ASME Turbo Expo 2009: Power for Force Land, Coefficients Sea, and Air GT2009-59072 Hybrid Brush Seal ROTORDYNAMIC FORCE COEFFICIENTS OF A HYBRID BRUSH SEAL: MEASUREMENTS AND PREDICTIONS Luis San Andres Jose Baker Mast-Childs Professor Texas A&M University Project Engineer KBR, Inc. Adolfo Delgado Mechanical Engineer GE Global Research Center ASME GT2009-59175 accepted for journal publication Supported by Siemens Power Generation GT2009-59072 Hybrid Brush Seal Force Coefficients Trends in High Performance Turbomachinery: - Higher speeds - Extreme operating temperatures and pressures Issues of Importance - Increase in secondary flows (parasitic leakage) - Increase in specific fuel consumption and COST - Reduction in power delivery - Potential for rotordynamic instability Improved Brush Seal Technology Offers: Higher engine performance with less parasitic leakage Improved engine stability and reduced engine vibration Lower operating and maintenance costs Justification GT2009-59072 Hybrid Brush Seal Force Coefficients Hybridseals Brush (HBS) Seals Brush vs.Seal Labyrinth Advantages Advantages Handle amplitudes vibration SBS All thelarge advantages of 1stofgeneration Less axial space Reduced leakage (~36%) over 1st generation shoed brush seal (90%) Reduced secondary flow leakage Pads connected via EDM-webs, no Disadvantages spot welds between pads and bristles Pressure differential limitation Higher axial stiffness, prevents pad Wear and local thermal distortion motions at high pressure differentials Shoed Brush Seal (SBS)

Advantages Incorporates metal shoes at the free end of the bristle Reduces and eliminates wear shoes lift off Bi-directional shaft rotations Disadvantages Pads roll over under high pressure differential GT2009-59072 Hybrid Brush Seal Force Coefficients LITERATURE REVIEW: Hybrid Brush Seals Chupp et al. (2006): comprehensive review on sealing technology, including brush seals and shoed brush seals. Justak introduces a film riding seal with hydrodynamic pad action. (U.S. Patent 7,182,345) Delgado and San Andrs (Sealing Technology, 2005) measure leakage and structural characteristics of a shoedbrush seal. San Andrs et al. (ASME GT2008-50532 ) measure leakage and power loss in a Hybrid Brush Seal built with interference. HBS has approximately 36% less leakage then a shoed brush seal. San Andrs & Ashton (2009 STLE Meeting) compare leakage performance of three seals at high temperature GT2009-59072 Hybrid Brush Seal Force Coefficients Experimental Facility Supply pressure inlet Supporting springs Rotor Structural parameters Kshaft= 243 lbf/in (42.5 kN/m) Ms+d= 9.8 lb (4.45 kg) Electromagnetic Shaker Stinger : 0.01 % (damping ratio) Eddy current sensosrs Installation: Pressure Vessel DC Motor Flexible coupling Quill shaft cm 10 6.550 diameter brush seal Max. air Pressure: 60 psig Shaker (20 lb max) High pressure air 20 30 40 Eddy current sensor

50 60 70 80 Roller bearing assembly 90 100 Flow Disk Seal Ps Ball bearing Pe Shaft Flow Flow Disc Spring Detail of brush seal test rig Flexible coupling to motor Hybrid brush seal GT2009-59072 Hybrid Brush Seal Force Coefficients Physical Properties Rotor diameter, Dj Brush seal (pads) inner diameter, Dsi Brush seal (retainer) outer diameter, Do Brush seal width, Bw Radial Interference between rotor and seal, Ri Number of pads Width of pads Bristle lay angle, Bristle modulus of elasticity, E Bristle density (circumference) SI unit 167.1 mm 166.4 mm 183.1 mm 8.53 mm 0.381 mm 20 7.23 mm 45 deg. 5 22.48x10 bar 850 bristle/cm English Unit 6.580 in 6.550 in 7.210 in 0.336 in 0.015 in 0.331 in 6 32.6x10 psi 2200 bristle/ in Bristle Bed Back Plate Pad or shoe * Close-up courtesy of Advanced Technologies Group, Inc.

