itp-tvc-eurofighter

17
11-1 Industria de Turbo Propulsores, S.A. Thrust Vectoring Nozzle for Modern Military Aircraft Daniel Ikaza Industria de Turbo Propulsores S.A. (ITP) Parque Tecnológico, edificio 300 48170 Zamudio, Spain [email protected] presented at NATO R&T ORGANIZATION Symposium on ACTIVE CONTROL TECHNOLOGY FOR ENHANCED PERFORMANCE OPERATIONAL CAPABILITIES OF MILITARY AIRCRAFT, LAND VEHICLES AND SEA VEHICLES Braunschweig, Germany 8 th -11 th May 2000 Fig. 1.- CFD Model of a TVN ABSTRACT This paper describes the technical features of the Thrust Vectoring Nozzle (TVN) developed by ITP and its advantages for modern military aircraft. It is presented in conjunction with two other papers by DASA (Thrust Vectoring for Advanced Fighter Aircraft High Angle of Attack Intake Investigations) and MTU-München (Integrated Thrust Vector Jet Engine Control) respectively. Thrust Vectoring: advantages and technology Thrust Vectoring is a relatively new technology which has been talked about for some time, and it can provide modern military aircraft with a number of advantages regarding performance (improved manoeuverability, shorter take-off and landing runs, extended flight envelope, etc...) and survivability (control possible in post-stall condition, faster reaction in combat, etc...). Additionally, as a byproduct of Thrust Vectoring, there is also the capacity to independently control the exit area of the nozzle, which allows to have always an “adapted” nozzle to every flight condition and engine power setting. This means an improvement in thrust which in cases can be as high as 7%. There are several types of Thrust Vectoring Nozzles. For example, there are 2-D (or single-axis; or Pitch-only) Thrust Vectoring Nozzles, and there are 3-D (or multi-axis; or Pitch and Yaw) Thrust Vectoring Nozzles. The ITP Nozzle is a full 3-D Vectoring Nozzle. Also, there are different ways to achieve the deflection of the gas jet: the most efficient one is by mechanically deflecting the divergent section only, hence minimizing the effect on the engine upstream of the throat (sonic) section. The major aerodynamic aspects of the design of a Thrust Vectoring Nozzle include the correct dimensioning of the vectoring envelope, accounting for the difference between “geometrical” and “effective” vectoring; as well as the sealing pattern of the master and slave petals at and around the throat section. Paper presented at the RTO AVT Symposium on “Active Control Technology for Enhanced Performance Operational Capabilities of Military Aircraft, Land Vehicles and Sea Vehicles”, held in Braunschweig, Germany, 8-11 May 2000, and published in RTO MP-051.

Upload: defencedog-khanz

Post on 23-Mar-2016

260 views

Category:

Documents


6 download

DESCRIPTION

itp-tvc-eurofighter

TRANSCRIPT

11-1

Industria de Turbo Propulsores, S.A.

Thrust Vectoring Nozzlefor Modern Military Aircraft

Daniel Ikaza

Industria de Turbo Propulsores S.A. (ITP)Parque Tecnológico, edificio 300

48170 Zamudio, [email protected]

presented atNATO R&T ORGANIZATION Symposium

onACTIVE CONTROL TECHNOLOGY FOR ENHANCED PERFORMANCE

OPERATIONAL CAPABILITIES OF MILITARY AIRCRAFT,LAND VEHICLES AND SEA VEHICLES

Braunschweig, Germany8th-11th May 2000

Fig. 1.- CFD Model of a TVN

ABSTRACT

This paper describes the technical features of the ThrustVectoring Nozzle (TVN) developed by ITP and itsadvantages for modern military aircraft. It is presented inconjunction with two other papers by DASA (ThrustVectoring for Advanced Fighter Aircraft High Angle ofAttack Intake Investigations) and MTU-München (IntegratedThrust Vector Jet Engine Control) respectively.

Thrust Vectoring: advantages and technology

Thrust Vectoring is a relatively new technology which hasbeen talked about for some time, and it can provide modernmilitary aircraft with a number of advantages regardingperformance (improved manoeuverability, shorter take-offand landing runs, extended flight envelope, etc...) andsurvivability (control possible in post-stall condition, fasterreaction in combat, etc...).

Additionally, as a byproduct of Thrust Vectoring, there isalso the capacity to independently control the exit area of thenozzle, which allows to have always an “adapted” nozzle toevery flight condition and engine power setting. This meansan improvement in thrust which in cases can be as high as7%.

There are several types of Thrust Vectoring Nozzles. Forexample, there are 2-D (or single-axis; or Pitch-only) ThrustVectoring Nozzles, and there are 3-D (or multi-axis; or Pitchand Yaw) Thrust Vectoring Nozzles. The ITP Nozzle is a full3-D Vectoring Nozzle. Also, there are different ways toachieve the deflection of the gas jet: the most efficient one isby mechanically deflecting the divergent section only, henceminimizing the effect on the engine upstream of the throat(sonic) section.

The major aerodynamic aspects of the design of a ThrustVectoring Nozzle include the correct dimensioning of thevectoring envelope, accounting for the difference between“geometrical” and “effective” vectoring; as well as thesealing pattern of the master and slave petals at and aroundthe throat section.

Paper presented at the RTO AVT Symposium on “Active Control Technology forEnhanced Performance Operational Capabilities of Military Aircraft, Land Vehicles and Sea Vehicles”,

held in Braunschweig, Germany, 8-11 May 2000, and published in RTO MP-051.

MY
Highlight

11-2

Fig. 2.- TVN ground tests at ITP

ITP Design

The ITP concept consists of a patented design featuring theso-called “Three-Ring-System”, which allows all nozzlefunctions (Throat Area, Exit Area, Pitch Vectoring and YawVectoring) to be performed with a minimum number ofactuators, which, in turn, leads to an optimized mass andoverall engine efficiency.

The nozzle is controlled by four independent hydraulicactuators, each one with its own servovalve and positiontransducer. The level of redundancy will depend on the exactapplication.

There is also, only at design level, a simplified variant of thenozzle, with three actuators only, which has basically thesame functions of the 4-actuator one, except the independentexit area control. This variant would be a little lighter, but itmisses the thrust improvement capability.

The reaction bars of the ITP nozzle present an arrangementwhich allows for high deflection angles, without the risk ofpetal overlapping and/or disengagement. The presentprototype has demonstrated deflections up to 23º, but studieshave been performed of variants of the nozzle with deflectionangles of up to 30-35º.

Finally, the ITP design makes use of a partial “Balance-Beam” effect, which takes advantage of the energy of the gasstream, on one hand to help close the nozzle under highpressures, hence reducing the maximum load required fromthe actuators; on the other hand to allow self-closing of thenozzle in case of hydraulic loss under low pressureconditions, specially interesting to retain thrust for take-off.

ITP TVN programme - Past, Present and Future

ITP has dedicated a research programme on Thrust Vectoringtechnology which started back in 1991, and which met animportant milestone as is the ground testing of a prototypenozzle at the ITP facilities in Ajalvir, near Madrid, in 1998.Altitude testing is scheduled for 2000.

The next major goal will be the realisation of a flightprogramme, in order to validate the system in flight, andevaluate the capabilities and performance of the system as ameans of flight control.

The design used for the ground prototype has been a simpleone, for short life and limited safety. For the flight standard,a number of changes will be introduced, in order to comply

with the very demanding aircraft requirements in terms ofweight, life, safety, etc...

