AUTHORMI
(50-70 MILLIRADIANS)

* PRRIOUISITE FOR INTEGRATED FIRE/FLIGHT CONTROL

* MOR IMPRESSIVE FROM PILOTS VIEW

Figure 68. Pitch Pointing (al) Control Mode

102
- NADC-82240-60

* DIRECTIONAL ATTITUDE CONTROL

* CONSTANT FLIGHT PATH
* NO SIDEFORCE
* 50 SIDESLIP AVAILABLE

• PREREQUISITE TO INTEGRATED FIRE/FLIGHT CONTROL

* MORE IMPRESSIVE FROM PILOT'S VIEW

Figure 69. Yaw Pointing (01 ) Control Mode.

* VERTICAL VELOCITY CONTROL

s CONSTANT PITCH ATTITUDE

IDEAL FOR SMALL VERTICAL

POSITION CHANGES SUCH AS
FORMATION FLYING OR FLIGHT
PATH ADJUSTMENTS

Figure 70. Vertical Translation (, 2) Control Mode.

*SIDE VELOCITY CONTROL

*CONSTANT HEADING
* 68 FT/SEC AVAILABLE

* USED FOR SMALL LATERAL

POSITION CORRECTION
0 IDEAL FOR CROSSWIND
CORRECTION -

Figure 71. Lateral Translation (02) Control Mode.

1* 103
 "* NADC-82240-60

Figure 72 thru 74 inclusive illustrate how the use of these decoupled modes improve the operational
capability of an aircraft.

a. Longitudinal Decoupling -

" - Since normal acceleration and pitch rate are the variables to be decoupled, these sensed quantities
are fed back as shown in the Air Combat mode example of Figure 75. The sidestick controller pro-
vides direct lift commands and the throttle grip provides pointing and translation commands. A
.* third feedback is angle of attack. Since the airframe is statically unstable over most of the subsonic
flight envelope, a combination of angle of attack and pitch control law decouples this new "pre-
stabilized" coupled system. Decoupling relies on the symmetrical deflection of the trailing-edge
flaps as well as the symmetrical deflections of the horizontal tail. An important limiting feature is
the angle-of-attack limiting. Pilot command limiting is also provided. The design also includes a
" flap "washout" function. To prevent the flap from achieving steady-state offset from its optimum
flap schedule, the flap command is fed back in a manner which tends to return the flap to the flap
position prescribed by the optimum flap schedule. The flap remains at (or returns to) the position
dictated by the optimum flap schedule during relatively quiescent periods of controller activity.

Another important functional element of the system is the surface command crossfeeds. The accel-
eration error, which becomes ultimately the primary flap command, is also scaled by a different value
and sent to the horizontal tail. The scale factor applied is theoretically the one required to alter the
tail position by the amount required to cancel the change in pitching moment due to the new flap
position. Conversely, the pitch-rate error is crossfed through an appropriate scale factor to the flap
to cancel the change in normal acceleration created by the new tail position. The crossfeed of nor-
mal acceleration to the tail isan influential path, because even though the flap is primarily a lift-
producing surface, it is also a non-negligible moment producer. On the other hand, the pitch-rate
error-to-flap corssfeed is quite insignificant at most flight conditions because tail position changes
produce relatively small lift changes.

b. Lateral-Directional Decoupling

To generate lateral-directional decoupled motions, the control system must be able to constrain the
value of any two of the three directional states (sideforce, sideslip, heading) in addition to minimiz-
ing roll rate. In order to achieve this capability three basic design concepts are employed. The Air-
to Surface bombing flat turn mode of Figure 76 is an example. (a) A dedicated feedback loop is
designed for each lateral-direction state to be controlled. Roll rate is fed back to the aileron controls;
beta-dot is fed back to the rudder, and sideforce is fed back to the canards. (b) In each of the con-
trol loops, a proportional-plus-integral network is engaged to force steady state command errors to
zero. (c) The underlying mathematical theory of the decoupling concept assumes that the com-
mands from the dedicated control loops can produce pure forces or moments. For example, the
roll loop command to the ailerons (flaperons) would produce a rolling moment without attendant
yawing moment or sideforce. Similarly, the force-loop command would produce surface deflections
do produce sideforce, rolling moment, and yawing moments, interconnect command signals are
generated in order to cancel the undesirable effects. In theory, a total of 11 interconnect signals
*l of this sort would be required to cancel the cross-coupling between system states. However, all but
four of these signals can be neglected and satisfactory decoupling performance of the Standard
Normal system; the roll-rate-error integrator is engaged either during gunfiring operations or when
a directional command is input via the pedals. There are also two decoupling signals added to the
aileron command. One signal is the canard-to-aileron interconnect; the other cancels the rolling
moment induced by the sideslip which is generated during pointing or translational maneuvers.

6I 104
 NADC-82240-60

U-

ca t; 'UC

LL. CC

&U,

CC LLI
L(U

Ll.W

dc!h-. L

Q CS

L3L

105
q NADC-82240-60

0 U.

LU

z ca C

CLU -
W U-
U.
2= L6 a

LU- 4-

040

LUA

0 l

-J.

LU 4
u

Z L c

0 >

CA U
dc. @0i
j0.L

ac

L6 0106
 NADC-82240-60

us
11 I
L==
c-c
cc
26-

Z C6

LL
rL
E

L gJU

0, Cc

107
 4 NADC-82240-60

-- 0r

U.o

U.U
C-

14-

190
E

La.

cca

44
2:

a. 4c u
L. & I
C2
a -, rcts
, w ci 4r A-L

~108
 NADC-82240-60

CIO'

ICA
=iL UL
cc:E2
P

I-CIO

'00

aL

uCu

LUU

C02
-4-

'.V-.

0 'c

* 0a

aL

oU J
ci

*. uJ~~ =uJ -
~a 0.0LIJU).

9.109
 NADC-82240-60

The most important differences between the Standard Normal and Decoupled systems for lateral-
directional augmentation is the choice of feedback states used to control yawing moments. The
decoupled system used a synthesized beta-dot feedback for the following reasons: (a) Beta-dot is
a key quantity that must be precisely regulated to maintain decoupling purity, and its direct use as
a feedback will maximize the degree of control over it. (b) Better dutch-roll damping can be ob-
tained from a beta-dot system in spite of the special decoupling interconnects that are required and
tend to destabilize the system. This consideration is much more important because directional
motions will be intentionally induced as a control mechanism during precision tracking tasks. (c)
Better turn coordination can be obtained from the beta-dot system. (d) The beta-dot system pro-
vides a natural inner loop for an IFFC system. A maximum of 2 g's can be commanded in the
direct sideforce maneuver. However, at most flight conditions the canard effectiveness or struc-
tural limits are not adequate to generate that amount of force while sideslip is being constrained to
zero. Consequently, when the canards do saturate and the recoupling option has been selected, the
canard-to-rudder decoupling signal is reduced to allow sideslip to develop. In most cases, this side-
slip will generate enough additional sideforce to satisfy the pilot command.

The descriptions of the advanced flight control modes of the AFTI/F-16 were obtained from Ref-
erence 11.

.
~110

r'
 NADC-82240-60

THE ADVANCED FLIGHT CONTROL ACTUATION SYSTEM (AFCAS) PROGRAM

The development of Advanced Flight Control Actuation Systems (AFCAS) for next generation air-
craft has been a joint undertaking by the Navy and the Rockwell International Corporation since
1972. The sixth and final phase of the program was completed in November 1979.

The complexity of flight control systems has increased until present initial costs and required
maintenance time are approaching prohibitive levels. This situation is due primarily to the design
philosophy that improvements and refinements are best achieved by adding on accessories and/or
components to proven, traditional systems. Broad new approaches and technologies involving ad-
vances in power generation, transmission, control, and actuation will be required to alleviate com-
plexity in future Navy aircraft The AFCAS Program is a significant step in this direction.

Phase I of the AFCAS Program established that a direct-drive flow control valve, modular con-
figured actuator, and a localized power package could be readily integrated into a computer-
operated, fly-by-wire system. Adoption of AFCAS concepts should enhance flight control system
maintainability, reliability, combat survivability, and lower initial costs, Reference 12.

Efforts to confirm the practicality of Phase I concepts were begun in Phase II with the design and
fabrication of an engineering model, 8000 psi (55 MPa), control-by-wire, modular configured
aircraft-type hydraulic servo actuator, Reference 13. Electrical inputs were applied to force (torque)
motors employing cobalt samarium permanent magnets. Motor output was connected directly to
single stage spool/sleeve type flow control valves. The force motors and flow control valves could
be integrated into dual tandem, dual parallel, or single actuator configurations.
Phase III involved conducting laboratory performance tests on the engineering model actuator(s)
built in Phase II, Reference 14. Static and dynamic tests were conducted on the force motors, motor/
valve subassemblies, electronic drive unit, and actuator assemblies including dual system tandem,
dual system parallel, and single system configurations. The dual tandem actuator was tested under
load. Major achievements accomplished in Phase III were:

- Successful operation of a direct electrical control "muscle" actuator for primary flight control
surfaces.
* Use of building-block elements to assemble dual tandem, dual parallel, and single actuator
configurations.
9 Successful operation of a control-by-wire hydraulic actuator, utilizing 8000 psi (55 MPa)
operating pressure.
Successful performance of a laboratory-type electronic drive unit which provided high immun-
4 ity to circuitry failures.
In Phase IV, an 8000 psi (55 MPa) control-by-wire, modular rudder actuator was designed and fab-
ricated for future flight testing on a T-2C airplane, Reference 15. Actuator design criteria were based
on T-2C aerodynamic considerations, envelope constraints, and single system hydraulics. Actuator
output was commanded by a single stage spool/sleeve valve driven directly by a permanent magnet
force motor. The force motor was powered by an electronic drive unit which received inputs from a

4 111
4I NADC-82240-60

force transducer in the rudder system and position transducers on the actuator. A localized hydrau-
lic power unit supplied 8000 psi (55 MPa) pressure for the rudder actuator.

In Phase V, a direct-drive control-by-wire muscle actuator, powered by a localized 8000 psi (55 MPa)
hydraulic system, was used to control the directional flight of a T-2C, Reference 16. Successful op-
eration of the test installation represented a significant milestone in the development of advanced
flight controls. No problems were encountered; the system functioned exceptionally well and pilot
response was favorable. The test results confirmed analyses and laboratory investigations reported
in References 12 through 15. The ease with which flight testing was accomplished verified that
AFCAS-type systems can be designed, fabricated and maintained without special techniques or
state-of-the-art advances.

In Phase VI the feasibility of Direct Digital Drive (D3 ) for an AFCAS was successfully flight tested
in the T-2C aircraft, Reference 16. The test installation contained a digitally controlled direct drive
rudder actuator, microcomputer, electronic drive unit, force transducers, and a localized 8000 psi
q (55 MPa) hydraulic power supply. The system met all laboratory and flight test objectives and
demonstrated direct microcomputer control of primary flight control surfaces. Phase VI demon-
strated an approach that will improve performance and reliability of FBW control systems by reduc-
ing system complexity.

DESIGN PHILOSOPHY OF THE AFCAS CONCEPT

Flight control systems need to be as simple, direct and foolproof as possible with regard to design,
operation, inspection and maintenance. This requirement, as defined by MI L-F-18372, continues
to be an excellent basis for the design philosophy of primary controls and needs to be stressed in
evolving fly-by-wire (FBW) systems. Basically there are two ways of achieving the level of reliability
required for fully powered primary flight controls. The first is to develop highly reliable parts for
use in the systems. The second is to design reliable systems utilizing redundancy techniques. Neither
approach leads to an acceptable solution of the immediate problem. Development and verification
of components having the necessary reliability is a long term project while indiscriminant system
redundancy leads to impractical complexity and unwieldy packages. A judicious blend of the two
approaches is necessary to effect an acceptable near-term system design.

The first step of fundamental importance in achieving high reliability in FBW control systems is
minimizing the number of component parts. Pursuing this approach leads to a sirqflified FBW sys-
tem utilizing dualized hydraulic power systems. To achieve reliability equal to or exceeding that
of its dual mechanical counterpart, the approach should follow the design philosophies used in con-
ventional mechanical systems. They are:

e Failures are never abrupt but are progressive over periods of extended use.
* System failures do not result in hardovers.
* The existance of failures does not degrade total system operation.
e Design margins are in excess of normally used values.
e Design to minimize maintenance actions precludes "Murphy Failures" or human errors.
* Success of the system is attributed to subsystem elements with high designed-in reliability
obtained with redundant configurations.

Fundamentally, design of FBW electronic circuits should center about the foregoing philosophy us-
ing feedback and redundant techniques. Present availability of small low cost electronic devices,

112
 NADC-82240-60

integrated circuits, (IC or LIC) permits consideration of this approach. Each function circuit would
contain several paths for independent operation similar to the present cable system. The goal is a
FBW system where the time between required maintenance actions approaches the aircraft overhaul
period.

DESCRIPTION OF THE AFCAS CONCEPT

The AFCAS concept employs control of the primary surface actuators directly by an onboard digi-
tal computer or by a standby processor. Command signals are applied directly to the single stage
control valve. Direct control of the surface actuator eliminates augmentation actuators. Actuator
power is provided by 8,000 psi localized electrically or mechanically driven hydraulic packages in
lieu of a centralized hydraulic system. Flexible building block components provides single or dual
system surface actuators having the desired simplicity and hardware commonality for standard act-
uator classification. A direct drive single stage servo valve is utilized which eliminates contaminant
sensitive small nozzles, flappers and jet-pipes. The aim is to provide simple, rugged, reliable compo-
nents which can be integrated with an advanced digital flight computer into a complete FBW system
for future advanced aircraft.

The control valve employs a highly reliable single stage design combining high performance with
simplicity and ruggedness. A single moving part consisting of a spring centered, 4-way spool is driven
by a high output torque motor. The basic valve is compatible with single and dual actuators. Re-
placement of the conventional two-stage hydraulic servo valve (8 ma control current) with a high
force single-stage concept results in a significant improvement in reliability because small moving
parts that are susceptible to minute fluid contamination are eliminated, and the fragile fine wire
wound torque motor is replaced with a large ruggedized magnetic coil.

The evolution of solid-stage power amplifiers has resulted in devices, such as silicon controlled rec-
tifiers and switches, that can now effectively and reliably handle the power required in going from
low power control signals to the relatively high power level necessary for single stage valves.