High Pressure Side Spring Lever Mechanism Low Pressure Side Seal Courtesy of Advanced Technologies Group Ps Pe Flow Test Hybrid Brush Seal (HBS) Disc GT2009-59072 Hybrid Brush Seal Force Coefficients Identification of Rotordynamic Force Coefficients Electromagnetic Shaker Kyy, Cyy Eddy current sensors Kyx, Cyx Load Cell Y Y Kxx, Cxx X X Disk Stinger HBS Kxy, Cxy For rotor M x operation: M centered K K xx xy xx M M K yy y yx yx x C xx C xy x Fx Fix K yy y C yx C yy y 0 Fiy xy GT2009-59072 Hybrid Brush Seal Force Coefficients Identification of HBS rotordynamic force Coefficients Identification at excitation frequency () rotor speed () Z xxx Z xy y Fx System Impedances Z yxx Z yy y 0 Impedance Function Z K M iC , x , y

2 Z xxZ yy and Z xy Z yx GT2009-59072 Hybrid Brush Seal Force Coefficients Location of displacement measurements Gyroscopic effects negligible for test rotor speeds (600 and 1,200 rpm [20 Hz]) Disk 0.1 S h a ft R a d iu s , [m ] Shaft Radius [m] Shaft Radius, Shaft Radius,[m] [m] Effect of rotor speed on rotor-HBS natural frequency 0.075 0.05 0.025 0 -0.025 0.1 0.075 Roller bearing 0.05 support Shaft 0.025 0 -0.025 -0.05 -0.075 Location of 0.3 0.35 external force -0.1 -0.05 0 0.05 0.1 0.15 0.2 Hybrid Brush Axial Location, Seal location -0.075 0.25

[m] -0.1 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 Axial Location, [m][m] Axial Location, Axial Location, z [m] Rotor Speed [RPM] 1st Backward Nat. Frequency, [Hz] 1st Forward Nat. Frequency, [Hz] 2nd Forward Nat. Frequency, [Hz] Forward Nat. Frequency, [Hz] 0 30.5 30.5 146 1351 600 29.7 31.4 154 1351 1200 28.8 32.2 163 1351 3rd GT2009-59072 Hybrid Brush Seal Force Coefficients Cross coupling effects under rotation X - Displacement [um] X-displacement [m] 1000 X For load along X Synchronous with speed (1X)

direction, rotor (X) motions 32 Hz 750 22 N Due to excitation 500 >>> 250 0 Load=22 N 600 rpm 40 40 120 200 Frequency [Hz] 280 cross (Y) motions 360 440 Frequency [Hz] Y - Displacement [um] Y-displacement [m] 1000 Y 750 1X motions always small compared to excitation 22 N Synchronous with speed (1X) 32 Hz Due to excitation 500 250 0 40 40 120 200 Frequency [Hz] Frequency 280 [Hz] 360 440 GT2009-59072 Hybrid Brush Seal Force Coefficients Dynamic Stiffness vs. Frequency Load = 22 N, frequency 25-80Hz

Fx x Z xx 2 (x y2 ) 5 M K xx K- 2M xx 2 xx xx 0 TESTs Model 5 Model Test Data 600 rpm, 1200 rpm, 600 rpm, 1200 20 40 rpm, 0 5 105 5 5 10 1 106 6 1 10 0 20 40 60 Frequency [Hz] 80 5 105 5 5 10 5 105 0 0 20 40 60 Frequency [Hz] 80 0 100 Test Data Model 5 105 5 105 6 1 10

1 106 40 60 Frequency [Hz] 0 R e (Z x x 1 ) [N /m ] 0 20 Test Data Model 80 5 10 5 6 1 10 40 60 0 Frequ ency [Hz] 20 0 5 105 6 1 10 60 0 80 20 100 40 Freque ncy [Hz] Test Data Model Test Data Model 106 80 20 100 40 60 Frequen cy [Hz] 0 20 40 60 Frequency [Hz] 0 20 40 60 F re q ue ncy [H z] 100 80 100 Test Dat a Model 100 0 5 5 10 0 20

40 60 F req uen cy [H z ] Te st Data Mo de l 6 1 1 0 80 Test Data Model 5 105 1 10 6 5 5 1 0 100 0 5 105 1 0 80 5 5 10 0 Re(Zxx1) [N/m] Re(Zxx1) [N/m] 0 40 60 Frequency [Hz] 5 5 10 0 1 106 5 10 5 20 Test Dat a Model 5 5 10 R e (Z x x 1 ) [N /m ] R e ( Z x x 1 ) [ N / m ]R e ( Z x x 1 ) [ N / m ] 6 1 10 0 100 Test Data Model 5 5 10