The experience acquired during the ground test phase hashelped ITP learn a lot about the behaviour of the nozzle,showing areas in which the design has to be modified,providing data for the integration of nozzle and enginecontrols, etc... The main outcome of the tests has been theconfirmation that the concept designed is valid and it workssmoothly.

A decisive contribution is being done by ITP’s partnercompany MTU of Munich, Germany, by developing theelectronic Control System.

This programme is making the Thrust Vectoring technologyavailable in Europe for existing military aircraft such asEurofighter, in which the introduction of Thrust Vectoringcould be carried out with a relatively small number ofchanges to the aircraft and to the engine, and could provide itwith an improved performance.

CONTENTS

1.- DEFINITIONS AND ABBREVIATIONS

2.- BACKGROUND

3.- BENEFITS OF THRUST VECTORING AND NOZZLEEXIT AREA CONTROL

4.- TYPES OF THRUST VECTORING

5.- ITP DESIGN: BASELINE AND OPTIONS

6.- ITP THRUST VECTOR PROGRAMME

7.- CONCLUSIONS

8.- ACKNOWLEDGEMENTS

1. - DEFINITIONS AND ABBREVIATIONS

A8 Nozzle throat area

A9 Nozzle exit area

ATF Altitude Test Facility

CFD Computational Fluid Dynamics

Con-Di Convergent-Divergent

DECU Digital Engine Control Unit

DOF Degree of freedom

ESTOL Extremely Short Take-Off and Landing

FCS Flight Control System

R&D Research and Development

RCS Radar Cross Section

SFC Specific Fuel Consumption

SLS Sea Level Static

TVN Thrust Vectoring Nozzle

MY
Highlight
MY
Highlight

11-3

2.- BACKGROUND

The Thrust Vectoring Nozzle developed by ITP was initiallydesigned to fit and be compatible with an EJ200 engine,which powers the European Fighter Aircraft EF2000. Thisaircraft is developed by the European consortiumEurofighter, constituted by the companies British Aerospace(UK), DASA (Germany), Alenia (Italy) and CASA (Spain).Similarly, the above engine EJ200 is developed by theEuropean consortium Eurojet, constituted by the companiesRolls-Royce (UK), MTU (Germany), Fiat Avio (Italy) andITP (Spain).

The current EJ200 engine is equipped with a variable-geometry Convergent-Divergent (Con-Di) Nozzle, developedby ITP. This nozzle can modify the area to match the enginerunning point and afterburner setting, but it has no vectoringcapability.

Through a dedicated R&D programme, ITP have nowintroduced a new Thrust Vectoring Nozzle which could beapplied to EJ200 to significantly enhance the capabilities ofEF2000 Aircraft.

Introduction to Military Aircraft Nozzles

In a military aircraft engine with reheat (also calledafterburner or augmentor), the nozzle presents a convergentsection, which has the task to accelerate the gas jet in orderto generate thrust, yet with the characteristic that it must becapable of varying the throat area according to therequirement of the engine running point. These are called“variable geometry convergent” nozzles.

Some nozzles, additionally, comprise a divergent sectiondownstream of the convergent section, which overexpandsthe jet between the throat area and the exit area in order toextract yet some extra thrust. These are called “Variablegeometry convergent-divergent” (or Con-Di) nozzles.

Depending on the level of control upon this divergentsection, variable geometry Con-Di nozzles can be of twotypes:

• One-parameter Nozzles: also called 1-DOF nozzles; theConvergent section (hence Throat Area) is fullycontrolled, and Divergent section (hence Exit Area)follows a pre-defined relationship to the convergentsection behaviour (throat area). The current EJ200 nozzleis of this type.

• Two-parameter Nozzles: also called 2-DOF nozzles; theConvergent section (Throat Area) and Divergent section(Exit area) are fully controlled independently. This typecan match the Divergent section to the exact flightcondition in order to obtain an optimised thrust.

Also there are some intermediate solutions such as “floating”and some other “passive” means of exit area control, whichare outside the scope of this paper.

One solution or the other is chosen according to theparticular requirements of each case, in terms of weight, cost,reliability, thrust, priority missions, etc…

In the case of Thrust Vectoring Nozzles, they also have thetask to direct the jet to generate side thrust to transmit it tothe aircraft structure, so that the aircraft can make use of it asa means of flight and manoeuvre control.

3.- BENEFITS OF THRUST VECTORINGAND NOZZLE EXIT AREA CONTROL

Although the description of benefits of Thrust Vectoring andnozzle exit area control for modern military aircraft is in factthe subject of a separate paper by DASA, a brief descriptionof some of them is given here for reference.

They can be basically grouped in four categories:

• Enhanced performance in conventional flight

• Post-Stall flight

• Increased Safety

• Reduction of aero controls

Enhanced performance in conventional flight

The concept of Thrust Vectoring is often associated withspectacular loop-type manoeuvres performed by smallaircraft in airshow demonstrations or combat simulations,and the operational use of these capabilities is often regardedwith a lot of skepticism, due to the trends of modern aircombat. However, there is a lot more to Thrust Vectoringthan these funny manoeuvres, and in fact the greatestargument in favour of Thrust Vectoring is not found incombat characteristics but rather in conventionalperformance, as described in more detail below:

Stationary Flight Trimming

The use of the Nozzles as a complementary control surfaceallows the aircraft to better optimize its angle of attack instationary level flight for a given flight point and loadconfiguration, hence reducing the drag, which in turn leadsto strong benefits in SFC, and therefore range.

Fig. 3.- Increased Sustained Turn Rate with TVNs

Stationary and Transient Manoeuvres

Similarly to the above case, the nozzles can be used toincrease the maximum load factor that is achievable undercertain circumstances while maintaining the aircraft trimmed.This applies both for stationary manoeuvres (sustained turnrate) and for transient manoeuvres (rapid deceleration).

Nozzle Exit Area Control

As described in the Introduction to military aircraft nozzles,in one-parameter Con-Di nozzles the divergent section(hence A9) follows a pre-defined relationship to the

INCREASED SUSTAINED TURN RATE

Lift

Weight

Drag

Available

Thrust

Conventional Trimming : CASE STUDY:

30,000 ft Mach 1.8

Flaps 6º upMax. Lift: Ref.

Max. Load Factor: Ref.Max. Turn Rate: Ref.

Flaps 2º upNozzles 4º up

Lift Coef.: Ref +14% Load Factor: Ref +9%Turn Rate: Ref +9%

Velocity Vector

Lift

Weight

Drag

Thrust Vectoring Trimming:

Velocity Vector

Rear FlapsTrimming

Thrust VectoringTrimming

AvailableThrust

MY
Highlight
MY
Highlight
MY
Highlight
MY
Highlight
MY
Line
MY
Highlight
MY
Highlight
MY
Highlight
MY
Highlight
MY
QuestionMark
MY
Line
MY
Highlight
MY
Line
MY
Highlight

11-4

convergent section (hence A8). This relationship is optimisedfor an average of all missions, which normally means lowA9/A8 Ratio for dry operations (without reheat) and highA9/A8 Ratio for operations with reheat.

In rough terms, this is reasonably optimised for low speeddry operations (cruise, climb, etc...) and for high speed reheatoperations (high speed strike, etc...), but is not optimised forlow speed reheat operations (take-off) and high speed dryoperations (supersonic cruise).

