professor dave rugg -...

© Rolls-Royce plc 2010The information in this document is the property of Rolls-Royce plc and may not be copied or communicated to a third party, or used for any purpose other than that for which it is supplied without the express written consent of Rolls-Royce plc.This information is given in good faith based upon the latest information available to Rolls-Royce plc, no warranty or representation is given concerning such information, which must not be taken as establishing any contractual or other commitment binding upon Rolls-Royce plc or any of its subsidiary or associated companies.

Trends and Issues - Titanium Alloy use in Gas Turbines

Professor Dave RuggCorporate Specialist – Compressor and Nuclear Applications

Royal Society Industrial Fellow

2

Talk Structure History

Load regimes / material limitations Elastic regimes Elastic-plastic regimes Plastic regimes

The way forward Process / structure relationships Structure / property relationships

Conclusions

3

Engineering Challenge - The Gas Turbine

4Rolls-Royce Ti overview Ti Alloys account for 1/3 of gas turbine weight

Over 2000 tonnes per year used by RR >$100M PA

‘Conventional’ alloys Form a significant part of research

Improved component lifing Introduction via improved design and improved

manufacturing processes Seeking improvements in buy to fly ratio (near nett shape

technology) Academic interaction

Extensive collaboration with Academia Direct contract Via nationally funded programmes

5

Disc alloy temperature capabilityTemperatureCapability (C)

1960 1970 1980 1990 2000

350

600

α-β Alloys

High strength α-β Alloys

Near-α Alloys

Aerofoil ‘fire line’

6

Titanium in Discs – 3 shaft history

Increasing temperatureDecreasing space= Increased Ni alloy use

1980 1990 2000 2010

HP

IP

Ti685 Ti829 Ti834

Fan

7If titanium was your child - (global perspective)

Approx 50 years to mature

Half the typical growth rate (with respect to adoption through the engine)

Cost £2,000,000 per year whilst maturing(or £50,000 per mm of growth)

8Load regimes and material limitations

9Elastic properties (Texture)

•Fan blade untwist

-arise from centrifugal loads

-blades manufactured with ‘over-twist’

-running tip deflection > 15mm

-Tip angle controlled to < 1°

Shape = Mass flow = Thrust

Parallel rolling

Engine type (property differential L to T %) Trent xyz (1.4%)

Trent xyz (1.8%)Trent xyz (2.8%)

Trent xyz (3.5%)

RB211 derivative1 (8%)

10Elastic Properties - Resonance•Resonance

-blade design to minimise exposure and magnitude of HCF stress.

-design to 109 cycles is common.

-criteria based on stress threshold, not

cyclic life….

Modal frequency = HCF load = Product integrityIP compressor blade - high order resonantmode stress contour plot.Resonant frequency 18 KHz

11

Matls070289P

Hollow Fan Technology

Trent

MILITARY

Cross-Section ofHollow Blade

12Elastic – Plastic properties

Cold Dwell Fatigue

Notch Fatigue

Macrozone vs Effective Structural unit size…

13Plastic properties – the great divide ?

Strain rate

UTS(Quasi static)

UTS(High rate)

Low (Hours / days)

Medium(seconds)

High(milliseconds)

‘Yield’

‘Failure’

‘cyclic’yield

‘tensile’yield

Stre

ss

Notes;In all regimes strength level is controlled by degree of micro-structural refinement.Many alloys have similar cyclic yield strength.

Room temp props are dictated byDislocation creep….

relaxation

14Plastic Properties contd.

1970 1980 1990 2000

RB211-22B & 524

RB211-535E4

TRENT 700

TRENT 800

TRENT 500

TRENT 900

Medium Bird

1.5lb

30min Run-on

4lb Large Bird

Medium Bird

2.5lb

20min Run-on

4lb Large Bird

Medium Bird

2.5lb

20min Run-on

8lb Large Bird

Medium Bird

2.5lb

20min Run-on

8lb Large Bird

Medium Bird

2.5lb - 20min Run-on

8lb Large Bird, AND new

5.5lb Large Flocking Bird

20min Run-on

Medium Bird

1.5lb

5min Run-on

4lb Large Bird

Also consider containment,FOD, trailing blade integrity,Surface treatments (peening)etc.

