laser diagnostics for high-pressure gas turbine combustion€¦ · · 2011-09-14ignition,...
TRANSCRIPT
Institute of Combustion Technology
Laser Diagnostics for High-Pressure Gas Turbine Combustion
Wolfgang Meier
German Aerospace Center (DLR), Institute of Combustion Technology, Stuttgart, Germany
Workshop on Techniques for High-Pressure CombustionArgonne National Laboratory, August 29 - September 1, 2011
30.8.2011
Slide 2 Argonne Workshop 2011
Outline
- Introduction to DLR Institute of Combustion Technology
- Measurement techniques and test rigs
- Problems encountered in high-pressure combustion
- Examples and results
- Summary and conclusions
30.8.2011
Slide 3 Argonne Workshop 2011
Gas Turbine Combustion at DLR Stuttgart
Goals and research fields
• Reduction of Pollutants:NOx, soot, UHC
• Instationary combustion phenomena:Ignition, re-ignition, extinction, thermo-acoustic
• Fuel-flexibility:Kerosene, natural gas of different quality, syngases, alternative fuels
• Burners and combustion systems:FLOX®, hybrid power plant
30.8.2011
Slide 4 Argonne Workshop 2011
Laser Measuring Techniques for Different Quantities
Flow field velocity
Species (OH, CH, NO, CH2O, tracer), temperature, structures
Major species, mixing, temp..
Temperature
Soot
Atoms, Ions
Quantity Technique
Particle Image Velocimetry (PIV)
Laser induced fluorescence (LIF)
Laser Raman Scattering
Coherent anti-Stokes Raman scattering (CARS)
Laser induced incandescence (LII)
Laser Induced BreakdownSpectroscopy (LIBS)
Dim.
2D
2D
0D or 1D
0D
2D
0D
All methods applied as single-pulse techniques with repetition rates up to 10 kHz
30.8.2011
Slide 5 Argonne Workshop 2011
Mobile Laser Systems
Raman-LasersCARS-Lasersystem
- For measurement campaigns at test rigs, all over Europe.- Installation in containers: temperature stabilized, compact.
30.8.2011
Slide 7 Argonne Workshop 2011
High-Pressure Combustion Test Rig at Stuttgart
Air supply:Pressure: max. 40 barFlow rate: max. 1.3 kg/sInlet temperature: max. 1000 K
Gaseous and liquids fuels
Thermal power: max. 2 MW
Not traversable
optical measurement section
supply section
30.8.2011
Slide 8 Argonne Workshop 2011
High-Pressure Combustion Test Rig – Stuttgart
Typical combustion chamber with optical access
Example of a low-emission flame
30.8.2011
Slide 9 Argonne Workshop 2011
Problems Encountered in High-Pressure Combustion
- Limited optical access, laser beams and signals must pass several windows.
- Window degradation by high thermal loads.- Line broadening → signal reduction in LIF.- Signal trapping (LIF, chemiluminescence imaging).- Beam degradation and steering (CARS, Raman, LII).- Depolarization in windows by stress-induced birefringence. - Non-traversable combustors → measurement technique must be
traversed.- Costs.
30.8.2011
Slide 10 Argonne Workshop 2011
Influence of Heat Load on Quartz Windows
Quartz window after extensive use at high pressure
30.8.2011
Slide 11 Argonne Workshop 2011
Window Cooling in Optical Combustion Chamber
plane quartz plates
Flow of cooling air
Exhaust165 mm
275 mm
30.8.2011
Slide 12 Argonne Workshop 2011
LIF: Effect of Line Broadening- NO was seeded as a tracer to the fuel or the combustion gases.- LIF of NO A-X ( bands at λ~226 nm) at 15 bar, T~1200 K.
225,48 225,50 225,52 225,54 225,56 225,58 225,60 225,620
10
20
30
40
50
60
70
80
90
100
NO O2 * 6000
inte
nsity
(a.u
.)
wavelength (nm) (vacuum)
Q1(23)
- Lines not well resolved and broader than exciting laser radiation.- Overlapp with (hot) O2 spectrum (Schumann-Runge band).- Different tracer available?
Sadanandan et al., Appl. Phys B 2011DOI 10.1007/s00340-011-4655-4
30.8.2011
Slide 13 Argonne Workshop 2011
Absorption of Laser Radiation and LIF Signal
Excitation of OH in A-X (1,0) band in CH4/air flames at 1 atm.
Averaged LIF signal from turb. flame at T≈1900 K
Absorption of laser from left to right ≈45%
Excitation of Q1(8) at λ≈283nm
30.8.2011
Slide 14 Argonne Workshop 2011
Absorption of Laser Radiation and LIF Signal
Excitation of OH in A-X (1,0) band in CH4/air flames at 1 atm.