EXIT AREA OPTIMIZATION

Throat Area

OptimumA9/A8 ratioincludingafterbody

effects

Dry Reheat

Supersonic

Subsonic

FixedSchedule

IndependentA9 control

Supercruise:

7%Thrust increase

Take-off:

2%Thrust increase

Fig. 4.- Optimization of Nozzle Exit Area

The use of an independently controlled divergent sectionallows A9 to be optimised for any engine running conditionat any flight point, and has an improvement especially inthose conditions where one-parameter A9/A8 Ratio is notoptimized.

For example, for a supersonic cruise case (Mach 1.2, altitude36,000 ft, engine at Max Dry condition) of EJ200 engine onEurofighter, the use of independent A9 control could lead toan improvement of up to 7% in installed net thrust relative tothe current performance. This is due to the combination oftwo effects: increase of nozzle internal thrust; and reductionof nozzle external drag.

In addition to thrust increase, independent A9 control alsopermits reduction in SFC for certain flight points.

Reduction of take-off and landing runs

The rotation of the aircraft for take-off and landing can beaccelerated by using Thrust Vectoring. Also, ThrustVectoring can be used to increase angle of attack, hence lift,while maintaining a trimmed aircraft. The combination of allthese effects gives an important reduction in the take-off andlanding runs for an aircraft such as Eurofighter.

Global mission performance

The combined effect of all the above items across a typicalcombat mission could add up to some 3% Fuel saving byusing Thrust Vectoring.

Post-Stall Flight

The most spectacular benefit of Thrust Vectoring, althoughpossibly not the most important, is the fact that it can exertan active control of an aircraft while the main aerodynamicsurfaces are stalled, hence not suitable for control, and thisopens a whole new domain of flight conditions where flightused to be unthinkable.

Extension of flight envelope

The use of Thrust Vectoring permits the aircraft to holdstationary flight in an area of the envelope whereconventional controls are not sufficient.

In the Altitude/Mach-number envelope, Thrust Vectoringpermits an extension of the envelope in the low speed-medium height region. In the Altitude/Mach-number/Angle-of-attack envelope, Thrust Vectoring permits operation atmuch higher values of Angle of attack.

Air superiority

A better control of the aircraft is achieved with ThrustVectoring, especially at low speed conditions, whereconventional aerodynamic controls are not effective, andwhere a good number of combat scenarios are to take place.

ESTOL

The ESTOL concept (Extremely Short Take-Off andLanding) is becoming more and more appealing to militaryaircraft operators, and it consists of performing the Take-offand Landing manoeuvres with the aircraft stalled. It reducestake-off and landing runs by a large amount.

This is only possible with Thrust Vectoring Nozzles, thatoperate when the aerodynamic controls are no longer useful.

Increased safety

This is probably one of the strongest arguments in favour ofThrust Vectoring. The existence of redundant means ofaircraft control allows for a better survivability.

Fig. 5.- Redundant Flight Controls with TVNs

In peace time, an aircraft crash by loss of aerodynamiccontrol could be avoided by the use of Thrust Vectoring.

In war time, damage to aerodynamic control surfaces can becompensated with Thrust Vectoring.

Next Step: reduction of aero controls

Once the Thrust Vectoring system has been sufficientlyvalidated, it will be a primary control for the aircraft. This

Foreplane

Rudder

FlapsLeading EdgeSlats

Nozzles

MY
Highlight
MY
Highlight
MY
Highlight
MY
Line
MY
Highlight
MY
Line
MY
Highlight
MY
Highlight
MY
Highlight
MY
Line

Center of Gravity (cg)

For trimmed flight, no rotation about cg.

W· Wing Lift

dw• distance

Trimmed Aircraft

w T· Tail Lift

d t· distance

T

Equation: (W x~) + (T x d tl = 0

(Lift of Wing x distance from cg) + (Lift of Tail x distance from cd) - 0

MY
Highlight
MY
Typewritten Text
http://www.grc.nasa.gov/WWW/k-12/airplane/trim.html

11-5

means that it will allow a gradual reduction of existingconventional aerodynamic control surfaces such ashorizontal and vertical stabilizers. This will have an impact,and there will be a reduction in:

• Mass

• Drag

• Radar Cross Section (RCS)

The extent of these impact could only be properly assessed inthe future, and it will probably not be fully exploited until thenext generation of combat aircraft, but mass reductions of15%-20% of the total aircraft are conceivable.

4.- TYPES OF VECTORING NOZZLES

From the point of view of the type of actuation means, TVNscan be classified:

• Fluidic Actuation: The deflection of the gas flow isachieved by injection of secondary airflows. This type isspecially suitable for fixed-area high expansion nozzles,such as those used in rockets and missiles.

• Mechanical Actuation: The deflection of the gas flow isachieved by mechanical movement of the nozzle, whichis powered by hydraulic or pneumatic actuators. Thistype is specially suitable for variable geometry militaryaircraft nozzles.

* * *

From the point of view of the direction of vectoring, TVNscan be classified:

• Single-Axis TVNs: (also called 2-D or Pitch-only) Thedeflection of the gas flow is achieved in vertical directiononly. They replace and/or complement horizontal controlsurfaces. This type is suitable for all types of variablegeometry military aircraft nozzles.

• Multi-Axis TVNs: (also called 3-D or Pitch and Yaw)The deflection of the gas flow is achieved in anydirection. They replace and/or complement horizontaland vertical control surfaces. This type is speciallysuitable for round nozzles.

* * *

If we focus on 3-D, Con-Di military aircraft TVNs withmechanical actuation, there are several ways to materialisethe vectoring:

• Deflect whole nozzle. The disadvantages are: a largemass has to be moved; and there is a big impact onperformance upstream of the nozzle.

• External Flaps. The disadvantages are: there is a need foradditional mass; and the efficiency of vectoring is verylow.

• Deflect Divergent section. This is the preferred solution.the size of the nozzle is optimised and the effect onperformance is negligible. The ITP Nozzle is of this thirdtype.

EXISTING 3-D THRUST VECTORING SYSTEMSMechanical Actuation, Con-Di Military Nozzles

DEFLECT WHOLE NOZZLE EXTERNAL FLAPS DIVERGENT SECTION

Fig. 6.- Existing types of 3-D TVNs

Regarding the nozzles of the third type, that is, those thatdeflect the flow by orienting the divergent section only, theygenerally need actuation means for:

• Controlling convergent section (hence A8)

• Controlling divergent section (hence vectoring and A9)

Where other designs make use of two separate actuationsystems, the ITP design has a unified actuation system with aminimum total number of actuators.

5.- ITP DESIGN: BASELINE AND OPTIONS

One of the biggest problems encountered when designing aThrust Vectoring Nozzle is how to find a mechanicalconfiguration comprising casing, rings, etc…, which must becompatible with both functions of the nozzle: on one hand,open and close the convergent section to control throat area(optionally open and close the divergent section to controlexit area); and on the other hand to direct the nozzle indirections different to axial, to obtain the jet deflection thatprovides vectored thrust.

The other big problem of a Thrust Vectoring Nozzle is howto find an actuation system (hydraulic, pneumatic, electro-mechanical, mixed, etc..) capable of generating themovements required in the nozzle, to accomplish all theabove functions, and reasonably limited under criteria suchas weight, size, etc…

Many different configurations have been studied at ITP forTVNs, the result being a “baseline” configuration, plus aseries of options available for every particular application.

