FLIGHT TESTING OF MANUAL FLIGHT CONTROL FUNCTIONS FOR A SMALL TRANSPORT AIRCRAFT (PROJECT ATTAS-SAFIR)

Thomas Heintsch Robert Luckner Daimler-Benz Aerospace Airbus GmbH Flight Guidance and Control P.O. Box 95 01 09 21111 Hamburg, Germany

Klaus-Uwe Hahn lnstitut fgr Flugmechanik DLR Lilienthalplatz 7 38108 Braunschweig, Germany

7 Summary In a technology programme DASA has developed flight control laws (FCL) for an electronic flight control system of a small transport aircraft (lOOseater). In a cooperation between DASA and DLR, the flight control functions were tested on DLRs VFW614/ATTAS test aircraft. This paper gives an overview of the flight control law development and testing within the SAFIR (Small Airliner Flight Control Law !nvestigation and Refinement) flight test project. Design objectives of the flight control system for the 1 OO-seater are reviewed, a system overview is given, the flight control law functions are briefly explained and the development process is described. The testing procedure comprises the SAFIR experiment integration into the ATTAS test system, the definition of the flight tasks, the flight testing and the evaluation of the flight test results. 2 Nomenclature AOA ATTAS angle of attack Advanced Technologies System computer aided software

ND

2 PFD RA SAFIR

Navigation Display normal load factor Primary Flight Display radio altimeter Small Airliner Flight Control Law Investigation and Refinement time calibrated airspeed airspeed corresponding to o,,, maximum operational speed airspeed corresponding to o,,,, Central Communication Computer (Zentraler Kommunikationsrechner) angle of attack flight path angle elevator deflection pitch attitude aileron deflection lright hand) bank angle roll attitude rate

t VCAS V m,,m VMO V e.101 ZKR

0 V 0 Testing Aircraft fR 0 Q

CASE DASA DLR

engineering 3 Introduction

Daimler-Benz Aerospace Airbus Deutsche Forschungsanstalt fur Luft- und Raumfahrt Electronic Flight Control System Experimental and Control Computer (Experimental- und Regelrechner fly by wire flight control law Flight Control Law Computer Instrument maximum Landing System operational Mach number

EFCS ERR

FBW FCL FCLC ILS MM0

Today, the application of fly by wire (FBW) technology to civil transport aircraft can be considered as the state of the art. Airbus lndustrie was the first aircraft manufacturer to make use of this technology. Now a family consisting of five aircraft types (A31 9, A320, A321, A330, A3401 is under production with 700 aircraft in service and 1300 orders (as at April 19961. The projection for the end of the century is that more than 1000 aircraft with EFCS will be flying and more than 10,000 pilots will have been trained. Taking this scenario into account and considering advantages and disadvantages of mechanical and electronic FCL systems DASA decided to use fly-

Paper presented at the AGARD FVP Symposium on Advances in Flight Testing. held in Lisbon, Portugal, 23-26 September 1996, and published in CP-593.

26-2

by-wire technology for its loo-seater project. Facing the project it was clear that design to cost would be a major challenge in EFCS development for such a small transport aircraft because the price of a commercial transport is related to the number of passenger seats, and can be given in terms of dollars per seat. As the price of avionic systems - such as EFCS - is virtually independent of aircraft capacity (assuming equivalent performance), the EFCS share in the specific cost of a loo-seater is nearly twice that of a 200seater 111. 4 Flight Control System 4.1 Design Objectives

Minimum requirements for EFCS are defined by the certification regulations. Additional requirements come from market analysis and the perceptible progress of teChnOlOQy. Due to the unfavorably high share that EFCS contributes to the specific cost of a small transport aircraft, cost efficiency becomes the dOrt7inatinQ factor in the EFCS design process. Flight control law design, as part of EFCS design, will influence: - system development costs; - aircraft qualification costs (stability and control); - crew training costs; - modification costs. A cost-effective FCL approach must be simple, based on proven techniques that are accepted by the certification authorities, and should use familiar standard control functions in order to reduce crew training costs. Special attention has to be paid to the software design process as modern design tools (CASE tools) may significantly reduce software design costs if properly applied. With a scalable controller structure and modular design, DASAs FCL approach is not exclusively targeted to a specific aircraft but can be adapted to other aircraft types.