This approach is centered on the elimination of the hardover potential by:

a Developing sufficient force on the val 9 spool to insure shear-out of all contaminants.
e Design of the valve torque motor coils such that an open or short will not cause a hardover.
* Design of torque motor drive circuitry which precludes hardover conditions due to component
failures.
* Redundant feedback connections to preclude hardovers due to an open transducer.

The AFCAS configuration consists of a ..roportional control loop utilizing a 4-way spool and sleeve
valve driven by a high output torque motor. The spool has two lands and is flow force compensated.
Spool travel is 0.010 inches to insure null tracking over the entire operating temperature range. Flow
forces are below one pound. The torque motor produces at least 40 pounds drive force on the spool
at null position. The torque motor is kept dry to eliminate accumulation of contamination at the
permanent magnet. The spool centering springs have a sufficiently high rate to insure positive
centering at electrical zero and to provide sufficiently high frequency response. The valve housing
is compatible with modular dual actuators utilizing rip-stop desigr principles. The dual spools have
provision for rigid mechanical synchronization. The entire valve, including housing is made of
steel. The valve assembly weighs less than 5 pounds. Figure 77 shows the salient features of the
direct drive control valve package.

113
* NADC-82240-60

0 a,
0 Cu
U
LI.l CU
c.J
o
L~. CU

-5
_____________ C
0
C.,
a,
L.
0
4-
U
a,
1~
0

________________ 1~.a,

LL

-I
0

114
 NADC-82240-60

SERVO VALVE DESIGN REQUIREMENTS

The following general servo valve requirements were established during Phase I of the program to
facilitate evaluation of the various options available and to provide a common basis for comparison.
The requirements were kept as general and loose as possible to prevent unnecessary constraints.

* Single Stage

The valves shall incorporate only one hydraulic stage. Tt..3 is a basic goal of the proposed
approach. The aim is to eliminate Small orifices and passageways which can clog and produce
hard over failures.

9 8,000 PSI

The system operating pressure shall be 8,000 psi. VHP study effort to date indicates future
aircraft should utilize high pressures to effect considerable savings in weight and space of
hydraulic system components.

* Dual Tandem

The valve design must be compatible with dual tandem actuators. The proposed actuation
system approach consists of two channels, and two hydraulic power output elements con-
trolled by two electrical commands driving two valves connected in dual tandem.

e Shock

The valve shall be inherently rugged and able to withstand 1,000 g's shock in any axis.

0 Weight

The weight goal for the entire valve assembly shall be less than 5 pounds. This represents a
reduction of about 1 pound when compared to typical mechanical input surface actuator
valves.

e Flow

The basic design shall provide flow rates up to 15 gpm by resizing orifice area gradients and/or
flow compensation. The goal is to standardize valve parts. All classes of valves will be assem-
bled using common housings, torque motors, mountings, etc. Adjustable mechanical stops
shall be utilized to reduce the number of spool-sleeve designs to a minimum.
* Null Leakage

Null leakage shall be less than 0.5% of rated flow at 8,000 psi. This level is selected as a com-
* promise between power loss, overlap, and manufacturing tolerances.

e Natural Frequency

The natural frequency shall be at least 35 Hertz. This value places the valve at least one order
of magnitude beyond the actuator loop frequency thereby assuring minimal influence and
ease of loop design.

115
oQ NADC-82240-60

* Hysteresis

The hysteresis shall not exceed 4 percent of the rated input signal.

e Dead Zone

Dead zone in the flow vs. input current characteristic curve shall be less than 4%.

e Valve Stroke

The spool stroke shall be 0.040 inch.

* Life

The valve shall withstand 4,500,000 cycles at rated pressure and 2% of rated flow, and 500,000
cycles at 100% rated flow.

- Spool Drive Force

The electromechanical transducer shall provide at least 40 pounds spool drive force at null.
This requirement is a compromise between sufficient force to shear out material or contamina-
tion which could enter an orifice and reasonable size transducers.

e Linearity

The flow vs. current plot shall be linear within 15% of rated input current.

* Null Shift

The change in the null point shall not vary more than 5% throughout the operating
environment.

AFCAS CONFIGURATIONS TESTED

Three types of actuation configurations were fabricated, assembled and tested during the third phase
of the AFCAS program (reference 14) and are listed as follows:

(1) Single System Actuator (Figure 78)

(2) Dual System Parallel Actuator (Figure 79)

(3) Dual System Tandem Actuator (Figure 80)

The "building block" concept developed in Reference 12 was used as a basis for the design of mod-
* ular elements which can be assembled to form any of the three actuation configurations listed
above. The two dual configurations are controlled by two mechanically synchronized Direct Drive
Control Valve Assemblies. During the laboratory tests of the three actuator configurations, an
electronic interface unit supplied the power required to drive the force motors. Independent
current-drive circuits for each of the four windings in each motor were provided for redundancy.
Figure 81 is a photograph of a force motor. Figure 82 shows the interface between the Electronic
* Interface Unit and a dualized Control-By-Wire Actuator.

* 116
NADC-82240-60

I'E

4~S

iQ

117n
I
NADC-82240-60

NC NE

118
NADC-82240-60

cc

.4r co

119U
C NADC-82240-60

0
J6L

a,

Ole-

120)
4 NADC-82240-60

I- - - - - - - - - - - - - - - - - -31

i.-

- * - Co4E
L

0~0 -~
fl 4ie

. 00

121.
l NADC-82240-60

The laboratory performance tests conducted during Phase III were successful and the major accom-
plishments achieved during this program phase are listed as follows:

a Successful operation of a direct electrical control muscle actuator for primary flight control
surfaces.
-- . Use of building-block elements to assemble dual tandem, dual parallel, and single actuator
* -configurations.
e Successful operation of a control-by-wire hydraulic actuator utilizing 8,000 psi (55 MPa)
operating pressure.
* Successful performance of a laboratory-type electronic drive unit which provided high immun-
ity to circuitry failures.

FLIGHT VERIFICATION OF THE AFCAS CONCEPT IN THE T-2C AIRCRAFT

In Phase IV, an 8,000 psi (55 MPa) control-by-wire, modular rudder actuator was designed and fab-
ricated for future flight testing on a T-2C airplane, Reference 15. Actuator design criteria were
based on T-2C aerodynamic considerations, envelope constraints, and single system hydraulics.
Actuator output was commanded by a single stage spool/sleeve valve driven directly by a permanent
magnet force motor. The force motor was to be powered by an electronic drive unit which received
inputs from a force transducer in the rudder system and position transducers on the actuator. A
localized hydraulic power unit was planned to supply 8,000 psi (55 MPa) pressure for the rudder
actuator.

" . Figure 83 is a photograph of the Rudder Actuator Assembly designed and fabricated during Phase IV
and used in the flight testing of the AFCAS concept during Phase V.

The objective of Phase V was to design, fabricate, and test a subsystem to verify the feasibility of the
AFCAS concept in the flight environment. The test system was installed and flown in a T-2C twin
engine turbojet trainer shown in Figure 84.

1. Technical Approach

The directional control system in a T-2C airplane was changed to a full-powered control-by-wire test
installation containing:

e Hydraulic rudder actuator

e Electronic drive unit
* Localized hydraulic power unit
* Force transducer

The existing hydraulic system was altered to operate at two pressure levels: 3,000 psi (21 MPa) and
8,000 psi (55 MPa). Engine driven pumps powered the 3,000 psi system in the usual manner. A
localized motor/pump unit was added to power the rudder system which was formerly operated
manually by the pilot. The original 3,000 psi and newly added 8,000 psi systems shared the existing
reservoir and return lines. The T-2C electrical system was altered to power the localized motor/pump
unit and electronic drive unit. The modified system functioned the same as the basic T-2C system
except the rudder was hydraulically powered instead of manually operated.

The original cable system between the rudder pedals and rudder was changed to incorporate the
control-by-wire test installation. The rudder pedal cables were attached to a sector which was pre-
vented from rotating by a force transducer. Force on the pedals was converted to a proportional

* 122
NADC-82240-60

I-

111
'Ivi

123~
NADC-82240-60

49o

124-
 NADC-82240-60

electric voltage from the transducer. This command signal was conditioned by an electronic drive
unit which powered a torque motor on the rudder actuator. The torque motor in turn operated a
single stage flow control valve on the actuator.

The direct-drive, 8,000 psi rudder actuator designed and fabricated in Phase IV was modified to in-
corporate a bypass valve. This device allowed the rudder to seek the trail position if system pres-
sure were lost. In the event of a "hard-over" type electronic failure, the pilot could permit the rud-
der to trail by turning the 8,000 psi motor/pump unit "off."

The electronic drive unit was designed, fabricated, and packaged to be a flightworthy assembly. The
unit had dual channels with sub-circuits which were dualized. The circuitry was designed with re-
dundancy features which provided high immunity to component failures.

Requirements were established for an 8,000 psi localized hydraulic power supply. The pump used
was the same unit employed for flight testing in the Lightweight Hydraulic System (LHS) develop-
ment program except delivery was reduced to match rudder actuator flow rates and to lower input
power requirements. The pump was mated to an off-the-shelf, aircraft type 28 volt DC motor.

The force transducer incorporated in the test system was designed specifically for this application.
The transducer utilized two linear variable differential transformers mounted in series.

All major components in the test installation were assembled in the laboratory for integration test-
ing. Investigations were made to determine if detrimental pressure oscillations or surges were present
Motor current and system heat reiection were measured. Frequency response tests were conducted
on the actuator/system. Nine hours of simulated flight testing were performed to evaluate the en-
durance capability of system components.

The test system was installed in a bailed T-2C with instrumentation for monitoring pressures, flows,
temperatures, etc. Standard parameters such as air speed, altitude, engine RPM, etc., were also in-
strumented. Flight data were collected by photorecorder and telemetry systems.

Procedures were established for system checkout, ground demonstration, and flight testing. Approx-
imately ten hours of flight time were logged on the test system at various altitudes and airspeeds.
Pilot observations and instrumentation data were used as a basis for evaluating the AFCAS
installation.

2. T-2C Airplane

The T-2C "Buckeye" is built by the Columbus Aircraft Division of Rockwell International Corpora-
tion. The Buckeye is a two-place, subsonic trainer powered by twin turbojet engines. The aircraft
is designed for both land and carrier based operations. Distinguishing features include wide-track
tricycle landing gear, straight tapered wings, and low slung intake ducts.

The T-2C is used as a basic trainer for military pilots, and is equipped for cross-country flight, night
. flying, and low altitude, high speed navigation exercises. Maximum level flight speed of the Buckeye
is 465 knots (239 m/s) at 15,000 feet '4.6 kin); the service ceiling is 45,000 feet (13.7 km). Take-
off and landing speeds are in the ranqa of 95 to 110 knots (49 to 57 m/s). A typical take-off gross
weight is 13,000 pounds (5,900 kg).

Dual power sources are provided for the electrical, hydraulic, and air conditioning systems. The
flight control system includes hydraulic full-powered ailerons, a boosted elevator, and an electric

125
[k NADC-82240-60

trim system; rudder operation is manual. The aileron and elevator actuators are part of mechanical
linkage connecting the pilot's stick to the control surfaces. Thus, in the event of a hydraulic system
malfunction, control of the aircraft can be accomplished manually.

The T-2C has a 3,000 psi (21 MPa), Type II (-65 to +275 0 F) (-54 to +135 0 C) single hydraulic sys-
tern. Two pumps, one on each engine, provide power to operate the landing gear, speed brakes,
arresting hook, aileron actuator, and elevator boost package. The pumps are constant pressure,
variable delivery, axial piston designs. Each pump is capable of delivering 4.9 gpm (18.5 L/m) at
7,800 rpm. Hydraulic fluid (MI L-H-5606) is supplied to the pumps by an air/oil type reservoir
pressurized by engine bleed air. Fluid cleanliness is maintained by 5 micron absolute filters.

One pump can adequately handle all flow demands. However, if supply pressure should drop below
1,800 psi (12 MPa), a priority valve is used to insure operation of the aileron and elevator actuators.
A cockpit controlled shutoff valve is installed in the aileron/elevator subsystem to permit simulat-
ing loss of power for training purposes. The landing gear and arresting hook can be lowered and
locked by gravity, if desired. The wheel brakes have an independent hydraulic system.

Electrical power is sipplied by two 28 volt DC 300 ampere starter-generators, one mounted on each
engine. The generators are connected for parallel operation and power the primary bus. Output
voltages are regulated for varying loads and engine speeds.

Two nickel-cadmium 24 volt re-chargeable batteries are used for engine starting and emergency DC
power. The batteries are normally connected in parallel, but are used in series for engine starting.

A portion of the 28 volt DC power is converted to 115 volt 400 Hz AC power by two rotary in-
verters. Inverter No. 1 produces 500 volt-amperes for instruments; inverter No. 2 generates 1,500
volt-amperes for avionics and serves as a backup source for instrument power.

3. AFCAS Flight Test Installation

The directional (rudder) system in a bailed T-2C (BuNo. 152382) was changed from a manual to a
full-powered control-by-wire system for the AFCAS program. Principal components in the test in-
stallation were:

* Hydraulic rudder actuator

e Electronic drive unit
-.- . Localized hydraulic power unit
9 Force transducer

Modifications required in the T-2C to accommodate the new installation are discussed in the follow-
ing sections under four general headings: mechanical system, hydraulic system, electrical system,
and instrumentation.

a. Mechanical System

Elements of the mechanical system are depicted schematically on Figure 85. The salient mechanical
components of the T-2C rudder control system are listed as follows:

a Sector assembly
*e Sector support

* 126
 NADC-82240-60

00 E

0 0

u 1-1

'044 0 V C-4
0c $JJ 05
0U 0 to '
$4 0 x

0~ *10 U . -4 .- E.~
f
4

-4 Q. 14 -H 9
44 0 o /

C
oc
g
E

E-4-

pq.

rAu

127
 NADC-82240-60

o Bellcrank assembly
o Rudder actuator mounting bracket
o Force transducer mounting braket

The T-2C rudder has a travel of ±250. For safety reasons, rudder travel was reduced to ±120 in the
test installation by limiting actuator stroke. This permits the pilot to land safely with a "'hard-over"
rudder, opposite engine out, and three knot cross-wind.

The relationship between pedal force and rudder travel is approximately 7.5 Ib/deg (33.3 N/deg) of
rudder movement or 90 lb (0.4 kN) for full travel (120). Pedal displacement was small, approxi-
mately 0.50 in. (13mm), since the force transducer length changed only 0.025 in. (0.63mm) for full
rudder travel (pedal displacement was due primarily to cable stretch). In the original manual control
system, pedal displacement was approximately 4 in. (10.2 cm) for full rudder travel (250).