The main option is the A9 modulation capability, aimed atoptimising the thrust as described above in the chapter“Benefits of Thrust Vectoring and nozzle area control”.

Baseline

The baseline ITP TVN design is a Convergent-divergentaxisymmetric (round) nozzle with multi-axis ThrustVectoring, mechanically actuated, and where the deflectionof the gas flow is achieved by orienting the divergent sectiononly. This way the moving mass is minimized, and thedistortion to the engine turbomachinery upstream of thenozzle is negligible.

It has three degrees of freedom (DOFs), namely: Throat area(A8), Pitch vectoring and Yaw vectoring. Any obliquevectoring is made of a combination of pitch and yaw. Exitarea (A9) follows a certain relationship to A8.

The actuation system consists of only three independenthydraulic actuators, a fact which is made possible by thebasic feature of the design: the “Three-Ring-System”.

MY
Highlight
MY
Highlight
MY
Line
MY
Line
MY
Line
MY
Highlight
MY
Line
MY
Line
MY
Highlight
MY
Highlight
MY
Highlight
MY
Line
MY
Accepted

11-6

X

OUTERRING

ACTUATORS

Fig. 7.- Three Ring System (3 actuators)

This system consists of three concentric rings which arelinked by pins and form a universal (or “cardan”) joint. Theinner ring is linked to the convergent section of the nozzle,the outer ring is linked to the divergent section through thereaction bars, and the intermediate ring acts as the crossbarbetween the inner and outer rings. The actuators are linked tothe outer ring only. The design of the rings and reaction barsis such that a small tilt angle on the ring is amplified to alarge deflection angle on the divergent section.

The outer ring can be tilted in any direction while the innerring can only keep a normal orientation to the enginecentreline, but they both are forced to keep the same axialposition along the engine. This is the key factor that permitsa full control of the nozzle by acting on the outer ring only,hence minimizing the total number of actuators.

For pure throat area movements, all three actuators move inparallel, hence all three rings follow axially, and A8 is set tothe appropriate value. A9 follows a pre-defined relationshipto A8 according to the dimensions of the mechanism.

For Pitch and/or Yaw vectoring movements, the threeactuators move differently, hence defining a tilt plane of theouter ring. The divergent section will deflect in the directionof that plane. Throat area (A8) is not affected unless thismovement is combined with a throat area movement.

Fig. 8.- Ring Movement in vectoring

Fig. 9.- Nozzle Movement in vectoring

A9 Control option: optimised thrust

This option consists basically of the baseline design, exceptfor the fact that the outer ring is split in two halves, forming a“hinged” outer ring.

It has four degrees of freedom (DOFs), namely Throat Area(A8), Exit Area (A9), Pitch vectoring and Yaw vectoring.Again, any oblique vectoring is achieved by combination ofpitch and yaw.

The actuation system consists of four independent actuators,also linked to the outer ring only.

THREE RING SYSTEM Z

YX

PITCH, A9

YAW, A8

SPLITRING

ACTUATORS

Fig. 10.- Three Ring System (4 actuators)

The same Three Ring principle is used as in the threeactuator version, and A8 and vectoring movements areoperated in a similar way, yet this time with four instead ofthree actuators.

Additionally, pure A9 control movements are performed bymoving top and bottom actuators in parallel while the othertwo stay static, hence “hinging” the outer ring open or close.The divergent section opens or closes relative to the nominalposition, acquiring an “oval” shape. Hence this movement issometimes referred to as “ovalization”. Of course, A9movements can be combined with A8 movements and/orvectoring movements.

Click at left to close this box
Click at left to close this box
Click at left to close this box
Click at left to close this box
Click at left to close this box
Click at left to close this box
Click at left to close this box
MY
Highlight
MY
Highlight
MY
Line
MY
Highlight
MY
Highlight
MY
Line
MY
Highlight
MY
Highlight
MY
Line
MY
Accepted
MY
Line
MY
Highlight

11-7

Fig. 11.- Ring Movement in A9 control

Fig. 12.- Nozzle Movement in A9 control

With this configuration there could be an improvement ininstalled net thrust of up to 7% in certain conditions.

In fact, this A9 option could well be considered as thebaseline, leaving the non-A9 configuration as a “simplifiedoption”.

Third Member of the Family: "Two-Ring" Pitch-only Nozzle

This is a simplified version of the ITP Nozzle where theintermediate Ring is deleted, hence reducing some weightand complexity. Outer Ring is split in two as in previousversion.

It retains the four actuators and it has three DOFs (A8, A9,Pitch Vectoring).

It is suitable for application in aircraft with no Post-Stallcapability, but where the benefits in conventional flight areimportant.

SIMPLIFIED TWO-RING SYSTEM FOR PITCH-ONLYAPPLICATIONS

YX

PITCH, A9

A8

SPLITRING

Fig. 13.- "Two-Ring" Pitch-only Nozzle

Other features: "hinged" Reaction Bars

The design of the reaction bars presents “hinged struts”which allow an optimised smooth movement of petals.Where other designs are limited to about 20º geometricdeflection by the disengagement and/or interference betweenpetals, the ITP design allows for growth if required, andstudies have been carried out for deflections up to 30º-35º.

SIMPLE REACTION STRUTS

VECTORING

Disengagement

Interference

Fig. 14a.- Vectoring with simple reaction bars

HINGED REACTION STRUTS

VECTORING OptimizedMovement

Fig. 14b.- Vectoring with hinged reaction bars

Balance-Beam

The ITP TVN makes use of a partial balance-beam effect,which consists of taking advantage of the energy of the gasstream to help close the nozzle in high pressure conditions.

The closing movement of the nozzle is accompanied by anaxial displacement of the throat, so that the volume sweptagainst the gas pressure is modified, in particular morevolume is swept in the low pressure region of the nozzle, andless volume in the high pressure region.

Click at left to close this box
Click at left to close this box
Click at left to close this box
Click at left to close this box
Click at left to close this box
Click at left to close this box
Click at left to close this box
MY
Highlight
MY
Highlight
MY
Line
MY
Line
MY
Highlight
MY
Highlight
MY
QuestionMark

11-8

15% LOWERACTUATOR LOADS

NOZZLE SELF-CLOSINGIF HYDRAULIC LOSSDURING TAKE-OFF

BALANCE BEAM

Fig. 15.- Balance Beam effect

This has two benefitial effects:

• On one hand, in high pressure conditions, the total workperformed by the actuation system upon the gas streamis reduced by as much as 15%, which results in smalleractuator dimensions and better engine efficiency.

• On the other hand, in case of hydraulic loss in lowpressure conditions, the nozzle self-closes, which isparticularly interesting to retain thrust during take-off.

Actuation and Control System

The control system of the nozzle consists of three (baselinedesign) or four (A9 option) independent actuators, each withits own servovalve and position transducer. The servovalvesare powered by the engine hydraulic pump; the electroniccontrol loops and safety logic between servovalves andtransducers are performed by the TVN Control Unit, whichis built into the engine DECU, which, in turn, is connected tothe aircraft Flight Control System (FCS).

HYDRAULICPUMP(S)

FLIGHTCONTROLSYSTEM

SERVOVALVE 1

SERVOVALVE 2

SERVOVALVE 3

SERVOVALVE 4

ACT. 1

ACT. 2

ACT. 3

ACT. 4

DECU VNCU

TRANSD. 1

TRANSD. 2

TRANSD. 3

TRANSD. 4

OUTER RING

GEARBOX

INTEGRATED CONTROL SYSTEM

FCS BUS

Fig. 16.- TVN Control System

For a twin-engine application such as Eurofighter, a simplehydraulic system and dual electrical system provide enoughsafety for a primary control.