a rudder trim switch; two handwheels for control of the trimmable horizontal stabilizer; two priority push buttons; - a speed brake lever; - a slat/flap lever; - a thrust lever. Sensors that measure feedback parameters include: - three air data/inertial reference units; - two radio altimeters; - accelerometers. Additional sensors necessary e.g. for ground spoiler logic are: - wheel tachometer; - landing gear compressed switch; - thrust lever position sensor. For flight parameter indication, the primary flight display is used. Two lamps indicate priority if requested by one of the pilots. System status and failure conditions are displayed on the Engine Display and the System Display, respectively. 4.3 Flight Control Functions The FCLs provide both primary control functions (pitch, roll and yawl and secondary control functions (slat/flap, airbrake and ground spoiler). In addition, they calculate operational and limiting speeds as well as parameters dealing with the flight envelope protections, and display these on the PFD. Generally, the functional design is aimed at crew commonality with AIRBUS aircraft; all functions and the man-machine interface are similar in order to reduce the cost of transition training. However, everything that is hidden to the pilot - i.e. systems and software is realized differently. For normal operation (that is, as long as no systerns have been degraded due to failures) three modes are necessary: - ground mode; - inflight mode; - flare mode. The transition from one mode to the other has to be smooth, with no adverse effect on the pilots ability to control the aircraft. The normal laws provide complete flight envelope protection as follows (See Fig. 3): - load factor limitation; - high angle of attack protection; - high speed protection; - pitch attitude protection; - bank angle protection. Envelope protections are designed to prevent specified boundaries from being exceeded. They assist the pilot by initiating corrective action if necessary, but they do not assume the pilots decision-making role or his responsibility for safe

4.2 System Overview The principle of the loo-seater EFCS is shown in Fig. 1 as a block diagram. All flight control surfaces are electrically controlled and hydraulically activated (Fig. 2). The stabilizer and the rudder have an additional mechanical link as a back-up. In the cockpit, we find: - two side sticks for pitch and roll control (not mechanically coupled); - two pair of pedals for yaw control frigidly interconnected);

flight. A more detailed description of the protection functions is given in 141. In the event of multiple system failures, the FCLs shed protection functions or degrade from the normal law to the direct law, according to the number and nature of the successive failures. lnflipht Mode Pitch Normal Law. The pitch normal law is a load factor demand law with automatic trim function. At low speed, load factor is blended with pitch rate. With the side stick at neutral during level flight, this law provides short-term flight path stability and compensates for turbulence. Turn compensation is provided up to 33O bank angle. Roll Normal Law. Roll normal law is a roll rate demand/bank angle hold law. The roll rate demand is proportional to side stick deflection and limited to * 15/s. Bank angle hold is provided up to i33O bank with automatic turn coordination and turn compensation. This allows turns to be flown in normal airline operations without pitch input. Yaw Normal Law. The yaw normal law is a direct control-to-surface law (pedals to rudder) with maximum deflections limited by the rudder travel limitation function. Additionally, - yaw damping, - turn coordination, and - automatic trim in case of engine failure is provided. Flare Mode In order to provide a conventional flare (where the pilot has to pull the side stick back progressively to achieve a gently increasing pitch attitude during flare), longitudinal control changes at 50 ft from inflight to flare mode: - automatic trim is deactivated; - a modified normal law with load factor and pitch rate feedback is activated. Ground Mode On the ground, side stick deflections correspond directly to elevator, aileron and roll spoiler deflections. There is no automatic pitch trim. After takeoff, the flight mode is progressively blended in. Direct Laws In case of system or sensor failures which prohibit the correct performance of the normal laws, FCLs are degraded to the direct laws. The direct laws are direct control-to-surface laws without any protections and a rough scaled set of feedforward gains.