The maximum hinge moment normally applied to the T-2C rudder is based on pilot strength and is
2,200 lb-in (249 N-m). Maximum rudder deflection a pilot can achieve thus depends on air loads
present. The AFCAS rudder actuator can develop 13,000 lb-in (1,470 N-m). Because of the limited
rudder deflection (12' max.) the high moment capability of the rudder actuator required only minor
adjustment in the T-2C flight envelope to assure safety.

b. Hydraulic System

The original and modified hydraulic systems are compared schematically on Figure 86. Major
changes required in the T-2C hydraulic system to accommodate the test installation were:

o Addition of an electric motor driven 8,000 psi (55 MPa) variable delivery pump
o Addition of an 8,000 psi control-by-wire rudder actuator and bypass valve
e Addition of a suction line from the reservoir to the 8,000 psi pump, pressure line from the
pump to the rudder actuator, and actuator return line

. Addition of pump case drain return and shaft seal overboard lines
a Relief valve installed in the 8,000 psi system
o Heat exchanger installed in the 8,000 psi pump case drain line

The modified system is shown schematically on Figure 87. The 3,000 psi (21 MPa) and 8,000 psi
(55 MPa) systems shared a common reservoir and common return lines. All major components,
. except for the rudder actuator, were located in the fuselage compartment above the engines.
c. Electrical System

The electronic drive unit was designed and fabricated by the Columbus Aircraft Division of Rock-
well International Corporation. Circuit concepts employed in the unit were developed under com-
pany funded I R&D projects. Innovative application of redundancy and feedback techniques permit
EDU operation to be maintained with multiple component failures. Although the assembly was de-
signed and fabricated to be suitable for flight, the EDU was nevertheless an experimental model.
The assembly contained discrete components, test points, and external adjustments to facilitate data
acquisition. This resulted in a much larger package than would be needed for a production unit.
A production design EDU would have approximately 5% of the volume of the AFCAS unit.
i

4
128
 NADC-82240-60

RESERVOIR

Hydraulic Power
Shut-Off Valve

LATERAL

3000
PSI (ENGINE 30 S

80 PSI D000VPS 3000 PSICN.SS

IZPUMP PUM

MODIFIEDGEA
HYSYLCS.SE

4 Fiure 6. Oiginl an ModfiedHydraulic ystes

129v Sht-f
'Ul NADC-82240-60
AUX ILI AR Y

AU
IR ~PRESS
ENG INE --
XF!tIL
ROATOR

RUE~- AIR/OIL
All FRESERVOIR

VALVIg FILTER

-
r DIRECTIONAL AXIS
I TEST INSTALLATION

J
,J ~ THERMIAL SWITCH

ELC. T

R.H. ; S L.H. s
PuM [~ pjgp'FILTER T
3000 CD "7 3000 CD
PSS Io--L . PSII j

00I
000PSI
EXTf.PRESS VALVET UM
DISCONNECT
I I HEAT

SNUBBERI
I' p

PRESS.
SWITCH
RELIEF VALVE
FILTER
TO P
COCKPIT
I GAGE

AILERON ACTUATOR -
PRIORITY
ACTUATOR
WLVE ITY ==-
RELIEF VALVE

UTILITY SYSTEM -
SPEED BRAKES J I-
LANDIN~C GEAR &
ARRESTIGGA RRESING ELECTRONIC
EARDRIVE
CODE: UN IT

____ Pressure Line

Return Line
. - - Supply Lihe
,------ Electrical Connection TO
RUDDER
PEDALS FORCE
TRANSDUCER

Figure 87. Schematic Diagram of Modified Hydraulic System

130
 NADC-82240-60

The EDU was powered by 115 volt 400 ^. AC and basically had two channels with dual sub-circuits
(4 channels total). Bias pots were provided to adjust the input, feedback, and balance of each chan-
nel. Two power supplies provided ± 15 VDC for the signal amplifiers and system transducers. All
circuitry was contained on two identical printed circuit boards.

The force transducer housing assembly was designed and fabricated by CAD. The housing contains
two DC-operated LVDT force transducers built by Schaevitz Engineering in Camden, New Jersey.
The units have a maximum capacity of 500 lb (2.2 kN), a spring rate of approximately 8,000 lb/in
(1.40 MN/m), and an output of 0.01 v/lb (0.002 v/N) in tension or compression.

A simplified block diagram of all elements in the system is shown on Figure 88. Pilot inputs are
transmitted through the rudder pedals via cables, pulleys, and bellcranks to the force transducer
located inside the vertical stabilizer. Gearing multiplies pilot effort by 2.28. Transducer output is
the command signal (ei ) to the EDU. Amplifiers in the EDU process ei with a feedback signal (efb)
and power the force motor coils which drive the spool X i in the control valve. The valve ports
8,000 psi hydraulic fluid to the rudder actuator in response to X i . Actuator piston travel is sensed
by position transducers having an output of 5 v/in; this is the feedback signal (efh). Actuator piston
travel (± 1.75 in. max.) is converted through bellcrank and push rod to angular travel of the rudder
(+ 12 max.).

A simplified diagram of electrical tomponents in the test installation is presented on Figure 89.
System redundancy is illustrated on Figure 90. The system concept developed under CAD I R&D
studies is flexible in that various levels of redundancy could be employed (as required) for other
applications. AFCAS redundancy features include:

* Dual force (input) transducers

e Dual position (feedback) transducers
* Dual power supplies
* Quad electronics
" Feedback fault correction

Each of the four power amplifers in the EDU employes current feedback with a highly reliable
darlingto.,- power transistor configuration and independent power supplies. The circuitry is designed
so that in the event an output stage fails "hard-over," voltage applied to a motor coil will not exceed
its rated value. This limiting feature permits a subunit failure to be compensated or nullified by
another subunit. Closed loop tests reported in Reference 14 verified that operation of the redundant
subunits provided high immunity to component failures.
A math model of the idealized system is presented on Figure 91. The transfer functions are for
"small signal" inputs and do not reflect fluid flow saturation limitations or motor current limita-
tions imposed by coil inductance. System spring-mass effects (actuator loaded) were not included.
Optimum loop gain was 90; this provided a theoretical band width of 14.3 Hz and a damping ratio
of 0.7.

Performance characteristics of the test installation were higher than could be utilized in the T-2C
rudder system. To assure satisfactory operation, AFCAS dynamics were matched with T-2C di-
rectional system dynamics. This was accomplished by lowering loop gain to 20 and adding high
frequency roll-off filtering to reduce the possibility of system noise.

4131
 NADC-82240-60

bE

xS

n ..

rI - -

SI -I-132
".,1

r--

iL
 NADC-82240-60

U. Q0

00

o C
0
CL z

00

:Z- - j

Qc o 0
I- 0
2~~~- - 0 : 4 J

cm
I- u 2 w
C= U.

U(U

w..

'n, C6U

133
 NADC-B2240-60

I-AA

134 * ** ,
 NADC-82240-60

1-4

..
,4

00

C0

:oi
09 - I,,
-

I I "
En
- 'q 04 - 0,
+ "

'U

z -

135-' (U
U) 430t Z

c-c

a 0.
I0
!U.
= ,
-
 NADC-82240-60

d. Instrumentation

The T-2C was equipped with several flight data acquisition systems. Two were used in the AFCAS
program: (1) an 18 channel telemetry system, and (2) a 21 hole photo recorder system. The
telemetry oscillator package was located in the aft cockpit seat area; the photo recorder was in-
stalled inside the nose.

Telemetry data were recorded at the CAD Telemetry and Data Processing Center where a UHF
receiving/tracking system provided rJl-time data acquisition and direct read-out on strip charts.
Audio communication with the pilot was available for convenience and safety monitoring.

Pilot instrumentation controls were located above the cockpit instrument panel and on the control
stick. Data in the two recording systems were related by means of correlator numbers printed on
the photo recorder film, and correlator blips on the TM strip chart. A correlator counter could be
read by the pilot for reference purposes.

New equipment installed to permit the pilot to monitor and control the AFCAS system were:

. An indicator was provided for direct readout of the motor/pump discharge pressure
* A switch was provided to turn the motor/pump unit "on" and "off"
e An "oil hot" light was set to illuminate when hydraulic fluid in the motor/pump suction line
exceeded approximately +200°F (93 0 C)

4. Flight Plan

The primary objective was to verify the feasibility of the Advanced Flight Control Actuation System
(AFCAS) concept by flight testing a control-by-wire, direct-drive actuation system powered by a
localized 8,000 psi (55 MPa) motor/pump unit. Demonstration of flying qualities was not part of
the program, however, pilot comments were encouraged. Ten flight hours were expected to be suf-
ficient to evaluate AFCAS performance, confirm prior analyses and laboratory testing, and provide
a measure of confidence in system reliability.

The flight plan was designed to determine directional control characteristics at several altitudes up
to 30,000 ft. (9.1 km) and various speeds up to 340 knots (174 m/s). The first two flights were
. dedicated to confirming satisfactory operation. Subsequent flights were scheduled to evaluate sys-
tern performance and reliability while accumulating 10 flight hours. Flight plan details are given in
Appendix A of Reference 16.

5. Flight Test Results

The AFCAS flights are summarized on Table V of Reference 16. Two pilots participated in the pro-
gram and prepared reports detailing each test flight. Additional comments were made during flight
de-briefings. Both pilots stated that performance of the AFCAS test installation was completely
*satisfactory. Comments made by the pilots concerning their flights were:
* * * The AFCAS installation worked exactly as designed
e No malfunctions occurred
e System pressure was steady
* Hydraulic fluid temperatures were normal

136
 NADC-82240-60

a Directional control response was judged to be superior to the production T-2C

* Pilot adaptation to "force control" of the rudder was quickly and easily acquired. Reaction
of the aircraft provided the clues to close the loop.
a The force system had an advantage during take-offs and landings in high cross-winds. The
fixed pedals provide full rudder and allow much easier braking (in combination) without severe
leg and foot extension that is required for conventional deflection controls.

6. Summary

A direct-drive control-by-wire muscle actuator, powered by a localized 8,000 psi hydraulic system,
was used to control the flight of a T-2C. Successful operation of the test installation represented a
significant milestone in the development of advanced flight controls. No problems whatsoever were
encountered; the system functioned exceptionally well and p,,ot response was favorable. The test
results confirmed analyses and laboratory investigations reported in References 12 and 15. The ease
with which flight testing was accomplished verified that AFCAS type systems can be designed, fab-
ricated, and maintained without special techniques or state-of-the-art advances.

The AFCAS concept is intended for application to automatic, computer operated flight control sys-
tems. The AFCAS flights described in Reference 16 did not demonstrate the full performance
capabilities of the test hardware since the T-2C did nothave computer operated controls. Company
funded investigations at the Columbus Aircraft Division verified the feasibility of controlling
AFCAS actuators directly by a digital computer.

FLIGHT VERIFICATION OF DIRECT DIGITAL DRIVE FOR AN AFCAS IN THE T-2C

Al RCRAFT

The objective of the sixth and final phase of the AFCAS program was to demonstrate that AFCAS-
type actuators can be directly controlled by a digital computer. The computer control was demon-
strated in system laboratory tests and in flight tests in a T-2C twin-engine turbojet trainer.

1. Technical Approach
The directional control system of a T-2C aircraft was changed to incorporate a full-powered Digital
Fly-By-Wire (DFBW) mode with an Analog Back-Up (ABU) mode. The test installation contained:

9 Hydraulic rudder actuator

* Localized hydraulic power unit
* Digital microcomputer
* Electronic drive unit (EDU)
- Associated sensors, wiring and power supplies

The original cable system between the rudder pedals and rudder was changed to incorporate the fly-
by-wire test installation. The rudder pedal cables were attached to a sector which was prevented from
rotating by a force transducer. Force on the pedals was converted to a proportional electrical sig-
nal from the transducer. This command signal was supplied to a microcomputer where it was summed
with a feedback signal, and processed into a pulse width modulated (PWM) error signal. The PWM
signal was power amplified in the EDU which powered to torque motor of a direct drive hydraulic
rudder actuator. The modified system provided a microcomputer controlled, hydraulically powered
rudder, instead of the manually operated rudder of the basic T-2C aircraft
The hydraulic system, the direct drive actuator, the EDU, the LVDT actuator position feedback
transducer, and the pedal force transducers were installed and flight tested in the T-2C during

137
*I NADC-82240-60

Phase V. The Phase VI system was designed so that the signals from the transducers could be
switched from the microcomputer unit directly into the EDU to provide an ABU mode with the
same control capability as the Phase V flight system. The ABU mode could be selected manually,
or selected automatically if the microcomputer monitor detected abnormal operation.

The safety provisions of the Phase V program were included in this phase. The direct-drive 8,000
psi (55 MPa) rudder actuator, designed and fabricated in Phase IV, was equipped with a bypass
valve. This device allowed the rudder to seek the trail position if system pressure were lost. In the
event of a system failure, the pilot could permit the rudder to trail by turning the 8,000 psi (55 MPa)
supply "off."

All major components in the test installation were assembled in the laboratory for integration test-
ing. System operation was verified in the laboratory prior to aircraft installation. Frequency re-
sponse tests and temperature-altitude tests were performed simulating aircraft operation.

The test system was installed in a T-2C with instrumentation for electrical and hydraulic operation.
Standard parameters such as air speed, altitude, engine rpm, etc., also were instrumented. Flight
data were collected by photorecorder and telemetry systems.

Procedures were established for system checkout, ground demonstration, and flight testing. More
than four and one-half hours of flight time were logged on the test system at various altitudes and
airspeeds. Pilot observations and instrumentation data were used as a basis for evaluating the test
system.

2. Description of the AFCAS Direct Digital Drive (D3 ) Test Installation

The fly-by-wire rudder control system test installation, originally installed in the T-2C aircraft during
Phase V of the AFCAS program was modified to test a digital microcomputer generated PWM valve
drive signal (Phase VI of the AFCAS program). Principal components in the test installation are:
:" e EDU
* EDUPreviously installed and
* Localized Hydraulic Power Unit (8,000 psi) tested per Phase V of
e Force Transducers the AFCAS program.
e LVDT Position Transducers

e Microcomputer Assembly Installed and tested per

Phase VI of the AFCAS
o Microcomputer Power Supply program.

Two modes of system operation are provided, the DFBW mode and the ABU mode. In the DFBW
mode, the microcomputer converts the pedal force command and rudder position feedback outputs
into digital signals which are summed, amplified, and converted into PWM signals. The PWM signals
are sent to each of two channels in the EDU where the signals are amplified and power converted
into four torque motor PWM currents. In the ABU mode, the pedal force commands and rudder
position feedback outputs bypass the microcomputer and are connected directly to the EDU where
they are summed, amplified, and power converted into four torque motor currents.