On the other hand, for a single-engine application, there willprobably be a need for duplex hydraulic system and triplexelectrical system.

Changes to EJ200 engine

Relative to current EJ200 engine, the introduction of a TVNwith "full capability" implies a number of changes:

• Nozzle

• Nozzle actuators, including Servovalves and transducers

• Bigger Hydraulic Pump

• DECU, including Thrust Vectoring functions

• Casing reinforcement

• Slight modification to engine mounts

• Reheat Liner, especially rear attachment

• Dressings (pipes and harnesses)

However, a reduced-capability TVN version of EJ200 isfeasible with very minor changes.

In any case, these changes are small if compared with theadvantages obtained by introducing Thrust Vectoring.

Advantages of ITP design

In summary, the ITP design presents a number of advantagesrelative to other designs, such as:

• Minimum number of actuators, which leads to lowerweight and better overall engine efficiency.

• Unique reaction bar design for high deflection angles.

• Partial Balance-Beam effect for lower actuator loads.

• Nozzle self-closing in case of hydraulic loss during take-off allows thrust retention

• It is the only proved example of 3-D TVN for 20,000 lbfthrust engine class.

6.- ITP TVN PROGRAMME

ITP’s R&D programme on Thrust Vectoring technologystarted in 1991, and within this programme a good number ofgeneral studies have been performed, including:

• CFD analyses

• Performance studies

• Concept design: Baseline plus options

• Trade-off studies with side loads, number of petals, etc...

• Patents

• Mechanical / Kinematic simulations

• Mock-ups

• etc...

* * *

Additionally, a feasibility study has been carried out togetherwith DASA regarding the application of TVN forEurofighter. The outcome of this study includes thedefinition of the requirements for the TVN on the

MY
Highlight
MY
Highlight
MY
Highlight
MY
Line
MY
Highlight

11-9

Eurofighter, and some of the operational benefits expectedfor Eurofighter.

An initial study was done in 1994-95, and an update study isbeing conducted now 1998-2000, this time with MTU alsotaking part.

* * *

ITP and MTU have a special co-operation agreement underwhich MTU has developed the electronic Control Systemthat controls the ITP TVN and actuators.

Prototype Nozzle

In 1995 ITP launched what is called a “TechnologyDemonstration Phase” within the Thrust Vectoringtechnology R&D programme. This phase includes thedesign, construction and test of a prototype Thrust VectoringNozzle. The design of the prototype started in early 1996 andthe first run took place in July 1998, becoming the keymilestone in ITP Thrust Vectoring programme so far.

This prototype nozzle was aimed at demonstrating as muchas possible, even if some things were not necessarilyrequired from the aircraft point of view. Therefore it wasdesigned for high vector loads (30 kN) even if the aircraftrequirement will be not higher than 15 kN. Similarly, itincorporated the A9 option to optimise thrust. A deflection of20º was specified for any engine running condition.

The prototype nozzle was constructed for an EJ200 enginevehicle, but maintaining a minimum impact on currentEJ200, both regarding the hardware changes, as well asregarding the development programme.

In principle, only Sea Level Static (SLS) tests werescheduled, namely the ITP testbed in Ajalvir, near Madrid.However, the nozzle was specified to take the loads of thefull flight envelope, and real flight standard materials wereused in its construction, so that the mechanism could bevalidated as far as possible.

Most of the components were manufactured in ITP, hencekeeping a high degree of flexibility to introduce quickchanges in the design.

As part of the work associated to the tests of the prototypenozzle, a new detuner (exhaust duct) had to be installed inthe ITP testbed (Cell No.2) at Ajalvir. The need for this newdetuner was motivated both by the different flow pattern inthe cell, and also by the need for cooling.

MODIFICATIONS TO TEST CELL:NEW FRONT SECTION OF DETUNER

(WATER-COOLED SEGMENTS)

NOZZLE

DOUBLE SKIN WATER COOLING

WATER SUPPLY LINEEXISTINGREAR DETUNER

MODIFIEDDIAMETERAND LENGTH

EXTERNALFILM COOLING

Fig. 17.- Modifications to Test Cell

* * *

The test results obtained during the running of the prototypeinclude the following highlights:

• 80 running hours, including 15 with reheat

• Vectoring in all 360º directions, both dry and reheat

• 23,5º maximum vector angle

• 110º/sec maximum slew rate

• 20 kN maximum lateral force

• Programmed ramps and Joystick control

• Thermal case: sustained 20º vector in reheat for 5minutes

• Rapid transients Idle-Dry-Reheat while vectoring

• 100+ performance points run

• Exit area control: 2% thrust improvement

• Endurance: 6700+ vectoring cycles

• Endurance: 600+ throttle cycles (with sustained 20ºvector)

The nozzle performed smoothly and free of mechanicalfailures

* * *

The conclusion of the ground tests in Ajalvir represents thefulfilment of the Technology Demonstration Phase. Fromthis point onwards, the next steps to be taken include acontinuation of the general studies on Thrust Vectoring, aswell as the continuation of the feasibility study with DASAand MTU.

Additionally, altitude tests with the prototype nozzle arescheduled for the second quarter of 2000 at the Altitude TestFacility (ATF) in Stuttgart.

The next big milestone in he Thrust Vectoring programmewill necessarily be a flight programme, in order to validatethe TVN in flight condition. Consequently, ITP as well as allITP’s partners are strongly pursuing this possibility.

ITP THRUST VECTORING NOZZLE PROGRAMME

ITP THRUST VECTORING NOZZLE PROGRAMME

1990-94 1995 1996 1997 1998 1999-2002

Concept Design, Patents, Mock-upsPreliminary and Aerodynamic DesignDemonstrator Nozzle DesignAdvanced Material OrderingsComponent ManufactureManufacture of ActuatorsDesign of Control System (MTU)Component TestingManufacture of Control System (MTU)Actuator and Control System Testing (MTU)Test Bed ModificationPrototype Assembly and InstrumentationIntegration in EJ200 EngineGround TestingExtensive Instrumentation ResultsFlight Programme

Fig. 18.- ITP Thrust Vectoring programme

7.- CONCLUSIONS

• Thrust Vectoring offers great advantages for modernmilitary aircraft, in return for relatively small changes inthe aircraft, and is clearly the way to go for the future.

MY
Accepted
MY
Highlight
MY
Highlight
MY
Highlight

11-10

• Thrust Vectoring technology has become available inEurope, helped by the R&D programme conducted byITP, especially after the ground test of the prototypenozzle.

• The ITP design presents some advantages relative toother designs, which may prove vital on the long term.

• The aerospace community in Europe is actively in favourof this technology, and the institutions are willing tosupport this.

• With a very small number of changes to EF2000, ademonstration flight programme would be possible andproduce a very important stepping stone for theintroduction of this technology into service.

8.- ACKNOWLEDGEMENTS

The success of ITP’s programme has only been possible withthe contribution of partners and organizations, namely:

Spanish Ministries of Industry and Defence, with fundingthrough an R&D programme

MTU, of Munich, Germany, developed the electronic controlsystem under a dedicated agreement with ITP.