80 10 0 80 100 Pr Pr Pr Pr = 1.7 = 1.7 = 2.4 =60 2.4 Test Data Model Frequency [Hz] Test Data Model Frequency [Hz] Model reproduces the measured real part of impedance. Little effect of pressurization 0 5 5 10 Re(Zxx1) [N/m] 5 5 10 R e (Z x x 1 ) [N /m ] R e (Z x x 1 ) [N /m ] 5 10 R e ( Z x x 1 ) [N /m ] R e(Z xx1) [N/m ] Re (Zxx) [kN/m] 5 10 80 100 GT2009-59072 Hybrid Brush Seal Force Coefficients Cross-coupled Stiffness vs. Frequency Load = 22 N, frequency 20-80Hz 4 Z xy Z yx Z xx 4 4 10 4 10 Pr = 1.7, Zxy =- Zyx Pr = 2.4, Zxy =- Zyx

1200 rpm 1200 rpm 4 2 10 1 04 Stiffness Kxy Test[kN/m], Data 2 10 4 4 10 4 0 20 40 60 80 R e (F /X ) [N /m ] F re qu ency [H z] Zx y =-Zy x ( Tes t Dat a) Zx y =-Zy x ( Model ) 4 Zx y =Zy x (Test Data) 4 10 Zx y =Zy x (Model) 2 10 4 0 2 10 4 4 10 4 0 2 0 40 F requ ency [Hz] 60 80 4 1 04 2 1 04 2 1 04 4 1 04 R e (F /X ) [N /m ] Zx y=- Zy x ( T e s t Da ta ) Zx y=- Zy x ( Mo d e l) Zx y=Zy x (Te st D ata ) Zx y=Zy x (Mo d e l) 4

1 04 2 1 04 Model 4 Mass [kg.], Mxy 4 10 0 2 1 04 4 1 04 0 20 40 60 80 F requ ency [Hz] Zxy=-Z yx (Test Data) Zxy=-Z yx (Mod el ) Zxy=Zy x( Test Data) Zxy=Zy x( Model) 0 0 20 40 60 20 40 Frequency [Hz] Zxy=-Zyx (Test Data) Zxy=-Zyx (Model) R e (F /X ) [N /m ] 600 2 1 04 2 1 0 4 4 1 0 4 0 0 20 40 60 80

Freq uen cy [ Hz ] 15 1200 600 rpm 4 1 04 Zxy =- Zyx (Tes t Da ta) Zxy =- Zyx (Mode l) Zxy =Zy x ( Tes t D ata) Zxy =Zy x ( Model ) 4 104 2 104 0 Test Data 2 10 4 4 10 4 0 20 40 F requ ency [Hz] 60 8 0 60 80 Zxy =-Zy x ( T e st Da t a ) Zxy =-Zy x ( Mod e l) Zxy =Zyx ( T e st D a ta ) Zxy =Zyx ( Mo d el) 4 1 0 4 2 1 0 4 4 1 04 2 1 04 2 1 04 4 1 04 0 0 10 4 4 10 4 20 40

F requ ency [Hz] Z xy=-Zyx (Test Data) Z xy=-Zyx (Model) Z xy=Zyx ( Test D a ta) Z xy=Zyx ( Model) 0 2 0 20 40 60 2.7 Model 6.6 80 F requ ency [Hz ] 80 F requ en cy [H z] Z xy =-Zy x (Test Data) Z xy =-Zy x (Model) Z xy =Zyx (Tes t Data) Z xy =Zyx (Model ) 0 4 2 10 8.8 0 1200 R e ( F /X ) [N /m ] 1 04 2 0 R e ( F /X ) [N /m ] 4 600 Pr=2.4 R e ( F /X ) [N / m ] 4 2 10 R e ( F /X ) [N / m ] 0 Rotor Speed [rpm], 600 rpm Re(Zxy) [N/m] Pr=1.7 Pressure ratio R e ( F / X ) [N /m ] Re(Zxy) [N/m]

4 2 10 60 0 4 80 4 10 0 0 Zxy =- Zyx (Tes t Da ta) Zxy =- Zyx (Mo del ) Zxy =Zy x ( Test Data) Zxy =Zy x ( Model ) 20 0 40 Frequency [Hz] 60 0 80 Zxy=-Zyx (Test Data) Zxy=-Zyx (Model) Identified cross-coupled mass is nearly 0 kg. Identified cross-coupled stiffness (Kxx = -Kyx) is estimated as a constant independent of excitation frequency. Kxy doubles as rotor speed increases from 600 and 1,200 rpm for both pressure conditions y x GT2009-59072 Hybrid Brush Seal Force Coefficients Equivalent Viscous Damping (Cxx~Ceq) vs. Frequency 600 rpm Damping decreases with frequency, with little effect of supply pressure. Minimum value at test system natural frequency (~32 Hz) 1200 rpm Pr=1.7 Pr=2.4 Supply Pressure/Exhaust Pressure GT2009-59072 Hybrid Brush Seal Force Coefficients Force coefficiensts: system & seal Load = 22 N, frequency 25-80Hz No rotation Pressure ratio Pr = 1.0 Rotor Speed [rpm], 0