15Titanium – other criteria…

Compressor blade fire(Temp / mass flow criteria)

Compressor disc oxidation(Temp limit – alpha case cracking)

Residual stress(section size /total stress consideration)

Adaibatic shear(structural unit size)

Disc / blade contact fatigue(crushing stress)

Repair(microstructural control)

16The way forward

17Experimental techniques and models now available… (Imperial,

Manchester,Oxford,Swansea)

0

200

400

600

800

1000

1200

0 2 4 6

Position (micron)

Stre

ss yy

(MPa

)

time=0.01stime=0.02s

F Dunne

A Wilkinson

M Bache

M Preuss, J Fonseca

D Dye

18Local Microtexture - cross rolled plate;

AX

19Understanding and modelling properties - overview

Measure Represent Model Validate Use / interface

SAWEBSDBeam line

What / how tomeasure ?

How accurate Do the measurementsHave to be ?

Poly crystalrepresentation

When does strain /Slip reversal becomeA crack ?

Databank ?

Length scales

Misorientation distribution

texture

Aspect ratioTopographyGB type /area fraction

CRSSStrain rateElastic props

Failure criteria Hydrostatic stress

Boundary conditionsFrom continuum FE

Coherent framework is required to allow adoption by the end user. A similar logic can be applied to modelling TMP.

20Structural variables – real world issues…

Grain Boundary Morphology

Forge temperature, strain and strain rate,

transfer time and media for post forge

cooling

Primary Alpha Laths

Transfer time and media on post forge cooling,

Solution heat treatment temperature and time

Prior Beta Grain

Billet* preheat, temperature, time and

ramp rate, transfer time to press, strain and strain rate during forging, press

time and hold periods

Secondary “Fine” Alpha

Transfer time and media for post solution heat treatment cooling, Ageing temperature and time

Retained Beta


* Billet itself sets starting bulk chemistry, initial partitioning, macro/microstructure and crystallography

Grain Boundary Morphology

Forge temperature, strain and strain rate,

transfer time and media for post forge

cooling

Primary Alpha Laths

Transfer time and media on post forge cooling,

Solution heat treatment temperature and time

Prior Beta Grain

Billet* preheat, temperature, time and

ramp rate, transfer time to press, strain and strain rate during forging, press

time and hold periods

Secondary “Fine” Alpha


Retained Beta


* Billet itself sets starting bulk chemistry, initial partitioning, macro/microstructure and crystallography

21Symmetry variables – real world issues

Slip transmission function of:

(i) Relative orientation of αΙ and β and β to αΙΙgiving favourable or unfavourable alignment of preferred slip systems.

(ii) Orientation angle, θ

(iii) Stress, either macro applied or local due to crystal plasticity.

(iv) Length scale and slip planarity

(v) Loading rate

(vi) Local chemistry, driven by alloy partitioning, changing CRSS for individual slip systems

σ

σ

Slip

αΙ

β

αΙΙ

θ






(v) Loading rate


σ

σ

Slip

αΙ

β

αΙΙ






(v) Loading rate


σ

σ

Slip

αΙ

β

αΙΙ

θ

E = 145

E = 100

Elastic and plastic asymmetry;from single crystal to real engineering(complex) structures….

22Understanding and modelling the evolution of texture, structure and properties.

Important for the material supplier AND end user; Process route optimisation (large gains possible) Assessing manufacturing change and concessions Design for ‘local area properties’ Advanced micro-structural standards are coming...

Possible options; Physics based Statistics based Neural network Phase field

In reality, a combination of techniques is probable..

23Conclusions. Some limits for titanium application defined;

Max temperature in discs Max temp in aerofoils

Significant gains still to be had; Process improvements Improved component lifing ‘new’ manufacturing practice and repair All require detailed fundamental understanding combined

with the appropriate modelling framework

‘Joined up’ programmes and extensive interaction with Academia and supply chains are critical in achieving these gains.

professor dave rugg -...

Documents