Eurojet and the Partner Companies (Rolls-Royce, MTU andFiat-Avio), as ITP’s partners in the EJ200 developmentprogramme for Eurofighter, gave support to ITP.

CESA, of Getafe, Spain, designed the hydraulic actuators forthe TVN.

Sener, of Las Arenas, Spain, who started in the programmemany years ago, also contributed to the engineering work ofthe programme.

DASA, of Munich, Germany, as partner in the Eurofighterfeasibility study, provided the assessment of requirementsand benefits for EF2000 with TVN.

11-11

Paper#A11Q, by P. M. Lodge: What is the level of redundancy of the nozzle actuation?

A. (D. Ikaza) Simplex for the ground tests. Simplex will also be taken to flight for EF2000

���������������� ���� �� ��� ���

������������� ���� ����

The contract to supply Eurofighter's propulsion system was awarded to EuroJet Turbo GmbH in

1986. EuroJet is a consortium of companies from each partner nation and after workshare

changes; Rolls-Royce of the UK has 36%, MTU (Motoren-und Turbinen-Union) of Germany has

30%, FiatAvio of Italy has 20% and ITP (Industria de Turbo Propulsores) of Spain has 14%. The

partnership has resulted in the EJ200 advanced turbofan and associated management systems.

 EuroJet 200

Origins and Technology

The EJ200 started life in 1982 as the Rolls Royce/British MoD XG-40 Advanced Core Military Engine or ACME demonstrator. This

programme, split into three phases; technology (1982-88), engine (1984-89) and assessment (1989-95) developed new fan,

compressor, combustor, turbine (including high temperature life prediction) and augmentor systems using advanced materials and

new manufacturing processes. The first full engine commenced rig testing in December 1986 with the final XG-40 running for some

200 hours during 4000 cycles bringing the programme to a close in June 1995.

Upon formation of the EuroJet consortium in 1986 much of the

continuing XG-40 research was used for the new programme. The

requirements were for a powerplant capable of higher thrust, longer

life and less complexity than previous engines. The result was a

powerplant with similar dimensions to the Tornado's RB199 yet having

almost half as many parts (1800 against 2845 for the RB199) and

delivering nearly 50% more thrust. A very noticeable difference

between the two engines can be seen by comparing the turbine blade

designs. Compared to the RB199 the EJ200's blades are enormous and

show leanings towards sustained transonic and supersonic flight profiles.

The EJ200 is an advanced design based on a

fully modular augmented twin-spool low bypass

layout. The compressor utilises a three stage

Low Pressure Fan (LPF) and a five stage High

Pressure Compressor (HPC). The fan features

wide-chord single crystal blade/disc (blisk)

assemblies designed for low weight (including

the removal of guide vanes), high efficiency

operation. The three stages achieve a pressure

ratio of around 4.2:1 with an air mass flow of

some 77kg/s (or 170lb/s). Like the fan the five

stage compressor also features single crystal

blisk aerofoils. The use of single crystals and blade/disc units can both bring enormous potential

advantages to how the powerplant may be operated (see fact box).

Following the fan/compression stages fuel is injected via an annular combustor designed for low

smoke operation. The key factors in determining jet engine efficiency and achievable work are

the temperature and pressure differences attained between the engine inlet and combustor

outlet. In the EJ200's case the outlet stator temperature is in excess of 1800K with a pressure

ratio (achieved in just eight stages) of some 25:1.

Such a high combustor temperature requires special precautions be taken with the High

Pressure Turbine, or HPT which is directly downstream. To help reduce this problem the HPT uses

air cooled single crystal blades. However there is a limit to what can be achieved using air

cooling. In fact it eventually becomes detrimental to use cooling because it adversely effects the

achievable combustion temperature and thus reduces efficiency. To overcome this the EJ200's

HP turbine blades also utilise a special Thermal Barrier Coating, or TBC. This barrier is

comprised of two plasma deposited layers, a special bonding coat over which a top layer of a

Nickel-Chromium-Yttrium ceramic material is applied. Although this increases the life of the blade

and increases the achievable operating temperature it does require regular inspection to

ensure the coating remains viable. Following the single HPT is a further single Low Pressure

Turbine (or LPT) stage again employing single crystal blades. In both the HPT and LPT a powder

metallurgy disc is employed. A titanium alloy based mono-parametric convergent/divergent

(Con-Di) nozzle completes the engine improving achievable thrust while helping to optimise the

system for different flight profiles.

Turbo GmbH

Cross section of EJ200 © MTU

Cross section of rotor blades

Single Crystal Turbine Blades

On the microscopic scale all

metals (and their alloys) are an

arrangement of crystals. How

these crystals are arranged, their

size and distribution depends on

what the metal or alloy is and

how it was forged.

The vast majority of turbine

blades are now made from a

Nickel or Titanium based super-

alloy. The alloy will contain a

variety of elements (their

amounts are typically proprietary

and well guarded!) such as

Tungsten, Cobalt, Chromium, etc.

The resulting material is

extremely hard and difficult to

machine and thus blades tend to

be forged in a process called

investment casting. When forged

using traditional constant

temperature furnaces the blade

contains many small precipitates

of various compounds (Cobalt,

Tungsten, Ni3Al, Ni3Ti, etc.). This

resolves certain physical

problems with older generation

blades (such as problems with

power law creep) but

unfortunately at high

temperatures the blade can still

deform irreversibly due to a

process known as diffusional

creep.

To vastly reduce this creep

problem (which is basically

caused by the small, random

Overall the EJ200 employs a very low By-Pass Ratio (the

ratio of air which bypasses the core engine or compressor

stages) of 0.4:1 which gives it a near turbo-jet cycle. Such

a low BPR has the benefit of producing a cycle where the

maximum attainable non-afterburning thrust makes up a

greater percentage of total achievable output. At its

maximum dry thrust of 60kN (or 13,500lbf) the EJ200's

SFC is in the order of 23g/kN.s. With reheat the engine

delivers around 90-100kN (or 20,250-22,500lbf) of

thrust with an SFC of some 49g/kN.s. Compared to other

engines these figures may actually seem relatively high,

however such data must be used with caution and

evaluated with all other performance data to be of any

use. With reheat the engine weighs just 2286lb giving a

Thrust to Weight Ratio of around 9:1.

An interesting point to note is that the baseline production

engine is also capable of generating a further 15% dry

thrust (69kN or 15525lbf) and 5% reheat output (95kN

or 21263lbf) in a so called war setting. However utilising this capability will result in a reduced

life expectancy.

Much is currently being made about supercruise, that is the ability to cruise supersonically

without the use of reheat (afterburn) for extended periods of time. Although never stated

explicitly (as for example with the U.S. F-22) the Typhoon is capable of and has demonstrated

such an ability since early in its flight program according to all the Eurofighter partnets. Initial

comments indicated that, with a typical air to air combat load the aircraft was capable of

cruising at M1.2 at altitude (11000m/36000ft) without reheat and for extended periods. Later

information appeared to suggest this figure had increased to M1.3. However even more

recently EADS have stated a maximum upper limit of M1.5 is possible although the configuration

of the aircraft is not stated for this scenario (an essential factor in determining how useful such

a facility is). The ability to maintain transonic and supersonic flight regimes without resorting to

the use of reheat is achieved mainly thanks to the advanced materials and design of the EJ200.

For times when a quick sprint is required the Typhoon can employ reheat with an upper (design) limit of Mach 2.0.