Figure 92 shows a simplified block diagram of the T-2C Direct Digital Drive AFCAS test installation.
The DFBW engage/disengage functions are implemented by a cockpit control switch located adjacent

138
NADC-82240-60

00

to
adU

E- Z'
(D
CF

-4-

czv

N~ W

0 w 0

11
 NADC-82240-60

to the AFCAS power switch on the Pilot's Auxiliary Instrument Control Panel. After the AFCAS
analog system has been engaged, DFBW control will be selected by momentarily holding the DFBW
engage switch to "ON". The DFBW control relays will energize to switch the transducer outputs to
the microcomputer, and connect the microcomputer output to the EDU. When the microcomputer
functions are operating correctly, a +28 volt DC power ground will be supplied to the holding coil
of the cockpit DFBW engage switch. Disengagement of DFBW control will result from the follow-
ing: manually selecting the DFBW engage switch to "OFF", automatically by the loss of the micro-
computer supplied ground, or by manually selecting the AFCAS power switch to "OFF".

a. Microcomputer Assembly

The microcomputer assembly is housed in an enclosed unit, and consists of the following subassem-
blies:

* Motorola Mono-Board Microcomputer Module

e Analog-To-Digital (A/D) Converter Module
* Digital-To-Analog (D/A) Converter Module
* Card Cage & Mother Board Assembly
e Signal Conditioning Board

The mono-board microcomputer module is a complete computer-on-a-board having all the process-
ing and control required for a microcomputer-based system. It incorporates the MC 6800 MPU, 1 K
of Random Access Memory (RAM), provisions for 4 K of Programmable Read Only Memory (PROM),
timing and control, buffers, an Asychronous Interface Adapter (ACIA) and two Peripheral Interface
Adapters (PIA).

The A/D converter module consists of eight channels of A/D conversion of which four are utilized.
The D/A converter module consists of four channels of D/A conversion of which three are utilized.

The signal conditioning board contains four channels of sensor signal conditioning and a relay driver
that interfaces the microcomputer monitor output with the system control logic.

Additional information on the microcomputer assembly is contained in Appendix A of Reference

17.

A separate power supply converts single-phase, 115 VAC, 400 HZ to + 5 VDC, and + 12 VDC to
power the microcomputer assembly.

b. Software Description.

Software was developed to enable the microcomputer to perform two basic functions; a command/
feedback control function and a control monitor function.

The command/feedback control function sums the pilot command and rudder position signals to
produce an output signal proportional to the difference to drive the actuator.

The control monitor function measures the level of error between the pedal command the the
rudder actuator position feedback, and if a preset level is exceeded for a given period of time, the
engage command will be removed. Actuator control will then revert to the ABU mode. A contin-
uous check is also made on the transducer input A/D conversion hardware by comparing the two
digital feedback signals with each other and in a similar manner comparing the digital pedal signals.
Any differences exceeding preset levels for a given period of time will result in switching system
control to the ABU mode.

140
 NADC-82240-60

The control monitor function was incorporated as the Motorola microcomputer is a single channel
device which could generate a "hardover" command under certain failure conditions. The dual-
channel redundancy of the ABU mode prevents a "hardover" command of the rudder even if a
pedal transducer or rudder position transducer fails in a "hardover" condition.

c. Program Modules

The DFBW Microcomputer Program Flow Chart, Figure 93, illustrates the modular nature of the
software and the sequence in which the modules function. The program modules were designed,
coded, and initially checked as individual entities prior to being integrated.

POWIER
UP/RESET

INITIALIZE

INPUT-

ERROR

OUTPUT

INPUT-2

-4

F ONITOR

Figure 93. Digital Fly-By-Wire Microcomputer Program Flow Chart

141
 NADC-82240-60

Following is a brief description of the program modules:

Initialize - The Initialize module sets the D/A Converter (DAC) channel 4 to provide +5 VDC to
hold in a relay. The relay, in turn, holds the DFBW Engage switch in the;engage position. This
module also sets timing counters to ensure that the Monitor function does not immediately turn off
the DFBW Engage switch.

Input 1 - The Input-1 module, as the first in the repetitive loop, is used to start the PWM output
signals. This is done by setting-both DAC-1 and DAC-2 at +10 VDC. It then controls the A/D con-
versions of pilot command (CMD1) and rudder position (POS1). Inputs are scaled so that full scale,
±120 of rudder is t5 VDC, which is one-half of full range for the A/D channels. Since the force
transducer that provides CMD1 is not mechanically or electrically limited to ±5 VDC, a software
limit is provided to set CMD1 at either ±5 VDC, as appropriate, when that value is exceeded. Out-
put of the A/D converter is a 12 bit word, proportional to the voltage.

Error - The error module performs a double precision subtract of CMD1 from POS1 and sets com-
puter gain through a series of shifts. It then determines polarity of the error and transfers to the
appropriate output module.

Output - The output module sets countdown timers that establish the duration of the plus and
minus portions of the PWM output signal. It switches DAC-1 and DAC-2 to -10 VDC when the
"positive" counters have timed-out. When the "minus" counters time-out, it transfers control to
the Input 2 module.

Inpu 2 - The Input 2 module controls the conversion of CMD2 and POS2 and provides limits on
CMD2 in the same manner as Input 1. CMD2 and POS2 are for use in the Monitor functions.

Monitor - The Monitor module compares the redundant pilot command and rudder position input
signals. If a difference in either of 1.50 is detected for a period of 0.128 seconds, the program is
set to deenergize the DFBW holding relay and reverts control of the system into the ABU mode.
The monitor also checks the magnitude of the error signal. If it exceeds 1.50 for 2 seconds, the
DFBW holding relay is deenergized, and control of the system again reverts to the ABU mode. As
long as the monitor does not detect an error, it transfers control back to the Input 1 module.

d. Flight Test Program Software

Support software, trade name "Microbug ROM" was purchased with the microcomputer equipment
and enabled communications with the microcomputer via a Teletype Corp. Model 33TU teletype
keyboard/printer reader/punch unit. An RS-232-TO-TTY adapter unit provided the interface
through the Asynchronous Interface Adapter (AC IA), between the microcomputer and the teletype.

The communication consisted of entering both program and simulated input data, monitoring
microcomputer operation, and dumping of programs onto paper tape for storage.

After the software modules were operating satisfactorily they were then merged to become an
operational program. After checking the operational program the microcomputer integrated into
the rudder system (in the laboratory), the program was then loaded into a PROM. The PROM was
installed in the microcomputer and the operational program verified. All subsequent final system
response testing and calibration for the flight configuration was performed with this PROM installed
in the microcomputer.

A listing of the flight program software is contained in Appendix B of Reference 17. The program
was designed to function at a rate of 500 HZ, and occupies 462 bytes of the available 4096 bytes
of PROM and 18 bytes of 1024 bytes of "scratch pad" RAM. The PROM map is also shown.

* 142
 NADC-82240-60

e. Microcomputer Analog and Reduced Bit Resolution Program Software

In addition to the PROM software developed for the flight test program, additional software pro-
grams were developed to evaluate microcomputer performance as a function of reduced bit resolu-
tion and various PWM frequencies. These programs were designed for lab testing only, and were
therefore stored on paper tape and entered into the microcomputer memory via the teletype.

f. Laboratory Tests

Laboratory tests were performed to integrate the microcomputer into the existing Phase V AFCAS
system, evaluate system performance and compatibility with the microcomputer operational modes
and software, and to ensure the equipment would function properly in the T-2C aircraft environment
prior to installation in the aircraft.

The DFBW and ABU system operating modes were evaluated in the laboratory. In the ABU mode,
the microcomputer is disconnected from the loop, and the aircraft sensors directly control the actu-
ator through the EDU. This was the control used in Phase V of this program. Laboratory testing
was also accomplished with the microcomputer supplying an analog signal instead of a PWM signal
to the EDU. This testing was performed to obtain comparison data with the computer in the loop.

The actual aircraft hardware was used whenever possible in the lab test set-up to permit testing and
evaluation of the flight hardware and to eliminate potential problems during subsequent installation
and operation in the aircraft.

Included in the lab test set-up were the rudder LVDT feedback transducers, rudder actuator, EDU,
microcomputer and associated power supply, and the control panel switches and control relay used
in the aircraft. The 8000 psi (55 MPa) hydraulic pump, incorporated in the aircraft for the AFCAS
installation, was not included in the lab set-up since the pump performed successfully in the previous
Phase V of the AFCAS program and the potential benefits from including it in the lab set-up did not
warrant the added expense. An 8000 psi (55 MPa) laboratory hydraulic pump was utilized for all
tests requiring high pressure hydraulic flow.

A block diagram of the laboratory set-up is shown in Figure 94.

'.-43
1
 NADC-82240-60

E
E4-

uIE-4
9;x4 a."* I

E- Wdc a

i
W ____
_

___
I_
a _ __

Oz 00 -

IV 0 u -j

pa E______

IU
H MZ:: -

4c
iI~I
0:
4 1d
 NADC-82240-60

3. Flight Test Program

a. Flight Plan

The primary objective of the Flight Test Program was to evaluate performance of the computer-
controlled direct digital fly-by-wire characteristics of the AFCAS in a T-2C aircraft. Approximately
three flight hours were expected to be sufficient to evaluate performance, confirm prior analyses
and laboratory tests, and provide a measure of confidence in system reliability.

The flight plan was designed to determine digital control characteristics at altitudes up to 30,000
feet (9.1 km) and speeds up to 340 knots (175 m/sec.) and to compare performance between the
DFBW and ABU modes of operation.

Details of the flight plan are included in Appendix C of Reference 17.

b. Flight Test Results

Three flights were flown for a total of 4.7 hours. The pilot stated that performance of the Direct
Digital Drive AFCAS test installation was completely satisfactory. No difference in "feel" was
noted between the DFBW and ABU modes of operation. Additional pilot comments were similar
to those listed in the Phase V AFCAS flight test program contained in Reference 16, and included:

- Directional Control Response was judged to be superior to the production T-2C.

- Pilot adaptation to "force control" of the rudder was quickly cnd easily acquired.
- The fixed pedals provide full rudder and allow much easier braking (in combination) without
severe leg and foot extension that is required for conventional deflection controls.
- Hydraulic system fluid pressure and temperature were normal.
- No malfunctions occurred.

c. Flight Program Summary

Following is a summary of the Direct Digital Drive AFCAS flight program:

FLIGHT MAX. AIR SPEED & ALTITUDE NZ DURATION

- 1 250 KOAS & 20,000 FT. 3.0"g" 1.4 Hours

2 250 KOAS & 20,000 FT. 3.0"g" 1.5 Hours
3 340 KOAS & 30,000 FT. 5.5"g" 1.8 Hours

ROCKWELL'S STUDIES OF PULSE MODULATED DIRECT DIGITAL DRIVE (D3 ) CONCEPTS

In advanced aircraft using digital fly-by-wire control, direct-drive actuators show considerable advan-
4. tages because of their ruggedness, relative simplicity and high reliability. Both the Navy and Air
Force have research programs under way to develop direct drive actuation.

Previous work on this project has demonstrated that low level analog signals, such as are provided by
a D/A converter, can be suitably amplified into the high power forms required for closed loop con-
trol. Analog valve drivers using quadruple redundancy and optimized for linear (Class A) operation,
_ were successfully tested and demonstrated in the laboratory. A modified PDP-1 1 minicomputer with

4 q145
 NADC-82240-60

built-in A/D and D/A converters was used to close the loop both internal and external to the mini-
computer. Flight worthy analog valve drivers were developed with I R&D funds. This design was
fabricated and flown under Phase V of the AFCAS program.

Following Phase V,additional I R&D research was conducted to develop an approach that would
(a) eliminate the need for D/A and A/D converters and (2) reduce the cooling requirements of the
drive amplifiers.

Experiments were performed using error signals in various pulsed formats including "bang-bang"
pulse width modulation and time dwell modulation.

The use of pulsed drive waveforms offers several potential advantages:

e D/A converters are not needed, since digital circuitry can generate the pulsed waveforms.
* A/D converters for "wrap-around" monitoring are not needed.
* Pulsed waveforms allow the valve driver amplifiers to act as switches rather than as linear
amplifiers. As a result, the power dissipation in the drive transistors can be reduced by a
factor of 10 or more, thereby improving the reliability and reducing the weight of the DFBW
system is also improved as a result of the lower power dissipation.
e Dynamic pulsed waveforms are compatible with passive fault isolation schemes whereby a
hardover computer output can be blocked without the need for a disconnect arrangement.

*The results were sufficiently encouraging to warrant investigations into a new digital valve driver
amplifier concept better suited to pulsed waveforms than are the linear analog drivers, and designed
to exploit the potential advantages offered by the use of pulsed waveforms in aircraft having several
actuation systems under direct digital control.

The digital drive concept, illustrated in Figure 95, was constructed and tested. The typical wave-
. forms are illustrated in Figure 96. The circuit incorporates two features which reduce the size and
weight over that of an analog drive. First, the transistors operate in a switching mode to reduce the
internal power dissipation. Second, the circuit is designed to operate from a single polarity power
supply.

In addition the circuit can be designed for dynamic operation, that is, a pulse rate is required to
obtain an output. The absence of pulses results in zero output. If some form of AC coupling is
provided between the computers and the power amplifier, a hardover failure of a computer output
circuit would result in a passive failure and not a hardover failure. This permits the computers to
perform self-monitoring and remove themselves from the line if a failure occurs. It does not matter
if the computer output is zero or a plus voltage, it is still removed from the drive. This provides
opportunities for many forms of redundancy in the actuator drive.

The breadboard digital driver was tested as a part of the closed loop system. A direct drive torque
motor was used to provide a realistic load for the driver. A simulated actuator was configured to
enable studying the dynamic response of the loop, which was closed by the computer. The com-
puter generated the "surface error signal" in the pulse modulation format. The following specific
results were achieved'

1. The circuit concept was verified and a data base was established for use in optimizing future
4operational circuit designs.

146
 NADC-82240-60

2. The closed loop test results showed that the desired closed loop frequency response can be
obtained with proper compensation for the inductive characteristics of the valve coil.
3. The qualitative effects of pulsed waveforms on the torque motor were evaluated and found to
have little effect on the motor when operated at frequencies above 500 Hz.