Development of a system is an iterative process, and several different models are used to describe it. In the V-Model, the analytical steps are listed on the left leg, and the steps towards a synthesis on the right leg. The links between them define verification and validation activities on different levels (see Fig. 4). FCL development is part of the system development process, and a rational and methodical approach can reduce development costs significantly. The elements - CASE Tool: HOSTESS - test facilities: development flight simulator - flight test: VFW61 4/ATTAS flying test bed. play a keyrole in the optimization of the FCL development process (Fig. 5). 5.1 CASE Tool HOSTESS

Based on experience (e.g. with the center of gravity control computer for A300 and A310). DASA has developed the CASE tool HOSTESS (High Order Structuring 1001 for Embedded System Software). Its goal is to standardize software specification, to automate the coding process, to provide automatic checks and testing, to improve software documentation, end to facilitate configuration management. HOSTESS provides: - Software specification using a graphical block diagram language in conjunction with a symbol library and assembly rules which are easily understood by electronics and automation engineers; - a consistency check of each module; - automatic coding with different code Qeneraters (FORTRAN, ADA, etc.); - a hierarchical software structure. The benefits are: - a reduction in coding errors; - the automatization of routine activities during the software development cycle, especially for modifications; - standard, unambiguous software specification. 5.2 Development Flight Simulator

5 FCL Development

Process

Tests on the flight simulator are conducted in order to validate the FCL functions and their reconfiguration in a real-time environment. In particular, the transitions from one mode to another in combination with different pilot inputs and various flight States can Only be investigated in real time. The DASA flight simulator facility has been continously upQraded over the past years and now features: - high fidelity simulation models; - generic loo-seater cockpit (sidestick,

6 displays, etc.); - visual system; - sound system; data acquisition and analysis. More than 1000 simulator flight hours have been flown with engineers and various test and airline pilots at the controls. An ethernet data link between the development workstations where the FCLs are specified and coded by means of HOSTESS and the flight simulation computer allowed a highly efficient development process. 6 ATTAS 6.1 ATTAS Flight Test Facility Test Aircraft

the flight task. He can disengage the FBW control system by switching off or by overriding the control actuators. The actuators are force limited to avoid structural damages of the basic aircraft. For safety reasons the VFW61 4 flight envelope is restricted in the FBW and SIM modes: a maximum cruising altitude of 25000 ft, maximum cruising speed of 288 kts (0.65 Mach) calibrated airspeed (CA% and a rather low landing speed of about 100 kts is adequate for a transport aircraft flight regime representation. The high maneuverability is adequate for a broad spectrum of flight experiments L61. Fig. 8 illustrates the ATTAS FBW envelope available for the SAFIR flights. 6.2 ATTAS System Simulator

In the early 80s the flying simulator and demonstrator aircraft ATTAS (Advanced Technologies Testing Aircraft System) was developed as a testbed by DLR and DASA (former MBB) supported by the German Ministry of Research and Technology. The system architecture of the ATTAS in-flight simulator is outlined in several papers [2,3,51. ATTAS is based on a VFW614, twin-turbofan, short haul 44.passenger aircraft (Fig. 61. The aircraft is equipped with a complete flight test instrumentation. The heart of ATTAS testbed is the experimental fly-by-wire system l20ms cycle time) with interfaces to aircraft systems IARINC 429, MIL-Bus 15538, etc.). All on board computers are of MIL-specified LORALlROLM types IMSE 14 and HAWK 32). 1.4 MFLOPS are available for experimental functions on the HAWK/32 32-bit computer with 8 MB of rnemory. The network is based on a ring structured serial fibre optical bus system. In flight test ATTAS can be operated in three principal modes: - basic mode: VFW61 4 standard mechanical flight control; - FBW mode: FBW system active as a direct link; FBW system active with user- SIM mode: defined functions. The aircraft is operated by two pilots, an experimental and a safety pilot. The experimental pilot is sitting on the left-hand side. His side of the cockpit is equipped with a side stick controller and two displays (PFD and ND, Fig. 71. He controls the aircraft via the FBW computer system. The FBW actuator inputs are mechanically fed back to the safety pilots control column on the right-hand side which is mechanically linked to all controls. This feed back function enables him to monitor the movement of the control surfaces. This is an important safety aspect, because he can evaluate, whether the inputs are adequate to

