Active Turbine Tip Clearance Control Trade Space Analysis of an Advanced Geared Turbofan Engine

Active Turbine Tip Clearance

Control Trade Space Analysis of an

Advanced Geared Turbofan Engine

Jonathan Kratz (NASA GRC)

Jeffryes Chapman (NASA GRC)

AIAA Propulsion & Energy Forum

11 July 2018

7/10/2018 1 AIAA Propulsion & Energy Forum 2018

Outline

Summary • Illustration of a systematic approach for modeling high pressure turbine tip clearance • Application is to a conceptual N+3 geared turbofan with a compact gas turbine (CGT) • Presentation of a sensitivity study and a parametric study of actuators that provides preliminary requirement for implementing active turbine tip clearance control Outline • Background/Motivation • Modeling • Control Approach • Sensitivity Studies • Parametric Actuator Studies • Conclusions 7/10/2018 2 AIAA Propulsion & Energy Forum 2018

Background

Image is from: Lattime , S.B., and Steinetz , B.M., “Turbine engine clearance control systems: current practices and future directions,” American Control Conference, 2001.

• Tip clearance, especially in the high pressure turbine, can lead to inefficiencies • Poor sealing can lead to flow leakages that: • Reduce work extraction and thus efficiency • Reduce flow capacity • Produces tip vortices that can lead to increased noise • Makes the turbine work harder to spin the shaft  requires more fuel and higher • Performance sensitivity to turbine tip operating temperatures clearance is expected to be greater for CGTs • For a large gas turbine: • Tip clearance gap varies through the course • 10 mil of clearance  1% of a flight and is impacted by: difference in turbine efficiency & 10 ˚C in exhaust • Degradation gas temperature (impacts • Mechanical loads time on wing) • Thermal loads 7/10/2018 AIAA Propulsion & Energy Forum 2018 3

Background

• Current approach to turbine tip

clearance control

• Cooling air is blown over the casing around the turbine to contract the casing and close down the clearance, particularly during cruise • Can only actively effect the tip clearance by closing it • Scheduled control logic • No direct tip clearance measurement and limited feedback

• Proposed approach – Higher Bandwidth - Active Turbine Tip

Clearance Control

• Could be a variety of actuator types and system designs • Use of an actively controlled “faster” actuation system to maintain tighter clearance • Can effect tip clearance by both opening and closing the gap • Utilizes a direct tip clearance measurement or reliable estimator Image is from: Soghe , R.D., Andreini , A. “Numerical Characterization of Pressure Drop Across the Manifold of Turbine Casing Cooling System”, ASME Journal of Turbomachinery , Vol. 135, Issue 3 (2013) 7/10/2018 AIAA Propulsion & Energy Forum 2018 4

Modeling

• System level physics - based approach • Considers axisymmetric tip clearance variations due to centrifugal and thermal loads • Neglects asymmetric variations that are much less significant • Models the following components: • Blade • Rotor Disc • Shroud/Case • Thermal expansion calculations use a 1 - D linear elastic expansion model that utilizes the average temperature of the component • Rotor Disc & Shroud/Case – 1 - D finite difference method • Blade – Lumped capacitance model • Centrifugal expansion of the rotor disc & blade • Modeled with algebraic equations under assumptions of simplified geometries • Temperature dependent thermal & mechanical properties are considered • A MATLAB/Simulink Library, the Tip Clearance Modeling Library (TCML) was created to implement this approach in parameterized manner • Download available at https:// github.com/nasa/TCML 7/10/2018 AIAA Propulsion & Energy Forum 2018 5

Control Approach

• Baseline control approach • Implements cooling on the outer engine casing using bleed air from the bypass duct • A bleed schedule was designed to maintain a prescribed tip clearance that provide acceptable margin for transients • The schedule utilizes the existing sensor suite • HB - ATTCC • The actuator model is generic • A tip clearance sensor is assumed with perfect accuracy and instantaneous dynamics • The set - point tip clearance is a function of shaft speed (less ability to expand at high speed) 7/10/2018 AIAA Propulsion & Energy Forum 2018 6