EJ200 development

Since the first EJ200 ran in 1991 some 14 development engines have been

constructed. The first three plants were for design verification amassing some

~700 hours of bench test time. Another 11 engines were then constructed and

placed in Accelerated Simulated Mission Endurance Testing, or ASMET. These

prototypes (designated EJ200-O1A) were used to verify the engine design and

reliability. During this stage the first two Development Aircraft, DA1 and DA2

entered flight testing. Since the EJ200 had not been certified for flight these first

two aircraft were equipped with Tornado ADV class Turbo-Union RB199-104D

engines (the D signifies the removal of the thrust reversal buckets). These have

have a significantly lower dry thrust, some 42.5kN, than the EJ200 but are

approximately the same size. In mid-1998 these RB199's were replaced with

EJ200-03A models (see below).

So far over 10000 hours of combined rig testing have been achieved of which

some 2800 hours were in altitude testing facilities. In addition the EJ200 has

completed well over 650 real flights in the various Development Aircraft from sea

level to 15000m (50000ft) and from 135kts through M2.0.

The first Eurofighter to receive the (flight certified) EJ200-01A was the Italian DA3 in 1995 with

its first flight in June of that year. The remaining development aircraft also use the EJ200

powerplant (both the EJ200-01A and 01C), but the DA3 remains the primary engine integration

aircraft. The DA3 has been used for testing not only the EJ200 itself, but also the Full Authority

Digital Engine Control (FADEC) system and Auxiliary Power Unit (APU). In April 1997 EuroJet

completed and obtained flight certification on the full pre-production model engine, the EJ200-

03A.

In January 1998 EuroJet signed production and production investment contracts with NETMA for

some 1500 engines worth around DM12.5B covering the basic order of 620 Eurofighter's.

Following this in January 1999 EuroJet received official orders for the first 363 engines to equip

the 148 Tranche-1 Typhoon's as well as providing a number of spares. In June 1999 EuroJet

obtained flight clearance for the final production standard powerplant with production release

scheduled to occur by the end of 1999. The first two production engines were handed over to

BAE on the 12th July 2001 at Rolls Royce's Filton plant in Bristol. They were subsequently

integrated into IPA1, the first production Eurofighter, at BAE's Warton facility.

 Engine Management

The EJ200 is controlled by both an Engine Control Unit and a fuel management system supplied by LucasVarity Aerospace. The

primary engine control system developed by ENOSA and Technost SpA under the leadership of Dornier (an DASA company) and

Smiths Industries is a Full Authority Digital Engine Control, FADEC unit. Its responsibilities include overall engine management,

grain structure) a different

approach to manufacture is taken

in which giant single crystals are

grown. This is achieved by using

either a special furnace across

which a temperature gradient can

be applied or by slowly moving

the blade mould through the

furnace, a process called

Directional Solidification. The

resulting rotor is very resistant to

both diffusional creep and power

law creep and thus may be used

at higher operating temperatures.

Thus the efficiency of the jet

engine is improved and it

becomes possible to run the

engine under more extreme

conditions (such as cruising

supersonically) for longer.

Powder Metallurgy

Most metal structures and items

are constructed by milling a solid

piece of forged material. However

over recent years a different

method of forming has become

possible using powdered metals.

The major benefit of powder

metallurgy is that it can produce

a component with typically >95%

the density of a forged/machined

equivalent but at lower cost. The

actual procedure used is actually

quite straightforward. The metal

powder (or combination of

blended powders) is compressed

under high pressure into a mould

of the component to be produced.

It may also be possible to

incorporate special compounds in

the metal powder blend to

enhance for example,

temperature resistance or aid

lubrication. The mould containing

the compressed powder mix is

then be fired or sintered. Upon

cooling the mix crystallises to

form a solid component.

The result is a relatively cheap

component that can exhibit very

similar physical properties to

equivalent machined/forged items

but at lower cost.

EJ200 Specification

Length, m (ft,in) ~ 4.0 (~ 13'2")

Diameter, m (ft,in) ~ 0.85 (~ 2'9")

Dry Thrust, kN (lbf) >60 (>13500)

with Reheat, kN (lbf) >90 (>20250)

By-Pass Ratio 0.4:1

Total Pressure ratio 25:1

Fan Pressure ratio 4.2:1

Air mass flow, kg/s (lb/s) 77 (170)

SFC dry, g/kN.s 23

SFC reheat, g/kN.s 49

Production model EJ200 © Rolls Royce

including afterburn fuel flow and control of the nozzle area.

The third system comprises the Engine Monitoring Unit (EMU) developed by ENOSA, BAE Systems and Microtechnica. Its sole

purpose is the self diagnosis of engine faults, it thus augments the structural health monitoring system. The EMU takes its input from

both the fuel management and FADEC units as well as a full array of dedicated sensors within the engine.

 Future of the EJ200

Engine uprating

The EuroJet consortium were required to build an engine (often referred to as EJ2x0) which had at least a 20% growth potential.

There are already plans to carry out the necessary modifications to reach this higher (Stage-1) output in the 2000 to 2005

timeframe. Such an improvement will require a new Low Pressure Compressor (raising the pressure ratio to around 4.6) and an

upgraded fan (increasing flow by around 10%). This would result in the dry thrust increasing to some 72kN (or 16,200lbf ) with a

reheated output of around 103kN (or 23,100lbf). Given recent increases in the weight of the Typhoon it may not be unexpected to

find this upgrade performed in the near future.

More interestingly perhaps is Rolls-Royce and EuroJet's plan to increase the output 30% above the baseline specification as a

Stage-2 modification. Such an upgrade will require more substantial plantwide changes including a new LP compressor and turbine

and an improvement in the total pressure ratio. These upgrades would yield a new dry thrust of around 78kN (or 17,500lbf) with a

reheated output of around 120kN (or 27,000lbf). The indications are that these improvements will come on stream between 2005

and 2010, in time for the Typhoon's Mid Life Upgrade expected around 2016.

Thrust Vectoring Control

As well as the potential for increasing the EJ200's thrust there are also plans to incorporate a

Thrust Vectoring Control, or TVC nozzle.

The EJ200's TVC nozzle is a joint project lead by Spain's ITP and involving Germany's MTU.

Preliminary design of the system began in mid-1995 at ITP, the proceeding years involved work

by both ITP and MTU to deliver a fully functional EJ200 integrated system. The outcome of this

research led to the first 3DTVC equipped EJ200 undergoing rig trials in July 1998. The nozzle

requires relatively few modifications or additions to be made to the EJ200; a new hydraulic

pump, reheat liner attachment upgrades, casing reinforcement, new actuators and associated

feed equipment. More importantly the equipment fits within the engines current installation

envelope and therefore no changes will need to be made to the Typhoon to accommodate the

system.

There are essentially three types of vectoring nozzle; ones in which the entire post-turbine

section is moved, those which feature external nozzle attachments for directing thrust (e.g. the

X-31 paddles) or ones in which thrust is vectored within the divergent section. The ITP system

uses the later design requiring no external equipment (which adds weight and offers relatively

poor efficiency) and reducing distortion on the major engine structures (a problem with using

the first method).