Stiffness [kN/m], Kxx 120 Mass [kg.], Mxx R2 (correlation factor) Dynamic stiffness (Kxx Mxx2) HBS stiffness[kN/m], HBS Dry Friction coefficient, HBS Loss Factor coefficient, Tests with rotor spinning Ks Pr=1.7 Pr=2.4 600 1200 600 1200 108 98 130 124 2.11 2.62 2.54 2.43 2.39 0.97 0.98 0.98 0.98 0.98 103 89 135 128 0.39 0.36 0.37 0.38 0.29 0.26 0.33 0.34 93 0.66 0.42

GT2009-59072 Hybrid Brush Seal Force Coefficients HBS predicted stiffness vs. frequency Stiffness Coefficients [kN/m] Stiffness Coefficients [kN/m] Frequency 25-100 Hz 140 130 130 130 120 120 120 110 110 Ksxx = Ksyy, Pr = 1.0 Ksxx = Ksyy, Pr = 1.7 Ksxx = Ksyy, Pr = 2.4 Ksxx = Ksyy, 1,200 rpmPr = 1.0 Ksxx = Ksyy, Pr = 1.7 Ksxx = Ksyy, Pr = 2.4 110 100 100 100 90 90 8090 80 7080 70 HBS Measured Stiffness, Ks 6070 60 00 60 0 Code: 20 Predicted HBS stiffness (Ksxx) drops slightly in range from 20100 Hz. Tests show nearly constant Ksxx rpm Ksxx 600 = Ksyy, Pr = 1.0 Ksxx = Ksyy, Pr = 1.7 Ksxx = Ksyy, Pr = 2.4 HBS = Measured Ks Ksxx Ksyy, Pr Stiffness, = 1.0 Ksxx = Ksyy, Pr = 1.7 Ksxx = Ksyy, Pr = 2.4 140 140 2020 4040

6060 8080 Frequency [Hz] Frequency [Hz] 20 40 60 80 XLTPGASBEAR 40 60 80 ASME GT2004-53614 100 100 100 120 120 120 100 120 Pressure (Pr = Ps/Pd) has negligible effect on seal direct stiffness, Ksxx GT2009-59072 Hybrid Brush Seal Force Coefficients HBS viscous damping coeff. vs. frequency Frequency 25-100 Hz 1 11 1 1 0.60.6 0.60.6 0.9 0.50.5 0.50.5 0.8 0.40.4 0.40.4 0.7 0.30.3 0.30.3 0.6 0.20.2 0.20.2 0.5 0.10.1 0.10.1 0.4 0 00 0 0.3 2020 0 00 0 20 20 0.2 0.1 0.1 0.1 0.10.4 0 4040 40 40 60 60 60 60 80 80 80 80 100 100 100

100 1 0.7 0.70.7 0.6 0.6 0.60.9 0.5 0.5 0.50.8 0.4 0.4 0.40.7 0.3 0.3 0.30.6 0.2 0.2 0.20.5 000.3 0 0 120 00 20 0 120 120 120 0.2 Frequency [Hz] 0.1 1,200 rpm Cxx = Cyy, = 0.25 Cxx = Cyy, = 0.55 Equivalent Viscous Damping, Ceq Cxx = Cyy, = 0.25 Cxx = Cyy, = 0.55 Equivalent Viscous Damping, Ceq Increasing Cxx loss factor= (0.25 ) = Cyy, Cxx = Cyy, = 0.55 Increasing loss factor Pr = (2.4 Equivalent Viscous Damping, Ceq Cxx = Cyy, = 0.25 Cxx = Cyy, = 0.55 Equivalent Viscous Damping, Ceq 0.8 0.80.8 0.2 1200 rpm Pr = 1.7 0.9 0.90.9 Damping Coefficients [kN-s/m] Cxx = Cyy, = 0.25 Cxx = Cyy, = 0.55 0.8 Equivalent Viscous Damping, Ceq Cxx = Cyy, = 0.25 0.7 Cxx = Cyy, = 0.55 Equivalent Damping, 0.6 Increasing lossViscous factor () Ceq