The closed loop frequency response without any form of compensation is shown in Figure 97.
With computer compensation this bandwidth can be extended for small signals (1% of full travel)
by a factor of 4 to about 26 Hz. This is illustrated in Figure 98.

For certain dynamic situations, large changes in the surface displacement would be desired requiring
increased bandwidth for large signals.

Resistance in series with the motor coil is the simplest, most effective means of providing the needed
compensation. The resultant large stroke response is shown in Figure 98. This method also offers
the advantage of maintaining valve driver power bandwidth over the full range of error signal levels
- that is, the large signal response is identical with the small signal response. The closed loop band-
width will be limited only by the rate capacity of the actuator itself. The use of a series resistor
provides an additional advantage of current-limiting protection in the event of a hardover failure.

This work, performed under company I R&D funds, demonstrated a method to take maximum ad-
vantage of pulse modulated control. While pulse modulation performs satisfactorily with analog
(Class A) amplifiers, the use of drive circuits designed especially for pulse modulation produces a
significant reduction in size and power (heating) dissipation in the drive circuit. Because of the low
power dissipation of the electronics, the temperature rise is slight and the electronics could be pack-
aged directly into the surface actuator. A direct drive actuator design that utilizes electronics of
this type integrated into the actuator has been developed by Rockwell.

FLIGHT CONTROL COMPUTER COMPLEX - D3 ACTUATOR INTERFACE (obtained from

Reference 18)

While the simplification of the surface actuator is significant, the major advantage of the direct
drive concept is the reduced complexity of the computer-actuator interface. This is achieved be-
cause the direct drive concept provides methods of failure monitoring and redundancy management
that are more compatible with digital technology. The direct drive torque motor is compatible
with digital technology. The direct drive torque motor is compatible with digital (pulse modulation)
control circuits. Pulse modulated signals eliminate the need for analog signal transmission and can
be generated with digital timing circuits. Power amplification can be accomplished using highly
efficient switching amplifiers. The direct digital drive provides several options in failure monitoring
and redundancy management and can be used for triplex or quadruplex systems. It can also be used
in conventional redundancy management systems and is suitable for a fault tolerant design.

1. Pusle Modulation Command

The use of an electromechanical force motor to power the main control valve requires electrical
power amplification to replace the hydraulic power amplification. While servo power amplifiers,
* capable of handling the power requirement (25 to 50 watts) are readily available, the characteristics
of the force motor are advantageous for other methods of power amplification. The force motors
are limited travel devices with the windings having fixed resistance and inductance typical of electro-
magnetic coils rather than the characteristics of a motor. Utilizing the R/L time constant of the
coil winding as a filter makes the force motor well suited to the use of pulse modulated techniques.

147
 NADC-82240-60

B+

it 0ll 02

IN-4
IN 1 Al A2 -

+4 VALVE + VI
COIL

Q4 03

Figure 95. Digital Drive Unit Concept

SMALL ERROR LARGE ERROR

VIN 0 -;AF

(b)
" coil 0- 'ocv--
0 - , "np'
Voltag"
" I I I I
, I
Iurn I
-V] -V

0.8 Amp- 0.8 Amp-. I

(c) -DC-pple

Current 0-0

-4.8 Am- -0.8 Amp -

Figure 96. Digital Drive Concept Waveforms

148
 NADC-82240-60

Relative 5 - - --- -

output Amplitude
1

-151 - - - -1-

Phsase
Lag -100
(Degrees)

0 1 Frequency-Hz 2

Figure 97. Closed Loop Response (Uncompensated)

5 _____1 10 20 30 40

0 1~
-5 ~ ~Frequency-Hz Cmue
____________________________________________ ______Compustn

Amplitude-dB opnsto

-10 __ (%
0____ Full Travel)

t With Series Resistor,

I
-~

0 I(Full
____________________Compensation______

_______________ Travel)

Phase ShiftWihCmue
~100o S
Series_____

I 20o ___________I(Full
Travel IN \

4 Figure 98. Effect of Compensation Method on Closed Loop Response

 - --

NADC-82240-60

.Pulse
* modulation signals can be generated with simple digital timing circuits and the actual power
° . amplification can be accomplished in switching circuits, eliminating the need for analog servo ampli-
fiers.

Table 9 lists the characteristics of the torque motors used in the AFCAS program. Several forms
of pulse modulation have been evaluated including pulse width modulation (PWM), time dwell mod-
ulation, and an on-off or "bang-bang" control. While all three types of modulation are usable,
. pulse width modulation proved superior for small signals, offering the best response and linearity.
This occurs because PWM uses a constant carrier frequency with the signals producing only slight
changes in the plus and minus portions of the wave. Other forms of pulse modulation require a
much wider bandwidth to accommodate the frequency variations.

The modulation rate must be selected to meet the requirements of the electromechanical force
motor. The rate of modulation must be greater than the natural mechanical frequency of the force
motor and also well in excess of the R/L time constant of the motor coils. When these requirements
are met, the characteristic wave shapes shown in Figure 99 will be present at the motor. While the
signal voltage to the switching amplifier will be pulsed, the current through the force motor windings
will be well filtered by the inductance of the coils.

TABLE 9 - ELECTROMECHANICAL FORCE MOTOR DATA

COILS PER MOTOR - 4

VOLTS PER COIL - 10V
CURRENT PER COIL - 0.8A (RATED TORQUE)
COIL RESISTANCE - 10 OHMS
COIL INDUCTANCE - 0.6H.
POWER PER COIL - 8 WATTS (RATED TORQUE)
NATURAL FREQUENCY - 230 Hz
SHEAR OUT FORCE - 80 LBS (RATED)

The principle of the switching circuit is quite simple as illustrated in the schematic of Figure 95.
There is one switching amplifier for each channel of redundancy. This form of switching amplifier
offers several advantages. By using four switches operating in odd or even pairs (Q1 & Q3 or Q2 &
Q4) a single switch failure will not cause a hardover coil command. However, a single switch failure
will cause excessive current as half the cycle will produce a shorted supply, thus making current
monitoring a positive fault indicator.

Signal amplifiers Al and A2 are used to buffer the input signals. A failure in either of these stages
cannot cause a hardover command, but will result in a bias current in the switching amplifier that
can be readily detected and the power source subsequently turned off. If the pulse modulated in-
put signal is AC coupled, pulses are required for the signal to be transmitted to the switching ampli-
fier. A lack of pulses results in a fail passive circuit, even if the command voltage were to remain in
a full on condition.

The pulse modulated command signals can be generated by the computer if cycle time is not a con-
cern. The pulse modulated signal can also be generated by simple timing circuits consisting of
several logic gates and a counter. The exact design will vary with the application. The Motorola
4•timer MC6840 is one method of generating a PWM signal that is very flexible for test work. This

.* * .* * . *.*.'.. 150 . . . . .
 NADC-82240-60

chip is under computer control and the modulation characteristics can be readily varied. In flight
*tests performed at Rockwell using the direct drive actuator in the rudder system of the T-2C aircraft,
the pulse modulated signal was generated by the computer. For these test programs, closing the loop
through the computer had advantages for monitoring system operation, and computer cycle time
was not a concern.

The switching amplifier that produces the main power amplification can be located at the computer,
'. the actuator, or anywhere in between. The most desirable location will usually be at or in the actu-
ator. This eliminates the need for transmission of analog power signals and permits power monitor-
ing to verify the current used at the electromechanical force motor rather than on the transmission
side of the drive. The approach requires two additional wires per stage to transmit power to the
actuator.

2. D3 Redundancy Concept

The redundancy concept shown in Figure 100 for a quadruplex direct drive actuation is very similar
to existing redundancy concepts. The dual tandem actuator employs two electromechanical force
motors driving the main control valve. Each force motor contains four windings. Each of four
switching amplifiers drive two windings in series, one from each force motor. The switching ampli-
fiers are driven by a PWM signal from four separate computers, eliminating the need to transmit
analog information. As the switching amplifier is driven into saturation, neither the quality or exact
DC level of the power source are critical and will not effect loop gain. The power (current) supplied
to each switching amplifier is monitored at the source. When improper power measurements are ob-
served, the power source can be switched off. It is assumed that the aircraft electrical power
concept for the drive circuits will be the same as for the computers.

The switching amplifiers operate very efficiently with the switches either at zero voltage drop or
zero current. The power dissipation in the amplifier is no greater than the present class A servo
amplifiers used to power the electrohydraulic valves. The switching amplifiers are not gain sensitive,
and temperature effects on components will not affect system operation. This makes it practical to
locate the switching circuits in the higher temperature environment of the actuator. Locating the
switching amplifier at the actuator permits transmitting non-critical pulse modulated signals and
non-critical power to the actuator. The command signal is now a digital signal instead of a low level
analog signal.

Since the pulse modulated command signals are generated by timing circuits, it is possible to moni-
tor the command signals to a greater precision than possible with an analog drive. This enhances
the monitoring of the computer outputs and allows the computers to increase the accuracy of self
test. Computer redundancy requirements become less critical with improved output monitoring
which should make the triplex system very attractive to D3 actuation.

The redundancy concept shown in Figure 100 isfunctionally quite similar to present systems. The
only significant redundancy difference is the location of the force monitor and channel turn-off
switches at the computer. In present systems multiple hydraulic force monitors and solenoids must
4 be located in the actuator environment. This greatly increases the number of wires required between
the actuator and computer. Present systems need 80 to 100 wires compared to approximately 30
for the D3 system.

3. Fault Tolerant Redundancy

In the direct drive system presently under test, loop gains in excess of 90 (15 Hz) are possible be-
cause of the excellent dynamics of the direct drive torque motor. With this high loop gain, a hard-
over failure of one of the four channels will cause a bias shift at the actuator of less than 1%of its

,_ 151
 NADC-82240-60

COMMAND_____
{+5V __ ___

SIGNAL

-5V _ __

SIGNAL- - - _ _ -

VOLTAGE

-V

COIL ............ _ _ _
CURRENT

-A.

Figure 99. Pulse Modulated Waveforms

COMPUTER

CC~APU.ER

*~~~~* ____________ -PI

A -Ri

TOC,,E §It I J r w
MiOTORS ~-R2 -

- >-

ACTUATOR

*6 OYPuER

Figure 100. D3 Redundancy Concept

4 152
 NADC-82240-60

total travel. In some applications, surface responses of 15 Hz are not desirable. The use of lower
loop gains will cause a corresponding increase in the bias error due To a single channel failure.

One approach to reduce the bias error of a failure is to use four drivers per actuator stage (eight per
dual actuator). This will reduce the bias error of a single "hardover" failure by more than a factor
of 2 and will allow the unit to continue operation with up to 3 identical "hardover" failures in the
drive electronics. With this approach, current monitoring and switching would not be required. The
simplicity of the switching amplifier makes this a reasonable approach. The use of an interloop
feedback (monitoring force motor position) is also a valid approach to reducing effects of hardover
failures, but adds considerable complexity to the overall interface.

As shown earlier, single failures of the switching circuit, while not producing "hardover" failures,
can result in drawing excessive current from the power source. Therefore, regardless of which re-
dundancy method is selected, "short circuit" protection of the power source must be provided.
This protection also removes power from a failed channel. The actual redundancy method selected
will depend on the system application.

This system approach simplifies the actuation, the command electronics, and the interface by in-
creasing the use of digital electronics. The only area that remains analog is the feedback snesors. A
program is presently underway that will remove the analog sensors and reduce the electronic hazards
such as EMI, EMP, lightning, high energy radiation, and spurious signals.

153
'q NADC-8224-60

THE HYDRA-OPTIC FLIGHT CONTROL ACTUATION SYSTEM (HOFCAS)

DEVELOPMENT PROGRAM

The AFCAS concept developed by the Naval Air Development Center (NADC) is a direct-drive,
lightweight, hydraulic surface actuation system capable'of being controlled with a direct digital
command. This Direct Digital Drive (D3 ) actuation system and the lightweight hydraulics, have
been successfully flight tested and shown to be a practical approach to surface actuation. This
project utilizes AFCAS with new approaches incorporated to eliminate the conventional intercon-
necting cabling to the surface actuator.

A major concern for future fly-by-wire aircraft is protection against electronic hazards such as EM I,
EMP, lightning, high energy radiation, and spurious signals. If control signals can be transmitted via
light pulses, particularly to the actuation regions, then electromagnetic transient protection can be
greatly enhanced. By removing all electrical connections to the actuator, electromagnetic coupling
which would allow RFI, lightning or other transient signals to enter would be eliminated. The
electrical power required to receive and utilize the light pulses would be generated hydraulically
within the actuator. The electronics to restore the digital signal and drive the electrohydraulic valve
would also be contained within or on the actuator. The system developed for this project offers an
approach for meeting the goal of a direct drive surface actuator with no interconnecting cabling.

The AFCAS program that preceded this study was a six phase program that explored the feasibility
of direct drive, developed the concept, and evaluated it in a series of laboratory and flight test pro-
grams. Phase VI of this program incorporated the microcomputer and successfully demonstrated
the D3 capability of the AFCAS. Pulse modulated signals, generated in the microcomputer, were
power amplified in an Electronic Drive Unit (EDU) and used to control the direct drive actuator.
An analog backup capability was also provided, whereby the EDU controlled the actuator directly
and the microcomputer was not in the control loop.

The general objective of this program was to assemble and test a direct-drive, lightweight, hydraulic
surface actuator system that provides immunity to all forms of electromagnetic radiation. The
specific goal was to operate the digital AFCAS without the need for electrical power being trans-
mitted to the actuator. This was accomplished by modifying the AFCAS to operate with a
hydraulically-driven alternator supplying the electrical power, and adding an optical command link
between the microcomputer and the direct drive actuator electronics.

TECHNICAL APPROACH

The technical approach used to develop the Hydra-Optic Flinht Control Actuation System
(HOFCAS) was to utilize assets developed under the previou4 six-phase AFCAS program and to
demonstrate, through laboratory testing, that the HOFCAS concept is suitable for flight in the T-2C
Demonstrator Aircraft. The AFCAS actuator, the transducers, and the microcomputer power sup-
ply were used without change. The microcomputer and the EDU were modified to incorporate
optical data transmission. A Hydra-Powered Alternator (HPA) was added to supply the electrical
power needed for the actuator drive unit. Additional changes were made to the system switching
of operation.
to incorporate the new modes
and interconnecting wiring

154
 NADC-82240-60

The system configuration was designed and tested to verify that the HOFCAS is suitable for con-
trolling the rudder of a T-2C aircraft, shown in Figure 84. For this application the T-2C control
system will be modified from a mechanically powered cable system to a full powered Digital Fly-
By-Light (DFBL) system with an Analog Back-Up (ABU) mode of operation. The original cable
system between the rudder pedals and the rudder will be removed. The rudder pedal will be
* attached to force transducers. Force on the rudder pedal is converted to a proportional electrical
signal by these force transducers. This signal is supplied to a microcomputer where it is summed
* with the feedback signal and processed into a pulse modulated error signal. The pulse modulated
error signal is transmitted optically through fiber optic cables to the EDU where it is converted to
electrical signals, power amplified, and used to control the torque motor of a direct drive hydrau-
lic rudder actuator. This modified system provided full power DFBL control of the rudder,
instead Qf the manually operated rudder of the basic T-2C aircraft.