For testing and training on ground, DLR has build up the ATTAS system simulator which is a one by one duplicate of the ATTAS flight test sy stem. It consists of: - a fixed base flight simulator cockpit; - a simulation computer with a VFW61 4 simulation model; - the original ATTAS data processing system. The ATTAS system simulator was used for the verification of the FCL software that had been developed in DASAs flight simulator, the adaption to the ATTAS real world system, the experimental check out and the pilots familiarisation and test training. 7 SAFIR Project In April 1993 DASA and DLR launched the SAFIR project for the investigation of flight control laws developed by DASA for a loo-seater aircraft. Within this project FCLs were validated, demonstrated and evaluated in flight with DLRs VFW614/ATTAS as test aircraft. It was decided to adapt DASAs FCLs to the VFW61 4s flight dynamics, instead of using ATTAS as an inflight simulator for the lOOseater with the original loo-seater FCLs. Not more than half a year later, in October 1993 the first flight was carried out. The aim of the first test campaign (SAFIR II was to evaluate the functionality and performance of the FCLs within the whole flight envelope. The test campaign comprised 6 test flights. In 130 single tests all functions of the normal laws and protection laws were investigated. Special points of interest were control accuracy of the FCLs, handling qualities and the transition between the different operating modes. In SAFIR 2, the second project phase the emphasis was laid on the investigation of system aspects. For that purpose FCLs were doubled in two dissimilar software languages (FORTRAN,

ADA) to be implemented in a command and a monitor lane. Based on the same block diagram specification the design tool HOSTESS automatically generated the two software packages. This configuration enabled the investigation of: - monitoring between command and monitor lane; - data consolidation between command and monitor lane; - degradation from normal law to direct law. Within the SAFIR 2 campaign 6 test flights with 175 single tests carried out in the years 1995 and 1996 by several test pilots and an airline pilot having different flight experience. The total flight time for the whole SAFIR project was 32 hours with 305 single tests (Tab. 1). The evaluation of the flight test results lead to modifications in the FCLs which always were considered in the software for the next test flight. All in all 110 modifications were tested in 9 different software versions. 7.1 Experimental Set-Up 7.1 .l ATTAS Test System Configuration In preparation of the flight tests the FCL software had to be implemented in the ATTAS data processing system. A software interface had to be realized - that provided the input signals for the FCL software module in the appropriate format, - that connected the FCL output signals to control surfaces and displays. Additionally adaptions were necessary concerning: - data recording of user defined signals; - modification of the side stick force characteristic; - generation of automatic test functions for FCL input (e.g. synthetic side stick inputs); - assignment and interpretation of FCL control signals (e.g. activation of test functions, switching between modes); - Modifications of PFD and ND to indicate test relevant signals (FCL messages, characteristic speeds, synthtic ILS signal). 7.1.2 Software Implementation (SAFIR 1 I In the first project phase the FCLs were computed on the Experimental end Control Computer (ERR) of ATTAS. The software was transferred via floppy disc and copied one by one. With a configured ATTAS test system no additional modifications were necessary for implementing a new software on the assumption of an identical FCL interface. An essential stability criterion of a flight control

system is the total time delay (time between pilots input and control surface movement). For the ATTAS test system it can be determined by the fifthfold of the ERR cycle time. With a value of 35ms the time delay was 175ms. 7.1.3 Hardware Modifications (SAFIR 2)