The new Thrust Vectoring Nozzle, TVN is a

convergent/divergent type containing three

concentric rings linked via four pins forming a

unified Cardan joint. Each of these rings serves a

purpose, the inner ring is connected to the nozzle

throat area with the secondary ring forming a

cross-joint connection with the pivoting outer

ring. This outer ring is in turn connected to the

divergent section (green on the CAD diagram) via

several struts or reaction bars (black on the CAD

diagram to the left). The outer ring is controlled

by either three or four hydraulically powered

actuators situated at the North, South, East,

West, South West and South East positions. By minimising the number of required

actuators (either three or four) ITP claim there is little additional weight, reduced

actuator power demands and increased reliability over previous systems. Additionally

the nozzle utilises a partial balance-beam effect to minimise the actuator load

requirement. This effect uses the exhaust gases themselves to close the nozzle throat area, according to ITP this gives a 15%

reduction in actuator loads in certain circumstances.

The baseline vectoring configuration uses three actuators (North, South East and South West). By

moving each actuator either in or out the outer ring (red) can be tilted in any direction (see CAD

diagram to right, top picture) thus offering both pitch and yaw control. Any net directional movement

in the outer ring is then translated via the struts into a larger movement of the divergent section,

vectoring the thrust. As well as vectoring control (via movement of each actuator) it is possible to

alter the throat area directly by moving all three actuators outward or inward in parallel. In both

cases the outer pivot and the inner (green) throat area ring are fixed in the axial direction which

reduces the required number of actuators.

Beyond the baseline case the TVN includes a pro-baseline configuration offering the ability to alter the

divergent section exit area as well as vectoring thrust and altering the throat area. To achieve this

the outer ring is split into top and bottom halves and four actuators (in the N, E, S and W positions)

are utilised (see CAD diagram to right, bottom picture). By moving each actuator in a

Thrust Vectoring

Thrust Vectoring Control became

a big issue in the 1980's and

early 1990's with the majority of

aerospace companies pushing

ahead with related programs.

There are two basic types of TVC,

2D and 3D. Two dimensional

vectoring (2DTVC) works by

directing the exhaust up or down

(pitch vectoring), the F-22 Raptor

features such a system. While

three dimensional vectoring

(3DTVC) adds the ability to direct

the thrust left to right (yaw

vectoring).

The are several real benefits to

employing thrust vectoring, for

example; decreased take-off and

landing distances, higher

achievable angles of attack,

improved control at low

speeds/altitudes, reduction in size

and number of control surfaces

and reduced supersonic drag (by

using the vectoring equipment to

adjust trim rather than the control

surfaces). There are however

questions over just how useful

TVC will be in future air battles

with the increasing move towards

beyond visual range

engagements.

CAD image of nozzle during vectoring © ITP R&D

MY
Highlight
MY
Highlight
MY
Highlight
MY
Highlight
MY
Underline
MY
Underline
MY
Line

unified/combined manner the thrust can be vectored and the throat area altered. However by moving

just the N and S actuators the split ring hinge can be opened and closed. In turn this moves the

upper and lower strut series either in or out opening or closing the exit area. In a traditional Con-Di

nozzle the exit area is directly related to the throat area. The problem with this approach is that it is

extremely difficult to optimise the nozzle shape to different flight profiles, e.g. subsonic cruise,

supersonic dash. By allowing dynamic control of the exit area the nozzle shape can be altered on the

fly. According to ITP this allows for significant improvements in achievable thrust in all flight profiles.

The three ring system is not the only unique feature of the nozzle. In previous convergent/divergent

systems the reaction bars or struts have been connected to the divergent section at a single point.

This limits their deflection range thus imposing limits on achievable thrust vectoring (typically to no

more than 20°). The ITP TVN however uses a dual point hinged connection allowing a far greater 

range of movement to be achieved (according to ITP, studies indicate 30°+ can be achieved). By 

careful placement of the struts, problems with the nozzle petals overlapping or colliding are also

removed.

Since rig trials commenced in 1998 the TVC equipped EJ200-01A has run for 80

hours (February 2000) of which 15 hours were at full reheat (including sustained

five minute burns) during 85 runs. These trials have included over 6700 vectoring

movements at the most severe throttle setting and 600 throttling cycles under the

most demanding vectoring conditions. These trials demonstrated full, 360° 

deflection angles of 23.5° with a slew rate (the rate at which the nozzle can be

directed) of 110°/s and a side force generation of some 20kN (equal to

approximately to one third of the total EJ200 baseline output). These vectoring

trials have included both programmed ramp movements and active joystick control.

The studies have also verified the MTU developed DECU (Digital Engine Control

Unit) software and FCS connections.

During the summer of 2000 a round of altitude trials commenced at the University of

Stuttgart, Germany. These are focused on determining the effects of temperature

and pressure variation on the nozzle materials, shape and performance.

Additionally ITP are continuing work on further reducing the weight of the system.

In November 2000 ITP announced that an agreement had been reached with

Germany and the U.S. to utilise the X-31 VECTOR test aircraft for flight trials of the

nozzle. This will see a modified EJ200/TVN combination fitted to the X-31. The

modification work required will involve all members of the EuroJet consortium.

Additional input is likely from EADS and Boeing as well as NETMA in providing the

required EJ200's and equipping the X-31. The Spanish government has agreed to

pay for flight certification of the system and provide test pilots. The first flight trials

are expected in late 2002 to early 2003. In addition Eurofighter and EuroJet have

expressed a desire to commence flight trials of DA1 equipped with the nozzle sometime from 2003. How this fits in with the X-31 test

phase is currently unclear.

ITP have suggested that a Eurofighter fitted with the nozzle will benefit in a number of areas including; reduced after body drag

(through tighter nozzle shape control), an estimated 7% improvement in installed thrust for the supersonic cruise regime (M1.2 non-

reheat at 35000ft) and a 2% improvement in maximum take-off thrust.

At this stage there are no definite plans to fit the nozzle to any production Eurofighter. However Eurofighter, EuroJet and a number

of consortium nations and other companies have indicated a desire to include the nozzle (if possible) in Tranche-3 aircraft (due from

2010). This would fit with the stated desire of the four consortium nations to incorporate new technologies in sucessive Eurofighter

production runs. The current Eurofighter struture has already been strengthened in anticipation of increased loads created by TVC

as well as higher output EJ2x0 series powerplants.

The webmasters would like to express their thanks to ITP R&D for providing the information and material on the 3DTVC nozzle and Rolls Royce

for material on the XG-40 and EJ200.

Sources :

[1] : Rolls-Royce Online

[2] : Rolls-Royce, Press Office, Derby

[3] : EuroJet GmbH, Public Relations

[4] : Janes All the Worlds Aircraft 1996/97

[5] : Janes Avionics 96/97

[6] : Eurofighter 2000, Hugh Harkins, Key Publishing, 1997

[7] : Defence Data On-Line

[8] : DASA technical paper, Military Technology, December 1997

[9] : Engineering Materials 1, M. F. Ashby and D. R. H. Jones, Pergamon Press

[10] : Industria de Turbo Propulsores S.A., Spain

[11] : Smiths Industries plc.

[12] : Flight International, 16-22 June 1999

[13] : Bill Sweetman, World Air Power Journal, 38

[14] : EADS NV, Military Aircraft Division, Munich, Germany

Vectoring configurations © ITP R&D

Click either image for alternative versions

Rig trials of 3DTVC equipped EJ200 © ITP R&D

Cockpit | Defences | Flight Systems | Sensors | Structure | Weapons

History | News | Production | Squadron service | Main Index

All the source and data on this site is subject to copyright. Please read the disclaimer for more information

and contact the webmasters with any queries.

MY
Highlight