Modeling

• The tip clearance model was integrated with the AGTF30 engine model 7/10/2018 AIAA Propulsion & Energy Forum 2018 7

Baseline Models

Cruise Decel Descent Idle Climb Go - around Min. Tip Clearance Limit Re - accel Take - off Accel - cruise to full power • 2 different turbine configurations were considered, distinguished by the blade material • Inconel • 3 - D Angle Interlock SiC / SiC Hybrid Matrix Ceramic Matrix Composite (3DSiC) • Flight profile contains various potential worst case scenarios • The “stationary cold” tip clearance was adjusted in each case to enforce a minimum tip clearance of 25 mil (accounts for axisymmetric variation and other uncertainties/conservativeness that comes with the traditional control approach) 7/10/2018 AIAA Propulsion & Energy Forum 2018 8

Sensitivity Studies

TSFC • AGTF30 was run at the steady - state cruise condition and various tip clearances were enforced • ~5% change in turbine efficiency for a 30mil variation in tip clearance (~1.67% difference for every 10 mil) • Inter - turbine temperature (T ), fuel flow rate ( Wf ), and thrust specific fuel consumption (TSFC) are the most impacted parameters 7/10/2018 AIAA Propulsion & Energy Forum 2018 9

Sensitivity Studies

Material Parameter Inconel 718 3DSiC Units Density 15.89 4.7725 slug/ft - 5 - 5 Thermal expansion coefficient 1.05x10 0.144 x10 1/˚R Poisson’s ratio 0.4093 0.17 N/A Specific heat 5.47 8.73 Btu/(slug - ˚R) 9 9 2 Modulus of elasticity 1.19 x10 3.17 x10 lb /ft f • 3DSiC blades lower density and higher 3DSiC modulus of elasticity reduces its expansion considerably • However, it was found that the cooling schedule requires ~50% more bleed than the Inconel blade when paired with Inconel the Inconel casing 7/10/2018 AIAA Propulsion & Energy Forum 2018 10

Parametric Actuator Studies

• Evaluated bandwidth (BW), rate limit (RL), range (R), and deadband (DB) needs • Took a look at weight and force implication • Simulations utilized the flight profile below (R & DB) or an accel from the cruise point to full throttle (BW & RL studies) • For each set of actuator Idle Climb Cruise Decel Descent parameters, the tip clearance set - point schedule was adjusted such that the minimum tip clearance observed in the simulation was ~15mil (0.09% of the blade span) TC response Go - around 15 mil Take - off Re - accel Accel - cruise to full power 7/10/2018 AIAA Propulsion & Energy Forum 2018 11

Parametric Actuator Studies

Inconel Blade Configuration

• Large benefit

just from being

able to actively

control

• Relevant

bandwidth

range:

• Inconel: 0.01 –

1.5 rad/sec

• 3DSiC: 0.01 – 2

rad/sec

7/10/2018 AIAA Propulsion & Energy Forum 2018 12

Parametric Actuator Studies

Inconel Blade Configuration • Rate limit did not have much impact for the 3DSiC blade configuration • The entire range of tested rate limits appears to be applicable 7/10/2018 AIAA Propulsion & Energy Forum 2018 13

Parametric Actuator Studies

• Range: ~45mils

(Inconel), ~30mils

(3DSiC)

• For BW = 0.01 rad/sec

the tip clearance

reduces during the

deceleration period –

potential risk for

Inconel Blade Configuration

rubbing during a

subsequent

acceleration

• Re - evaluate speed

requirements: BW ≥

0.05 rad/sec & RL ≥

0.45 mil/sec

7/10/2018 AIAA Propulsion & Energy Forum 2018 14

Parametric Actuator Studies

• Solved the Breguet range equation to determine the amount of weight that could be added to the engine system on account of the actuation system for various flight ranges and fuel savings benefits Inconel Blade Configuration 7/10/2018 AIAA Propulsion & Energy Forum 2018 15