The HOFCAS mounting requirements, lead length, and hydraulic and electrical power sources
were selected to be compatible with the T-2C. The HOFCAS evaluation criteria were selected
to meet the dynamic performance requirement of the T-2C aircraft. Safety provisions, including
failure mode evaluation, were based on T-2C flight safety requirements. The previous AFCAS
flight test program included a hydraulic bypass valve on the direct drive actuator. This device
allows the rudder to seek a trail position if the rudder control system were lost. Flight testing
has established the aircraft can be safely landed with the rudder in the trail position. This fea-
ture establishes a third level of redundancy beyond the DFBL and the ABU modes of control.

All the major components needed to fly the HOFCAS in the T-2C were assembled in the labora-
tory for the system testing. System and component tests were performed to insure satisfactory
operation in the T-2C. Additional system tests were performed that exceeded T-2C require-
ments. These tests were performed to establish the performance levels possible with the present
HOFCAS and to identify areas where improvements could be made if necessary.

This approach makes use of existing Navy hardware to test and evaluate several critical portions
of an actuation system that eliminates the need for wiring to transmit power and signal to the
actuation region. The Navy currently has a separate program for developing an optical position
sensor to provide a feedback for this direct drive hardware. After the optical sensor, the HPA,
and the EDU are integrated into a common actuation package with the D3 actuator, the general
objectives of this program will be achieved.

DESCRIPTION OF THE HOFCAS

The HOFCAS utilizes a direct drive actuation system, commanded by light, with self contained
electrical power to control an aircraft rudder actuation system. The system hardware for this pro-
gram is a modification of the Phase VI Direct Digital Drive AFCAS program hardware previously
evaluated and described in Reference 17.

A simplified block diagram showin- the major components of the HOFCAS system is contained in
Figure 101. The shaded areas in Figure 101 indicate the units modified or added for the HOFCAS.
The microcomputer and the EDU were modified to incorporate the optical command link (new)
and the HOFCAS Hydra-Powered Alternator (HPA) is a new unit added to provide electrical power
for the EDU in the Digital Fly-By-Light (DFBL) mode. Two modes of system operation are pro-
vided, the DFBL and the ABU modes. Figure 102 contains a block diagram that illustrates the two
modes of operation.

155
I NADC-82240-60

__a_ >* < cn Q

> M co > L)

C.,.

z
LLI- w-

o X0c
- 0

0 0a

I- LL
00

O00

X w

0 (
156w
NADC-82240-60

0N

- cc-

157,
 NADC-82240-60

The DFBL mode is selected by momentarily holding the cockpit DFBL ENGAGE switch to ON,
energizing the DFB L control relays and resulting in the following:

* The pedal command and actuator feedback transducer outputs are connected to the
microcomputer input.

a The microcomputer output is connected to the EDU via the fiber optic control link.

* Hydraulic pressure is supplied to the HOFCAS HPA, and the 26 VAC, 1000 Hz HPA
output is used to power the EDU via a 26 VAC to 115 VAC step-up transformer. The
EDU input power is utilized to provide excitation for the transducers and to condition,
amplify, and convert the optical signals into electrical signals which are power amplified
and used to drive the torque motor coils.

If the microcomputer senses the system is functioning properly, a power ground is continuously
supplied to the DFBL ENGAGE switch holding coil by the microcomputer and the system remains
in the DFBL mode.

The ABU mode can be selected by manually placing the DFBL ENGAGE switch to OFF, or auto-
matically by the microcomputer removing the power ground from the DFBL ENGAGE switch hold-
ing coil if abnormal system operating errors occur.

Switching to the ABU mode results in the following:

0 The fiber optic control link output from the microcomputer is disconnected from the
EDU, and the pedal command and actuator feedback transducer outputs are connected
directly to the EDU where they are summed and amplified to drive the four torque motor
coils of the actuator direct drive valve.
* Hydraulic pressure is removed from the HOFCAS HPA, and the EDU is powered by

115 VAC, 400 Hz aircraft power.

The various system components are discussed in the following paragraphs.

1. Hydra-Powered Alternator (HPA)

The HOFCAS HPA assembly, Bertea P/N 287700, supplies electrical power to the EDU in
the DFBL mode. The alternator rated output, at 3000 psi (21 MPa) A P and a fluid flow
of 0.72 gpm (2.7 L/min), is 25 watts @26 VAC, 1000 Hz. The HPA was not used in the
previous Phase VI of the Digital AFCAS program and has been added for HOCFAS.

The unit consists of a hydraulically driven 2" (5.1 cm) diameter turbine wheel coupled
to a conventional permanent magnet electric alternator, spun by fluid discharged from a
.020" (.05 cm) diameter orifice. The 6-pole configuration of the alternator, Electro-
* Kinetics Model No. 4922-2, provides a frequency of 1000 Hz at 10,000 rpm.

The alternator output is nearly sinusoidal with a resistance load, and distorts to a pseudo
square wave when inductive loads are applied, (detailed data describing alternator output
and waveforms are contained in section 3.3.1 of Reference 19).

* 158
.a
.I ....
 NADC-82240-60

The HPA turbine wheel and alternator are shown in Figure 103 and the HPA assembly is
shown in Figure 104. Additional HPA data is supplied in Appendix B of Reference 19.

2. Electronic Drive Unit (EDU)

The EDU contains the electronics required for converting optical inputs to electrical in-
put signals, signal conditioning and amplification, signal summation, and power amplifi-
cation to current drive the actuator torque motor. The unit was designed by the North
American Aircraft Division (NAAD) and contains two independent channels, each sub-
divided into dual valve driver circuits.

Each of the dual valve driver circuits employ current feedback with a highly reliable
"Darlington" power transistor configuration. Redundant power supplies are used. The
circuitry is designed so that in the event an output stage fails "hardover", the voltage ap-
plied to a motor coil will not exceed its rated value. This limiting feature permits a sub-
unit failure to be compensated or nullified by another subunit. Closed loop tests showed
that operation of the redundant subunits provided high immunity to component failures.
The dual channel fiber optic receiver amplifier added for HOFCAS converts the micro-
computer optical output signals to electrical input signals for the EDU. Each receiver
amplifier channel contains a detector, two stages of gain, and a passive isolation circuit.
The optical signals received consist of PWM information, and the passive isolation feature
prevents a hardover of the rudder actuator if the drive signal is lost due to a microcom-
puter or fiber optic loop failure.

The dual channel fiber optic receiver amplifier and associated detectors are located on a
single module board, contained in a housing mounted on the EDU. All electrical and
optical connections to the module board were made to permit easy removal if return of
the EDU to its original configuration is desired.

The modified EDU with fiber optic cables is depicted in Figure 105 and is functional
schematic, including the fiber optic receiver amplifier, is included in Figure 2-3 of
Reference 19.

3. Microcomputer Assembly and Power Supply

The microcomputer assembly is housed in an enclosed unit, and consists of the following
subassemblies:

0 Motorola Monoboard Microcomputer Module

* Analog-To-Digital (A/D) Converter Module
* Digital-To-Analog (D/A) Converter Module
0 Card Cage and Mother Board Assembly
0* Signal Conditioning Board
0 Fiber Optics Transmitter Board Assembly

The microcomputer module is a complete computer-on-a-board having all the process-

ing and control required for a microcomputer-based system. It incorporates the MC
6800 MPU, 1 K of Random Access Memory (RAM), provisions for 4 K of Programmable
Read Only Memory (PROM), timing and control, buffers, an Asychronous Interface
Adapter (ACIA) and two Peripheral Interface Adapters (PIA).

- 159
 NADC-82240-60

Figure 103. HPA Turbine Wheel and Alternator

tMo

Figure 104. HPA Assembly

Figure 105. EDU With Fiber Optic Cables Attached

l 160
 NADC-82240-60

The A/D converter module consists of eight channels of A/D conversion of which four
are utilized. The D/A converter module consists of four channels of D/A conversion of
which three are utilized.

The signal conditioning board contains four channels of sensor signal conditioning and
a relay driver that interfaces the microcomputer monitor output with the system con-
trol logic.

The dual channel fiber optic transmitter board was added for HOFCAS. Located in
the microcomputer housing, the board contains a fiber optic photo diode and dropping
resistor for each channel, to convert the microcomputer PWM output to optical form for
transmission to the EDU. Two fiber optic cables, each 10 feet (3.05 m) long, intercon-
nect the microcomputer and EDU. The 10 foot (3.05 m) cables simulate the length
required to connect an actuator, with its electronic EDU Located at the rudder, to the
microcomputer located in the equipment bay. The microcomputer unit with the fiber
optic cables attached is shown in Figure 106.

A separate power supply (Motorola P/N M68MMPS1) is provided to supply power to

the microcomputer. The power supply converts single-phase, 115 VAC, 400 Hz to + 5
VDC and ± 12 VDC.

i WAdditional information on the microcomputer is contained in Appendix A of Reference

19.

The software of the microcomputer is the same as that used in the flight testing of the
D 3 AFCAS concept (Reference (7).

Figure 106. Microcomputer Unit With Fiber Optic Cables Attached

4
LABORATORY AND INTEGRATION TESTS

Laboratory tests were accomplished to integrate the HOFCAS with the existing Phase VI digital
AFCAS system, perform system functional tests, evaluate system performance and compatibility
with the HOFCAS modification, and to determine the effects of the HOFCAS modification on
system performance. Temperature tests were also performed to verify the equipment would func-
tion properly in the T-2C aircraft operating environment.

Laboratory tests were accomplished using the actual aircraft components to simulate flight con-
ditions in preparation for flight testing of the system at a later date.

4 161
 NADC-82240-60

Included in the lab setup were the rudder LVDT feedback transducers, rudder actuator, EDU,
microcomputer and associated power supply, HPA, fiber optic link including emitters, detectors
and fiber optic cables, and the switches and control relays used in the aircraft.

The lab wire harness was configured to simulate the aircraft wiring, and a terminal strip/intercon-
nection board provided control, test points, and the interface between the wire harness, system
components, and lab test equipment. The Lab Signal Simulator Box was included as part of the lab
test setup, and provided simulated pedal force commands, LVDT feedback signals, and microcom-
puter PWM output signals during certain phases of system testing. The two fiber optic cables, con-
necting the microcomputer output to the EDU, were each 10 feet (3.05 m) in length simulating the
actual aircraft cable length requirements. A portable lab hydraulic pressure source was used for all
lab testing.

The bandpass requirements and therefore the system loop gain were established by T-2C aircraft
dynamic requirements for the integration tests. System dynamic response tests were performed to
demonstrate that the HOFCAS operational modes meet these dynamic requirements and are satis-
factory for controlling the rudder of the T-2C aircraft. The lab test setup was configured to pro-
vide either HPA 1000 HZ or simulated aircraft 400 HZ power for the EDU in the HOFCAS DFBL
and ABU modes, and a DFBW mode of operation. A function generator provided sinusoidal and
step input signals to the system force transducer inputs and a 2-channel strip chart recorder was
used to record the function generator output and rudder actuator feedback signal.

ANALYSIS OF TEST RESULTS

. The test results demonstrate that the HOFCAS is satisfactory for operation in the rudder system
of the T-2C aircraft. The testing was divided into three categories. The first group of tests accom-
plished the functional testing of the added system components, especially the HPA. The second
group of tests evaluated the HOFCAS for operation with the T-2C rudder. The third set of tests
established performance limits and provided data for suggested improvements in the HOFCAS
.- concept.

1. Hydra-Powered Alternator (HPA)

The functional tests established the HPA operational capability and demonstrated the
unit is adequate to supply the electrical power needed to operate the rudder actuator of
a T-2C aircraft. The data obtained has the characteristics expected for the HPA design.
The design is described in Appendix B of Reference 19.

rThe speed (frequency) and the no load voltage of the alternator are directly related to
hydraulic flow, which is controlled by the turbine nozzle diameter and the applied
hydraulic pressure. Increased flow through the fixed nozzle results in increased losses
and additional heating of the hydraulic fluid.

The alternator used for this project was an adaptation of an existing missile power
source and is wound to provide alternator protection against overloads. This charac-
teristic becomes apparent in the tests data. As the output current is increased, the
voltage decreases and total power output remains the same. At the same time, the
unit shows very little change in speed (frequency) with increased current. As the
power output remains constant at rated load, it is apparent the hydraulic turbine load
is constant and the speed would not vary. In addition, the HPA has a very low effi-
ciency so that slight variations in extracted power have very little effect on total
torque required to drive the alternator.

162
 NADC-82240-60

Wave shapes of the alternator output showed the waveform with a resistive load to be
nearly sinusoidal. For this system application the output power is supplied to power
supplies in the EDU through a transformer and the wave shape is not critical to system
operation. When the HPA was used to power the EDU, the load became highly induc-
tive and the wave shape approached a square wave. While this does not directly affect
system operation, the distorted wave shape reduces transformer and power supply effi-
ciencies and further increases the loading effects on the alternator.

The HPA used for this project is satisfactory for flight testing, and will demonstrate the
concept of a D3 actuation system without electrical power and signals routed to the
actuation surface. Several areas of improvement in the HPA performance are desirable
and are discussed in Appendix B of Reference 19.
2. Fiber Optic Link
The fiber optic link performed satisfactorily as expected. No difference in system per-
formance was noted between the fiber optic link or direct wiring method of connecting
the microcomputer output to the EDU.

3. Flight Test Configuration

HOFCAS integration tests were performed using the hardware configured to the T-2C
aircraft installation. These tests consisted primarily of dynamic response and failure
mode evaluation. Frequency response tests showed a slight reduction in system perform-
ance with the EDU powered by the HPA 1000 Hz output compared to 400 Hz power.

The slight reduction in dynamic response, when the EDU is powered by the .HPA 1000
HZ output, is due to the limited power capability of the HPA which results in a voltage
drop at the torque motor coils when additional power is requested. The voltage drop
causes a reduction in loop gain and a corresponding loss of bandpass.