Due to the doubling of FCL software for command and monitor lane in the second project phase, the performance of the Experimental and Control Computer was exceeded. Therefore a high performance flight worthy Elight Control Law Computer (FCLC) a HARRIS Night Hawk 4401 has been installed in the ATTAS test system and linked to the Central Communication Computer (ZKRj (Fig. 9). The HARRIS computer I1 CPU VME-bus board with a MOTOROLA 88100 processor) computes the command and monitor lanes of the flight control laws within a cycle of 20ms. For the communication between FCLC and ZKR six ARINC 429 highspeed lanes (three input lanes, three output lanes) were provided (Fig. 10). The installation of the FCLC required major software modifications both to implement the FCL software into the FCLC including monitoring and data consolidation functions and to link the FCLC to the ATTAS test system. With a data transmission time of 80ms (ZKR->FCLC->ZKRj the total time delay was 180ms (520ms for ATTAS test system) which is in the same magnitude as for SAFIR 1. 7.2 Flight Tasks Definition Flight tasks for the validation of FCL functions within the normal flight envelope comprise side stick or pedal inputs in order to check: - the aircrafts dynamic reaction in terms of time constant, damping behaviour, etc.; - the ability of the normal FCLs to hold flight path angle and bank angle after side stick release; - the correct operation of functions like turn coordination, turn compensation, automatic pitch trim, one engine out compensation. Checking the performance of the flight envelope protections special attention is paid to: - a smooth transition between normal law and protection; - adequate maneuverability up to the target limits without overshoot. For investigations of handling quality, a special maneuver (Fig. 11) was defined by DLR, in which the pilot flew a vertical pattern in the normal law, running into the bank angle protection during the 45O bank turn. For evaluation a pilot questionnaire was prepared (Fig. 12) asking for Cooperl-

26-6 Harper ratings of specific tasks during the malSC!. Handling qualities as classified by the pilots were Level 1 within the normal flight control laws and -as expected- Level 2 within the bank angle protection, where the pilot had to stabilize the bank angle with the side stick (bank angle command) and comfort functions like turn compensation or turn coordination were deactivated. The FCL performance in normal airline operation was investigated with typical maneuvers like: - ILS approach; - heading changes in climb/descent; - step climb/descent with turns. As the minimum altitude in FBW and SIM mode was restricted to 500ft, the ILS approaches had to be carried out in higher altitudes. For that purpose a synthetic ILS signal was generated, with localizer and glideslope deviations indicated on the PFD and ND (Fig. 13). The raw data ILS approach (no flight director) included localizer and glideslope intercept with a 1ONM final spprosch. The synthetic ILS signal was activated via a switch operated by the flight test engineer. To ensure a maximum of efficiency flight tests had to be carefully prepared. The test flights consisted of single flight tasks, each specified on a test procedure card (Fig. 14). The test cards contained detailed information about the aircrafts initial flight condition and the test sequence in terms of a verbal description of every single step. Pilot marks indicate begin and end of test for later identification. Cooper/Harper ratings and pilot comments were noted. In advance of the test flight the complete test program was flown in the ATTAS system simulator with the purposes: - final check of the FCL software; - familiarization of the test pilot with the FCLs; - training of the test procedures; - optimization of the test programme. Typically a test flight comprised between 18 and 36 single tests within about 2.5h-3h of flight time. 8 Flight Testing 8.1 Execution of Experiment On board of ATTAS the DLR flight test crew was supported by DASA engineers, a flight test engineer and two FCL experts. The flight test engineer read out the test procedure cards and was responsible for the correct processing of the flight tasks. The FCL experts monitored the tests by means of a quick-look terminal which display ed test relevant signals online in numerical or graphical form. By a set of 15 switches online modifications in the test software during the flight tests was possible. Setting of the switches was defined in a switch declaration list (Fig. 15). The switches were used for e.g: _ degradation from normal law to direct law; - activation of the synthetic ILS signal; - variations of side stick characteristic (deadspace and slope); - activation of synthetic side stick step inputs. The evaluation of the flight began immediately after the flight in the debriefing with a complete walk through the test procedure cards. The pilot comments were discussed including comparisons between flight simulator and flight test and nOtation of unexpected events during the flight. A first analysis was possible a few hours after the flight when the first test results were available from the DLR data processing system. These overview time histories of the relevant experiment sequences were plotted on three wall-papers (six papers for SAFIR 2). each representing up to 33 flight test parameter. Although of limited resolution (due to data reduction) the time histories were well suited for an assessment and often most of the open items of the debriefing could be explained already in this early state of evaluation. The first two test flights of the SAFIR 1 programme were dedicated to adjustine the FCLs to the ATTAS F8W system as to: - sensor calibration and filtering; - fine tuning of gains (e.g. turn coordination at low speed); - adaotion to the force reduced FBW actuators. Minor software modifications became necessary. During the following flight tests the ATTAS adapted FCLs were tested within the whole flight envelope and showed good quality without any malfunctions. Modifications in the flight control software were necessary only in order to optimize gains and transitions between the different operating modes. As an important result, the FCL development methodology proved to be efficient. Software modifications could be performed within a short time, reliably and accurately, using HOSTESS. Simulator tests showed high fidelity with respect to the flight tests except for the well-known and accepted shortcomings: real-life/visual system disparity, motion, and pilot anxiety levels. Due to careful preparation flight tests were performed on high efficiency level. 8.2 FliQht Test Results