Conclusions

• HB - ATTCC would be advantageous for both of the evaluated configurations • Significant efficiency benefits could be enabled through implementing active control to both open and close the tip clearance and does NOT necessary need to be “fast” • Only needs to be fast enough to retreat to its set - point such that a re - accel pinch point will not violate the acceptable tip clearance margin • Use of the 3DSiC (CMC) blade reduced the tip clearance variation but with traditional control methods required significantly more bleed (~50% more) to cool and contract the casing • Observations from the sensitivity analysis: • Turbine efficiency changes by ~1.67% & fuel consumption by ~0.67% for every 10 mil of tip clearance • Largest impact is on turbine operating temperatures, fuel flow and TSFC • Young’s modulus & density of the blade play a significant role in the natural tip clearance variation • Effectiveness of cooling flow is limited and has diminishing effects as more flow is bled for cooling • Recommendations based on the parametric actuator study: • Bandwidth: Inconel – 0.05 - 1.5 rad/sec, 3DSiC – 0.05 - 2 rad/sec • Rate Limit: Inconel – 0.45 mil/sec, 3DSiC – 0.45 mil/sec • Range: Inconel – 45 - 50 mil (2.5 - 2.8% span), 3DSiC – 30 - 50 mil (1.7 - 2.8% span) • Deadband : Inconel – < 2 mil, 3DSiC - < 0.5 mil • Other observations: • Can increase η by 3.5 - 5.5%, reduce T by 10 - 25 ˚R, and reduce TSFC by 1.25 - 2.2% • Can save ~10 - 120 gal of fuel per flight compared to the baseline engine 7/10/2018 AIAA Propulsion & Energy Forum 2018 16

Acknowledgements

• This work supports the objectives and goals of

NASA’s Advanced Air Transportation Technology

(AATT) Project funded by the Aeronautics Research

Mission Directorate (ARMD).

• NASA civil servants & contractors who contributed

in some way to this effort: Sanjay Garg, Joe Grady,

Ram Bhatt, Jerry Lang, Vikram Shyam, Paht

Juangphanich , Dennis Culley, Joe Saus

7/10/2018 AIAA Propulsion & Energy Forum 2018 17

Questions?

Contact Information:

• Jonathan Kratz – jonathan.kratz@nasa.gov

• Jeffryes Chapman – jeffryes.w.chapman@nasa.gov

TCML Link:

https:// github.com/nasa/TCML

7/10/2018 AIAA Propulsion & Energy Forum 2018 18

EXTRA SLIDES

7/10/2018 AIAA Propulsion & Energy Forum 2018 19

The Engine

Advanced Geared Turbofan

30,000lb (AGTF30)

f

• Developed using the

Toolbox for Modeling and

Analysis of

Thermodynamic Systems

• Capable of producing

(T - MATS)

30,000lb of thrust at the

f

• Based on the NASA N+3

sea - level static condition

NPSS reference engine

• Includes a realistically

• Features a CGT and a

variable area fan nozzle

performing full - flight

envelop controller

7/10/2018 AIAA Propulsion & Energy Forum 2018 20

Parametric Actuator Studies

3DSiC Blade Configuration Inconel Blade Configuration

• Recommendations

• Inconel: DB < 2 mil

• 3DSiC DB < 0.5 mil

7/10/2018 AIAA Propulsion & Energy Forum 2018 21

Parametric Actuator Studies

3DSiC Blade Configuration 7/10/2018 AIAA Propulsion & Energy Forum 2018 22

Parametric Actuator Studies

• Pressure forces are expected to be dominant • Exact pressure forces may be dependent on design • Looked at pressure differentials ( ∆P ) for 2 basic configurations: (1) modulating the shroud and (2) modulating the shroud/casing assembly in the direction of the net pressure differential across the modulated object Cowl Cavity Case • For segmented actuators this suggests forces of several hundred to a few thousand pounds • Need to be able to respond to rapid changes in applied force • Need to overcome additional forces such as static friction, weight of the actuator, and inertial forces + produce excess force to meet dynamic requirements Core Gas Path 7/10/2018 AIAA Propulsion & Energy Forum 2018 23