The HPA used in this project is capable of a steady state 26 watt output with no capa-
bility of handling higher power demand transients. The EDU peak power requirement
is 50 watts, consisting of 32 watts for the torque motor and 18 watts for EDU amplifiers,
power supplies, and transducer excitation. The torque motor characteristics (see Figure
107) are such that at 26 watts total power (8 watts at the torque motor) the major por-
tion of the spool travel has occurred and the rated flow of the valve is already obtained.

The additional power requirement to the torque motor is needed to obtain the full
particle shear out capability and is not required for full hydraulic flow. For the T-2C
application, if the full torque motor force is required, the system will revert to the ABU
mode with aircraft power and full shear force will be available.

The step response characteristics determined during the tests again show the effects of
limited power to the EDU. This effect is more pronounced with step demands as they
require high power peaks.

4. HOFCAS Performance

Results of the dynamic tests performed in this project successfully demonstrated the
suitability for HOFCAS operation in a T-2C aircraft and provided additional data to
demonstrate the potential of a HOFCAS design.

The performance of the AFCAS actuator was evaluated in a previous flight test pro-
gram. The measured frequency response shown in Figure 108 is approximately 13
Hz. This response was considerably above that required for the flight test program.
163
 NADC-82240-60

.n +.040 -

+.030-

0
0

cn +.010

I I I l
32 24 16 8 8 16 24 32
-.010 MOTOR POWER, WATTS

-.202--

-.030--

C-.040

Figure 107. Torque Motor Characteristics

The D3 operation of this actuator used the same analog EDU to drive tne torque motor
of the actuator. Since the EDU was not optimized for digital drive some bandpass was
lost, and the response is that shown on Figure 109 for aircraft power. Rockwell lab-
oratory tests of a circuit optimized for direct digital drive have produced the frequency
response (25 Hz) shown in Figure 110. As the natural frequency of the torque motor
is approximately 230 Hz, the dynamic response is limited either by the valve design or
the electronic drive design.
3
The data of Figure 109 shows the dynamic response obtained with the present D
design and that obtained with the HOFCAS design. As mentioned above the bandpass
with aircraft power matches that of Phase VI of the AFCAS program. The reduced
response of the HOFCAS design is produced by the limited power available from the
HPA. The limited HPA power reduces the power available to the torque motor coils
resulting in a reduced loop gain under dynamic conditions. This produces reduced
response but no decrease in static accuracy.

For comparison purposes the response curve of the ABU mode is shown in Figure
111. This response is the same as that of the two previous flight test programs and
demonstrates that the modifications for HOFCAS did not affect the ABU operation.

RECOMMENDATIONS

The results of the laboratory tests demonstrate that the HOFCAS configuration is satisfactory for
flight testing in a T-2C aircraft. The tests prove the HOFCAS concept is a practical approach for
operating remotely located surface actuation with immunity to electromagnetic interference. The
- D3 actuator is ideally suited for this application because of the integrated actuator concept and
the ease of adapting optical control the actuator.

164
 .,o --. -- r1 . . . V .v-
- r r.-.r.; r r V. . .r ."" " • - " " - - - . . ". -

NADC-82240-60

The next logical step in developing the HOFCAS would be to demonstrate the concept in flight.
Several additional developments, some of which are already in work, are needed to establish this
concept for use in future high performance aircraft. These items are:

, A suitable optical actuator feedback link.

- High temperature digital electronics suitable for packaging inside the actuator.

* * An improved rotary type valve-torque motor suitable for packaging inside the
actuator.

0 An improved HPA with the capability of handling the peak transient power
requirements.

* Design of the integrated actuator package.

The optical feedback link and the valve-torque motor are already under development in separate
Navy programs. The drive electronics would be the next needed development. The new electronics
and tho improved valve-torque motor can be expected to reduce the power requirements of the
HPA.

Other studies of interest would include research into methods of improving the HPA efficiency
without increased size, weight or complexity. A design study to show the advantages of the
HPA over the "brute force" approach (shielding, filtering, etc.) to protect against EMP would
also be desirable.

The conclusions of this study show that HOFCAS approach incorporating a D3 actuation sys-
tem will produce a simple highly reliable actuation system that is immune to the effects of
electromagnetic interference.

Figure 112 shows an actuator concept in which the digital electronics, rotary value torque
motor, and feedback transducers are integrated inside the actuator body.

NII ERFACE OF FLIGHT CONTROL COMPUTER COMPLEX, D3 ACTUATION,

/!ND HOFCAS
The HOFCAS could be configured for either three or four channel redundancy. With the improve-
ments in digital technology and self monitoring techniques, it is anticipated that future systems
will be triplex. A triplex redundancy concept for HOFCAS is illustrated in Figure 113.

The force motors will be designed with three windings per motor and will use a switching ampli-
fier for each coil. This redundancy level will remove the need to disconnect a failed unit. Short
circuit protection will be included at all six switching amplifiers. The amplifier is commanded
with a PWM signal. The optical sensors are designed to supply a digital feedback signal to the
* computer. This system produces an actuation loop that is completely digitally controlled. The
removal of interconnecting wires between the computer and the actuator will greatly reduce the
susceptibility to electromagnetic interference. This system is still in the early stages of develop-
ment and the redundancy concepts may be modified as additional test data is obtained on system
components.

* 165
 NADC-82240-60

-50

10 -100

0 -150

-200 0
-.10-

-20-

-30

.11.0 FREQUENCY(Hz) 10 100

Figure 108. AFCAS Frequency Response

+10 AIRCRAFT POWER (400 HZ)

00

* HPA POWER (100 HZ)

r -20

11.0 FREQUENCY (Hz) 10 100

Figure 109. Digital Fly-By-Flight Mode System Response, Maximum Capability, HPA
Power (1000 Hz) and Aircraft Power (400 Hz)

II 166
 NADC-82240-60

-50

10 -100

m 0 -150
GAINw
Z0
a -10 -200

-20

-30

.1 1.0 FREQUENCY (Hz) 10 100

3
Figure 110. Frequency Response, AFCAS Designed for D

-50

GAIN
.10 -100

-20

-301
.1 1.0 FREQUENCY (Hz) 10 100

Figure I111. Analog Back-Up Mode System Response, Flight Gain

Configuration, Aircraft Power (400 Hz)

167
q NADC-82240-60

Figure 112. HOFCAS System Redundancy Concept

OPICF/
OPTIC F/B

OPI
Fiue13 HFA yte eudnc ocp
4 168
 ,,. .,.V r r. - --

NADC-82240-60

L-. FAULT-TOLERANT ACTUATION CONCEPT FOR A RESEARCH TEST AI RCRAFT

BACKGROUND INFORMATION

Current hovering aircraft capabilities demand considerable visual contact flight before final landing,
with the pilot providing most of the attitude-stabilizing, position-fixing, height-controlling and deck
motion compensating functions. Pilot workload, even in clear weather operations, is excessive.
This places additional demands of considerable magnitude upon the pilot and the flight control
capabilities of the hovering vehicle. The development of an advanced, precise, and highly reliable
flight control/guidance system concept to meet operational goals is a prime requirement. The man-
ual and automatic modes of the flight control system must have sufficient authority to perform
their required functions during the critical vertical take-off and landing operations. This implies
that the conflict between automatic control authority and flight safety must be resolved by incorpo-
rating fail-safe and fault-tolerant features in the Flight Control System (FCS).

The study vehicle selected for the integration and evaluation of the 4-valve fault-tolerant actuation
concept, developed by Bell Helicopter Textron Corporation, is a helicopter with a Type II power
boosted FCS. The Type II FCS is a reversible control system wherein the pilot effort, which is
exerted through a set of mechanical linkages, is at some point in these linkages boosted by a power
source, which in this application is hydraulic. Flight control commands from the pilot and the
Automatic Flight Control System (AFCS) are algebraically summed and then transmitted to the
*flight control surfaces.

Automatic landing capability is h.,, iy desired, which in turn dictates a large control power require-
ment. Therefore, the AFCS is anticipated to be a high (50 percent) authority system. During auto-
* matic landings on small ships, all ship kinematics, aircraft range, and range rate information will be
transmitted to the helicopter via the Landing Guidance Sensor (LGS) data link. All computations
of the flight control laws required for execution of the landing task will be performed by the Flight
Control Computers (FCC) which will then transmit the appropriate commands to the flight control
surfaces via the 4-valve fault tolerant AFCS actuator.

.. The actuation system for the AFCS is recognized as a critical technology in the development of a
research flight test vehicle. It is for a dual fail-operate requirement that the 4-valve actuation system
*1 is designed.

FLIGHT CONTROL ACTUATOR REQUIREMENTS

In order to resolve the conflict between automatic control authority and flight safety the laboratory
versions of the 4-valve AFCS and the primary boost actuators were required to have the following
design features and capabilities:

(a) The hydraulic and electrical power supplies shall be configured to provide a Failure Tolerance
Level (FTL) of single-fail/operate for the primary actuators as well as for the AFCS series
hydraulic actuators.

169
 NADC-82240-60

(b) The AFCS series actuators shall incorporate the following features:

(1) Be dualized and include a feature that automatically centers and mechanically locks
the piston shaft when the AFCS is disengaged or in the event all hydraulic or all elec-
trical power is lost. The centering function shall center the series actuators at a rate
that shall not overburden or disorient the pilot.
" (2) Be interfaced with the primary boost actuators in such a manner that full manual re-
version capability is provided in the event all hydraulic power is lost.

(3) Operate in conjunction with parallel trim actuators in such a manner that they shall
normally be working about the center point of authority.

(4) Be capable of receiving triplex electrical command signals from the AFCS and adding
them algebraically to the pilot's mechanical input

• (5) Possess at least a single-fail/operate capability with inherent fault-tolerant features.

* (6) Provide 50% of the total (stop to stop) command authority for the triplex electrical
signals from the AFCS.
(7) Contain provisions to disengage the triplex electrical signals. Disengagement will not

effect a transient that could be considered hazardous.

(8) Be powered by two independent 1,000 psi hydraulic systems.

(9) Be installed so that triplex signals shall not be reflected back to the pilot's cockpit
input device unless the summed input from the pilot and series actuator is in excess of
the total allowable control travel.

(10) Simultaneous application. of large pilot AFCS electrical inputs shall not overload the
output mechanical devices downstream of the actuator (stops downstream of the
summed output). "Smart" constraints for structural protection should be considered.

(11) Be efficiently packaged for installation in confined areas.

INTERFACE OF THE PRIMARY AND AFCS ACTUATORS

*l The actuator configurations shown in Figures 114 and 115 functionally describe two concepts or
how the 4-valve AFCS actuator can be operationally interfaced with the primary boost actuators to
effect a series-type control input.
The configuration shown in Figure 114 uses two grounded-base actuators with the two outputs dif-
ferentially summed at a specific ratio to establish the desired AFCS authority. Stops are located
downstream of the summing to limit total travel of the controlled output and hence to prevent over-
travel of the basic controlled element. The AFCS actuator is dualized and uses 4-position feedback
sensors, a centering spring/lock mechanism, and 4 electrohydraulic servo valves (EHSVI that are
controlled by quad-redundant control paths. The quad-redundant electronics can be interfaced with
digital or other analog systems.

6- 170
 NAAL
FUTUREWARMINSTER 3/2
I
7TO-RUS0624 FLIGHT CONTROL
AIRCRAFT(U)
MNE NAYRL AIR DEVELOPMENTFORCENTER
TECHNOLOGIES PR
U h AIRCRAFT AND CREN SYSTEMS TECHNOLOGY DIRECTORATE
p NCLASSIFIED N U KANIUKA SEP 82 NADC-82248-60 F/G 1/3 N
 !4

2.

2-1 25
LLA

UsI' 1j30

MICROCOPY RESOLUTION TEST CHART

NATIONAL
BUREAUOF StANDARDS,1963.A

,
0S ) .
q- - --
0 .-
q NADC-82240-60

CLIU.
1zm

0J 0

00

U, z

0 0 0

0
L'U

000
LU Z
z U. u-
00

-4-

1r -1-
0U
0
cn

0 Cd3n

0 00 a
II > x

LA.

0 "
ca J LL J L - L-

1Z

4u

Lu

02

4 171
 ND-82240-60

x wu

U
L(U
za
j
U> x

II> L LL.

oo cm

-U E
0.o

U-
0~

z0- 0- -I<

172W w
 ..- -. - ..--
,-"- -- -. t•-.-- -. v--v..-- - .,

4 NADC-82240-60

Operationally the configuration shown in Figure 115 is the same as the Figure 114 configuration
with the exception that the output of the AFCS actuator is mixed with the pilot's input to effect
a differentially-summed output to the control valve for the primary actuator. This configuration
uses the primary actuator for the muscle which allows the summing to be accomplished at a lower
force level. Hence, a smaller actuator can be used. The centering spring/lock mechanism, however,
will be required to have the force capability of centering in the event of a total loss of hydraulic
power. Bell Helicopter Textron (BHT) has used an actuator configuration of this type in a 3-axis
SCAS for the JetRanger helicopter.
This configuration was selected for laboratory because it offers a weight and size advantage as well
as a comparatively low hydraulic flow requirement

THE BELL HELICOPTER TEXTRON FOUR-VALVE, DIRECT DRIVE, ACTUATION CONCEPT

This actuation concept was originally developed by the BHT Corporation for use in 4-channel heli-
copter FBW control systems. It does not require the use of secondary actuators and its 4-valve con-
figuration enables it to be hignly fault tolerant The 4-valve concept has high hinge moment and
stiffness capabilities and can be used to actuate large swash plates and conventional flight control
surfaces. It is readily adaptable for use in high performance fixed wing aircraft.

In this program for the development of a fault-tolerant actuation concept for a test aircraft the 4-
valve concept will be used only for adding AFCS signals to those of the pilot's. It was chosen for
this flight safety critical task because of its fault-tolerant features and high inherent reliability.
Therefore, its force output requirement for this application is very low, being required only to
actuate the boost valve of the primary boost actuator (see Figure 115).

1. Summarized Description of Concept

The basic fault-tolerant actuation system consists of dual hydraulic primary actuators, quadruplex
electrical control paths, a failure management system, and electrical drive signals. Two electrical
control paths are used for each piston. The failure management system is mechanically interfaced
with the electrical control paths to provide maximum security. It provides automatic disengagement
of a control path and also provides track error signals that are used in the control paths for auto-
matic alignment of the four valves.