From the whole number of single flight tests four representative time histories are presented.

26-7

Bank Angle Protection Fig. 16 shows the results of a maneuver for testing roll control. Pilots task was a doublette of side stick maximum roll commands ILH, RH) without rudder and thrust inputs. Up to 33O bank angle, roll control is a roll rate command/bank angle hold function with a maximum of 15Ols commanded roll rate, which in the test is achieved just when reaching the -33 bank angle. Beyond -33, bank angle protection with a bank angle command becomes active limiting the maximum bank angle to -45 (according to ATTAS flight envelope). The transition between roll rate command and bank angle command is smooth, the maximum bank angle is kept without overshoot. During the following bank to bank maneuver the aircraft rotates with maximum admissible roll rate of 15/s up to +45O bank. The test sequence ends with a side stick release which lets the aircraft automatically roll back to 33O bank angle. Here the bank angle hold function stabilizes the constant turn as required. High AOA Protection The operation of high AOA protection is illustrated in Fig. 17. The pilot decelerates the aircraft by reducing thrust to idle without side stick input. At t 6 2Os, AOA reaches (I,,,, and the high AOA protection is activated resulting in a transition from n,-command to o-command (stick to neutral commands ear.,* full back stick commands maximum admissible angle-of-attack o,,, which is 2O-3 beneath the stall-AOA). The aircraft stabilizes u,,, with a reduction in flight path angle due to thrust setting. The aircraft descends with stabilized o=o,,., and V,,, = V,,,,,. At t = 85s the pilot commands (I,,, by pulling the side stick full back. Despite this step input the nose up reaction is smooth. (I,,, is reached without overshoot. Side stick release lets the aircraft automatically stabilize o,,,, again. High AOA Protection (dynamic) Fig. 18 shows that the high AOA protection operates precisely even in dynamic maneuvers, such as the side stick full back step input at V CAs=180kts. The aircraft reacts with a An, =0.8g pitch up movement up to 22O pitch attitude. The high AOA protection counters this dynamic nose-up maneuver with a 4O nose down elevator command. The aircraft stabilizes at u = qm and V,,, = V,,,,. In neither of the high AOA protection tests the maximum angle-of-attack o,,, has been exceeded. High Speed Protection The high speed protection limits the maximum speed beyond V,, lV,,,,,,,s=255ktsl. For a sudden full nose down command the aircraft must not exceed V,,+30kts tM,,+O.O7l. for a long term full nose down command, maximum airspeed is limited to V,, + 16kts IM,, + 0.04).

The time histories in Fig. 19 show the aircraft reaction on a sudden full nose down command initiated at a speed of V,,, = 225kts. The aircraft rapidly pitches down and is stabilized at 0 =-15O by the pitch attitude protection. With a speed advance of 1Okts due to pitch attitude, side stick command and horizontal acceleration, high speed protection is activated at V,,, - 240kts. With a combination of side stick nose down command reduction and nose up load factor command a smooth intercept maneuver is performed. The aircraft achieves a maximum speed of 285kts i = V,, + 30kts.T and then slowly decelerates towards V,, + 16kts until side stick pitch command is released. The pitch behaviour fullfills the requirements precisely although additionally a turn maneuver with alternating full side stick roll inputs is performed. Due to a bank angle limitation of 30 in High Speed Protection the aircraft rolls up to f 30bank angle without overshoot. After side stick roll command release wings levelled out again as required (positive spiral stability). As with these examples, all other FCL functions have been tested successfully. Pilots judged the inflight results to be consistent with the simulator results despite transition processes due to mode switching, particularly from the n,-command to the o-command. This transition, which is sensitive to pitch rate and load factor cues, could only be optimized by means of flight testing. 9 Conclusions