Averaged LIF signal from turb. flame at T≈1900 K
Absorption of laser from left to right ≈45%
306 308 310 312 314 316 318 3200
20
40
60
80
100(1,1)(0,0)
inte
nsity
(a.u
.)
wavelength (nm)
no abs. with abs.
Emission spectrum after laser excitation
Absorption path: 54 mm through exhaust gas at T≈2100 K.
Excitation of Q1(8) at λ≈283nm
Sadanandan et al., Appl. Phys B 2011accepted for publication
30.8.2011
Slide 15 Argonne Workshop 2011
1D Laser Raman Scattering at the High-Pressure Rig
Goals- Application to lean premixed GT combustors.- Simultaneous single-shot measurement of major species
concentrations, mixture fraction and temperature along a line.- Identification of effects of unmixedness and finite-rate chemistry.- Data for model validation.
Challenges- Introduce high-power laser pulses with small beam waist into
combustion chamber.- Move laser and detection optics synchronously to change
measurement position.- Perform calibration measurements inside pressure housing.- Cope with window contamination, alignment shifts, ….
30.8.2011
Slide 16 Argonne Workshop 2011
Laser system + pulse stretcher
Spektrographf = 300 mm
Beam expansion3,3 : 1
ICCDcamera
Cylinder lensesconvex
f = 300 mm
Convex lens
Focus inside thecombustion chamber
Apochromatf ≈ 280 mm
Notchfilter
532 nm
Power meter
Measurement Setup
3 double pulse Nd:YAG lasersλ = 532 nm, 6 x 0.3 J / Pulsrepetition rate: 10 Hz
Stopper et al., Exp Thermal FluidSci. 34, 396-403 (2010)
30.8.2011
Slide 17 Argonne Workshop 2011
Laser system + pulse stretcher
Spektrographf = 300 mm
Beam expansion3,3 : 1
ICCDcamera
Cylinder lensesconvex
f = 300 mm
Convex lens
Focus inside thecombustion chamber
Apochromatf ≈ 280 mm
Notchfilter
532 nm
Power meter
Measurement Setup
3 double pulse Nd:YAG lasersλ = 532 nm, 6 x 0.3 J / Pulsrepetition rate: 10 Hz
30.8.2011
Slide 18 Argonne Workshop 2011
Pulse Stretcher for Laser Pulses
8 ns 6 single pulsesPeak power: ~ 35 MW
1 long pulsePeak power: ~ 7 MW350 ns
Beam is splitted by beam splitters and recombined after having travelled different distances.
30.8.2011
Slide 19 Argonne Workshop 2011
ChallengeOptical breakdown in focus
Reduced with smallbeam divergence
Damage of the combustor windows
Requires largebeam divergence
large beam waist
large cross section on windows
30.8.2011
Slide 20 Argonne Workshop 2011
Our Approach: Generate Astigmatism
astigmatic fosus
Use of 2 cylindrical lenses
Beam waist about 0.5 mm
F1 F2
30.8.2011
Slide 21 Argonne Workshop 2011
Raman Measurements in Scaled GT Combustor- Lean premixed natural gas /air flames (p=4 bar, Ptherm.=1.2 MW).- Staged combustion: 1 pilot burner, 8 main burners.- One goal: Interaction between pilot and main burners.
pilot burner
main burners
Lückerath et al., Proc. ASME Turbo Expo 2011, GT2011-45790
30.8.2011
Slide 22 Argonne Workshop 2011
Scatterplot of Temperature vs. Mixture Fraction
- Symbols represents results from single-shot measurements.- Different states of mixing and reaction progress.
adiab.equil.
non-reacted stoich. f
30.8.2011
Slide 23 Argonne Workshop 2011
Flame Shape from Chemiluminescence
cone of pilot burner.
OH* chemiluminescence, Abel-transformed
flow frommain burners
region of mixing and flame stabilization
30.8.2011
Slide 25 Argonne Workshop 2011
Wish List for Raman Measurements
- Laser with ~1 μs pulse duration, pulse energy >1 J, good beam profile, wavelength close to 400 nm.
- Detection camera with high quantum efficiency, low noise, gateable.
- Window material with better heat resistance or improved cooling concept.
30.8.2011
Slide 26 Argonne Workshop 2011
Summary and Conclusions
- Laser measurements at high-pressure test rigs could be performed successfully.
- Window degradation is one of the most serious problems in our high-pressure combustion chambers.
- Effects of line broadening, beam steering, etc. reduce the quality of the results.
- Possible improvements: fs-CARS? Long pulse lasers for Raman?