A flight test model of this system would include a master control panel and an annunciator panel.
The control panel would provide the necessary control functions, preflight checkout capability, and
a manual reset for each control channel. The annunciator panel would indicate the operating con-
dition of the control paths and would operate in conjunction with the control panel for the pre-
flight checkout.

2. FBW Control Paths

A control axis of the basic 4-valve actuation concept consists of four FBW control paths and a dual
piston power cylinder. The four control paths connect the pilot's control input to the power act-
uator cylinder and include the four electrohydraulic servovalves as shown in Figure 116.

The dual actuator schematically depicted is conventional, except that the control head (spool valve
assembly) has been replaced with the four electrohydraulic servovalves.

173
 r NADC-82240-60

-4-

u 0 C4

I' a
L)cn II

H. 1-4 E1-

174z
 NADC-82240-60

Figure 116 is a functional schematic of the follow-up system; i.e., the dual actuator is slaved to the
control inputs. All control paths are identical and operate simultaneously. A control input to the
amplifiers proportionally opens the valves and drives the actuator until the dual feedback trans-
ducers provide feedback signals that cancel the command signals at the amplifer, which closes the
valves and hence stops the actuator at a new position. The four valves are continuously and auto-
matically aligned by a limited authority signal that is inherent in the failure management system
(this feature is discussed in more detail later). The dual feedback transducers can be single elements
and separately located to reduce vulnerability to battle damage if desired. The response of the
actuator can be shaped to improve handling qualities as required.

The failure logic for the system shown in Figure 116 operates in the following manner. If a control
path fails (e.g., path la), the path is automatically disengaged and Valve la is cut off to prevent
leakage of fluid from one side of the piston to the other. A second path failure will be disengaged
in the same manner. If the second failure should be path lb, the logic circuitry will automatically
engage a pressure-operated hydraulic bypass across the common piston so that the failure will not
restrict the operation of the other piston. It is pointed out that if a first failure should disable the
failure management system (described in the next section), the control path system, shown in Figure
*-: 116, has the inherent capability of absorbing a second failure. This is possible because, for example,
- if Valve la should fail and remain hardover, the other three valves will go in the opposite direction
to oppose the actuator motion. This will effect a bypass around the piston common to Valves 1a
and lb and, hence, will allow the other piston to operate without any appreciable degradation.
This inherent feature appreciably improves the overall reliability of the system and allows a com-
paratively simple failure management system to be used in place of a conventional voting scheme.
The overall actuation system can be characterized as a forgiving type system.
3. Failure Management System

The failure management system is an integral part of the total system. It is required, as a minimum,
to manage only one failure since the basic FBW system has an inherent failure tolerance level of
single-fail/operate. Each control path has its own respective failure management unit which operates
with a maximum of independence from the other failure management units. These units are me-
. •chanically interfaced with their respective control paths to provide operational security. In the
event of a failure, the faulty control path is automatically disengaged. If two control paths that
share a common piston fail, and hence are disengaged, the failure logic circuitry will effect a by-pass
across the piston so that it will not adversely affect the operation of the other piston. A condition
monitor is provided as a part of the failure management system to inform the pilot of the operational
status of the FBW system. The monitor indicates soft failures as well as failed and disengaged con-
trol paths. In addition, the monitor provides a track error signal to each control path to provide
continuous, automatic alignment of the servo valves which are the end elements of the paths.

The failure management system consists of a failure sensing function and an automatic detection
disengage function. These functions are conceptually described in the following paragraphs.

a. Failure Sensing Function

* The addition of an LVDT-type position transducer to the porting stage of the servovalves (second
stage on conventional servovalves) allows the failure management circuitry to be mechanically inter-
faced with the control paths. This feature affords a more secure means of sensing a failed or degraded
control path without reducing the reliability of the control path, and, hence, the transducer can be
used to cover failures up to the power piston. Several other ways to provide a valve feedback signal
for this failure monitor concept were considered. For example, differential pressure across the
second stage of a conventional servovalve can be used. Also, the current flow through the first
stage (flapper valve coil) can be used and is more economical. However, neither of these approaches

4175
 .- •v • - ~w
. , . ._, .I W! e T-- 702 %W

NADC-82240-60

will provide 100 percent failure coverage and were discarded in favor of the valve position trans-
ducer approach. Valves of this type are currently available.

The basic failure sensing function for each power actuator channel is provided simply by using four
equal value resistors in conjunction with the 4-valve position transducers. Connection of the re-
sistors as shown in Figure 115 constitutes a very simple and reliable voting concept that allows each
control path to comparatively monitor itself, determine a failure, and disengage itself.

Figure 117 is a simplified schematic of the basic concept for sensing a failure. For normal opera-
tion, the voltage across the valve position transducers should be the same. Since the voltages across
the transducers are the same regardless of valve position, there will be no appreciable current in the
resistors. Current will flow only in the resistors when the valve positions are not in agreement If
one control path has a "hard" failure, the respective porting stage will fully displace while the others
will partially displace in the opposite direction. The voltage differences will cause a current in the
resistor associated with the failed control path that is several times higher than the current in the
other resistors, thus providing a means for identifying the failed path. For example, if Valve 2b is
hardover, the other three will be displaced a small amount (depending on the actuator load) in the
opposite direction and each will produce a transducer output voltage.

See References 20 and 21 for a more detailed description of the failure sensing function.

b. Automatic Detection/Disengage Function

This function is the part of the failure management system that detects the occurrence of a failure
and isolates the fault by disengaging the affected control path. If the fault is not of sufficient mag-
nitude to warrant a disengagement, it is presented to the pilot as a soft-fail (e.g., high null) indica-
tion. The soft-fail feature is a cautionary device for the pilot and constitutes a BITE function.

Several approaches for detecting failures were considered. One approach was to simply compare
the magnitude of the failure voltage across each sense resistor with a set threshold. The second
approach used a scheme for comparing the four failure voltages to determine a failure. This approach
is not as simple as the first approach, but it appeared to be more tolerant to failures and was success-
fully used in the BHT 4-valve FBW feasibility program. Subsequent studies, however, indicated that
an improved arrangement of the first approach has some advantages and, as stated above, was se-
lected as the preferred approach. The major advantage of the first approach is that it has less failure
modes and has a high degree of operational independence. The selected failure detection technique
affords a failure management system that is very simple when compared to the more conventional
voting schemes.

See References 20 and 21 for a more detailed description of the automatic detection/disengage
function.

LABORATORY TESTS AND DEMONSTRATIONS

The objective of the laboratory test and demonstration program was to provide a means of apprais-
ing the 4-valve actuation system concept as a candidate actuation concept for use in the FCS of a
potential test vehicle. The test program allowed the 4-valve actuation concept to be evaluated in
terms of operational suitability audits ability to tolerate failures.

Tests were conducted to investigate the effects of the basic types of failures that can occur. The
intent was to validate the 4-valve actuation concept as a viable fault-tolerant actuation system that

176

----- " '"' - -- "- ... ? ,.' ", , ' . .

NADC-82240-60

-4-
<00.

0 U0

<00
>a
CL

cr--

LLJ -
F- -

>.
_j)

UU-

177
 NADC-82240-60

can be interfaced with the FCS of the test aircraft. The AFCS control paths, up to and including
the Electro-Hydraulic Servo Valves (EHSV's), were tested to assure a Fault Tolerance-Level (FTL)
of dual fail-operate for the worst conditions. The electrical and hydraulic power systems were
tested to assure that the failure effects on the total system would result in an FTL of single fail-
operate and dual fail-safe. The failure modes were simulated using the switches on a failure sim-
ulation panel, which contained four electrical power switches, two hydraulic hand valves, and
combinations of these input devices. Pertinent parameters were measured and recorded to define
failure effects.

The test results demonstrated that the 4-valve actuation system concept is suitable for use in the
FCS of the test aircraft Tests validated the concept in terms of operational suitability and ability
to tolerate failvres

The methodology used in the Integrated Test Program and the test results are fully described in
Reference 20.

178
 NADC-82240-60

REFERENCES

1. Rosemount Inc., Bulletin 1014C, 1974

2. Weinstein, W., Feasibility and Design Studies of an Integrated Sensory Subsystem (ISS) for
Advanced V/STOL Aircraft, Grumman Aerospace Corporation, NADC 76259-30,
March 1978.

3. Hooker, D.S., et al, Definition Study for an Advanced Fighter Digital Flight Control
System, AFFDL-TR-75-59, June 1975.

4. Weinstein, W., Development of an Advanced Skewed Sensory Electronic Triad (ASSET)

System for Flight Control, (Phases I and II. Analytical and Laboratory Investigations),
Grumman Aerospace Corporation, NADC 76295-30, October 1976.

5. Solomon, R., Development of an Advanced Skewed Sensory Electronic Triad (ASSET)

System for Flight Control, (Phases I IIA and B. Laboratory Verification and Flight Test).
Grumman Aerospace Corporation, NADC 77043-30, June 1979.

6. Grobert, K., Donnelly, H., Toolan, W., and Zislin, A., Development of an Integrated Sensory
Subsystem (ISS) for Advanced Aircraft (Phase II, Design and Laboratory Verification).
Grumman Aerospace Corporation, NADC-78083-60, January 1981.

7. Miyashiro, S.K.and Morris, F.E., "VTOL/Helicopter Approach and Landing Guidance

Sensors for Navy Ship Applications," presented at Navy/NASA V/STOL Flying Qualities
Workshop Naval Postgraduate School, Monterey, California, April 26-29, 1977.

8. Edwards, A., Jr., "The State of Strapdown Inertial Guidance and Navigation," ION Journal,
Winter 1971-72, p. 386.

9. W. C. Wi,,, "F-16 Flight Control Maintenance Training Manual," General Dynamics - Fort
Worth Division, Report No. 16PR840A, 23 January 1978.

10. Davis, J. D. L., "Side Stick Controller Grip Safety of Flight Test Report (CDRL 1045C),"
General Dynamics- Fort Worth Division, Report No. 20PR086A, 9 February 1981.

11. Ramage, J. K., Abrams, C. R., and Watson, J. H., "AFTI/F-16 Digital Flight Control System
Status Report," presented at the 4th Digital Avionics System Conference, St. Louis,
Missouri, November 1981.

4 12. NR72H-240,"Feasibility Study for Advanced Flight Control Actuation System (AFCAS)",
Rockwell International Corporation, Columbus Aircraft Division, Contract N62269-72-C-
0108, June 1972, Unclassified. AD 767 058.

13. NR73H-107, "Control-By-Wire Actuator Model Development for AFCAS", Rockwell Inter-
national Corporation, Columbus Aircraft Division, Contract N62269-73-C-0405,
4| January 1974, Unclassified. AD 772 588.

179
., NADC-82240-60

REFERENCES (Continued)

14. NR75H-1, "Control-By-Wire Modulator Actuator Tests (AFCAS)", Rockwell International

Corporation, Columbus Aircraft Division, Contract N62269-73-C-0405, January 1975,
Unclassified. AD A-006 371.

15. NR76H-1, "Design and Fabrication of an 8000 psi Control-By-Wire Actuator for Flight
Testing in a T-2C Airplane," Rockwell International Corporation, Columbus Aircraft Division,
Contract N62269-75-C-03 11, January 1976, Unclassified. AD-A024 487/IG I.

16. Demarchi, J. N. and Haning, R. K., "Flight Verification of the Advanced Flight Control
Actuation System (AFCAS)" in the T-2C Aircraft, NAVAIRDEVCEN 75287-60, Columbus
Aircraft Division, Rockwell International Corporation, Contract N62269-76-C-0201,
June 1978, Unclassified.

17. Kohnhorst, L. L. and Magnacca, D. A., "Flight Verification of Direct Digital Drive for an
Advanced Flight Control Actuation System (AFCAS) in the T-2C Aircraft," NAVAIRDEV-
CEN 78207-60, North American Aviation Division, Rockwell International Corporation,
Contract N62269-76-C-0201, November 1979.

18. Kohnhorst, L. L, "Direct Digital Drive Actuation," presented at the 4th Digital Avionics
System Conference, St. Louis, Missouri, November 1981.

19. Kohnhorst, L. L. and Magnacca, D. A., "Design and Test of a Hydra-Optic Flight Control
Actuation System (HOFCAS) Concept," NAVAl RDEVCEN 79156-60, North American
Aviation Division, Rockwell International Corporation, Contract N62269-79-C-0709,
December 1980.

20. Haskins, D. E., "Fault-Tolerant Actuation Concept for a Research Test Aircraft," Report
No. NADC-79125-60, Bell Helicopter Textron, Contract No. N62269-79-C-0292,
October 1980.

21. Murphy, M. R., "Four-Valve, Direct Drive, Advanced Actuation System," presented at the
Advanced Flight Control Symposium, USAF Academy, Colorado Springs, CO.

180
 NADC-82240-60

ACKNOWLEDGEMENTS

Most of the descriptive, graphical, and photographic material in this report was obtained from
the engineering reports published by the contractors who advanced the development of the described
technologies and concepts under contracts issued by the Navy and Air Force.

I18

181
 NADC-82240-60

Distribution List

No. of Copies

Grumman Aerospace Corporation

Bethpage, Long Island, NY 11714
A ttn : K . Grobert ...................................................

Rockwell International
Columbus, OH 43216
A ttn: L. Kohnhorst .............................................. ..

Bell Helicopter Textron

Fort Worth, TX 76101
A ttn: M . M urphy ...................................................

General Dynamics Corporation

Forth Worth, TX 76108
A ttn: J. W atson ....................................................

Superintendent
Naval Postgraduate School
M onterey, CA 93940 .................................................

Commander
Naval Air Systems Command (Al R00D4)
Department of the Navy
Washington, DC 20361
(2 - for retention)
(1 -for Air 340D) ..................................................... 3
Administrator
Defense Technical Information Center
,. Cameron Station, Building 5
Alexandria, VA 22314
A ttn: DT IC-D D A -1 ................................................. 12

AFFDL
Wright Patterson AF B,
O H 4 5433 ... .. .. ...... . ... .. .. .... ...... ...... ... ... ........ ..... .
NAVAI RDEVCEN
* Warminister, PA 18974
3 for Code 8131)
(1 for Code 60)
II for Code 601)
(1 ior Code 6012)
(20 for Code 60122)
S(1 for Code 6013)
(1 for Code 60 134) ..... ......................................... ... 28
 NADC-82240-60

Distribution List (Continued)

No. of Copies
National Aeronautics and Space Administration
Ames Research Center
Systems Integration B ranch
Moffett Field, CA 94035
A ttn: D. H. Brocker ................................................. 2
 4 ''-14.

-q - .

j.0.
4r,

V OD
k- 4.1

411

4VI