A comprehensive set of manual flight control functions is being developed for implementation in an EFCS for a small transport aircraft on the premise of an AIRBUS similar functional design. The FCLs have been validated by flight simulator tests and have been favorably received by test and airline pilots. For inflight tests, they were adapted to the dynamics and operational constraints of the DLRs VFW614/ATTAS flying test bed. Supported by a small but well trained flight crew ATTAS was ideally suited to offer the possibility of using its FEW capability to investigate and operate the Iuncertified) experimental FCL software. It permitted flight tests in a real world environment with a short term experiment preparation at low cost and low risks. Within the SAFIR project performed by a joint DASA/DLR team, a very effective and well organized infrastructure has been built up between the partners. In more than 300 single flight tasks all FCL functions were investigated in detail with different test and airline pilots. The flight test results showed high precision of the FCL perfor-

mance with respect to the requirements. The handling qualities for airline typical operations were rated as Level 1. 10 Acknowledgements The authors would like to thank their collegues from DASA Airbus in Hamburg and DLR in Braunschweig, who contributed to the success of the SAFIR flight test campaign.
11

f31

D. Hanke, J.-M. Bauschat, H.-H. Lange, H. Heutger: In-Flight-Simulator ATTAS-System Design and Capabilities DGLR Symposium In-Flight Simulation for the 90s DGLR Conference Proceedings 91-05 Braunschweig, July 1990 R. Luckner, T. Heintsch Development of Manual Flight Control Functions for a Small Transport Aircraft Vortrag ICAS-94-7.6.1, Anaheim, Sept. 1994 J.-M. Bauschat, H.-H. Lange ATTAS and its Contributions to System Design and In-Flight Simulation International Conference Aircraft Flight Safety, Conference Proceedings, pp. 403-413, 1993 J.J. Buchholz, J.-M. Bauschat, K.-U. Hahn, H.J. Pausder ATTAS & ATTHES In-Flight Simulators Flight Vehicle Integration Panel Symposium, AGARD CP 577, Braunschweig, May 1995

[41

References H. Seidel: Cockpit and Avionics for a Small Jet Transport Aircraft ERA Technology Ltd. Avionics Conference, London, December 1992

Ill

151

I21 D. Hanke, H.-H. Lange: Flight Evaluation of the ATTAS Digital Flyby-Wire/Light Flight Control System Vortrag ICAS-88-3.6.1 Jerusalem, Aug./Sept. 1988

[61

Fip.

EFCS Block Diagram

Fia.

Control

Surfaces

Fia.

Normal

Flight Control

Law Modes

-----t-----j-----------

A------~wai.~-------. Hardware Prcduction

Fia. V-Model for FCL Development Process

----

Fip.

Flight Control

Law Development

Process

Fip.

VFW614/ATTAS

Test Aircraft

Fip.

VFW614/ATTAS

Expermental Cockpit

Fip.

ATTAS FBW Envelope

Fia. Tab 1. A SAFIR Flight Test Overview

FCLC Integration into ATTAS Test System

Dktance x

Diitance y

Fia. 10:

Communication between ZKR and FCLC

!
Ficl. SAFIR Maneuver

FiQ. 12: Pilot Comment Card and Questionnaire

Primary Flight Display

NAV Display

Fia. 13: Synthetic ILS Approach

Fia. 15: SAFIR Switch Declaration List

Fia. 14: SAFIR Test Procedure Card

Fia. 16: Bank Angle Protection

Fia.

High AOA Protection

..

Fia. 16: High AQA Protection (dynamic)

Fia. 19: High Speed Protection