Cxx = Cyy, = 0.25 Cxx =Pr = 0.55 0.5 Increasing loss factor (Cyy, = 2.4 Equivalent Viscous Damping, Ceq Cxx = Cyy, = 0.25 0.4 Cxx = Cyy, = 0.55 Equivalent Viscous Damping,0.3 Ceq 0.80.8 0.80.8 Damping Coefficients [kN-s/m] 11 1 600 rpm 0.9 Pr = 1.7 0.90.9 0.90.9 0.70.7 0.70.7 1200 rpm1 20 2040 20 40 4060 40 60 6080 80 80 80100 120 100 100 100 Frequency [Hz] 0.1 0 0 0 20 40 60 80 100 0 120 20 40 60 80 HBS direct damping (Csxx) decreases with excitation frequency. Loss factor coefficient () represents physically seal structural (hysteresis) damping 100

1 GT2009-59072 Hybrid Brush Seal Force Coefficients Conclusions Prior to shaft rotation, seal pads lift-off due to hydrostatic effect from pressurization. Break-away torque is 90% less when seal is pressurized. Rotor speed has negligible effect on HBS drag torque (power loss) and leakage. A structural loss factor () and a dry friction coefficient() effectively characterize the energy dissipation mechanism of the HBS. HBS direct stiffness (Ksxx = Ksyy) decreases minimally with rotor increasing rotor speed for Pr = 1.7 and 2.4. HBS cross-coupled stiffness (Ksxy = -Ksyx) is one order of magnitude smaller than direct stiffnesses. HBS direct viscous damping coefficients decrease with increasing excitation frequency. GT2009-59072 Hybrid Brush Seal Force Coefficients Acknowledgments Thanks to Siemens Power Generation, Advanced Technologies Group (Mr. John Justak) www.advancedtg.com Learn more at http://phn.tamu.edu/TRIBGroup Questions ? GT2009-59072 Hybrid Brush Seal Force Coefficients HIGH TEMPERATURE SEAL FACILITY Properties Specific gas constant, R Supply pressure, Ps Inlet temperature, T Exhaust pressure, Pe Ambient temperature Magnitude 287 J/kg-K 101-760 kPa 298-573 K 101 kPa 298 K Voltage 240 V 90 V Heater Motor Flow in (supply pressure) Maximum Power Output 12 kW 300C 850 W 3,000 rpm 1 3 6 3 Flow out (ambient pressure) Pe =101 kPa 8 2 7 4 0 10 20

cm 1 2 3 4 Hot air inlet Pressurized cylinder & shaft Radial support bearings Disc and test seal location 5 6 7 8 Optical displacement sensor Centering mechanism Coupling and quill shaft Electric drive motor 2009 STLE Annual Meeting, May 2009 GT2009-59072 Hybrid Brush Seal Force Coefficients Compare Leakage from Three Seals Flow factor m T Ps D 35 40 30 30 25 25 20 20 15 15 10 10 5 5 0 0 Flow Factor [kg-K 0.5/(MPa-m-s)] 40 1.0 Air temperature and rotor speed affect little the flow factor. Show comparisons at max conditions 35 1.5 HBS has lower flow factor than both the labyrinth seal (38%) and the brush seal (61%) Labyrinth Seal Brush Seal Hybrid Brush Seal

1.02.0 1.52.5 2.03.0 Pressure Ratio [Ps/Pe] 2.53.5 3.04.0 3.5 The brush seal and HBS begin with same clearance. HBS 4.0 is more effective in reducing leakage Max. air temperature (300C) & rotor speed (3 krpm) 2009 STLE Annual Meeting, May 2009 GT2009-59072 Hybrid Brush Seal Force Coefficients Backup slides GT2009-59072 Hybrid Brush Seal Force Coefficients HBS Pad Lift Off upon Pressurization Hybrid brush seal profile section HP Front plate LP Back plate Back plate 16.674 mm Bristle pack Cantilever beam Pad profile 0.254 mm 4.445 mm 7.750 mm ROTOR Ps Pd ASME Journal of Engineering for Gas Turbines and Power, 2009, 131(1), pp. 012505 GT2009-59072 Hybrid Brush Seal Force Coefficients HBS Leakage (No Shaft Rotation) Supply Pressure = 5 to 30 psig 30.0 30.0 30.0 MassLeakage Flow Rate[g/s] [g/s] 25.0 25.0 25.0 20.0 20.0 20.0 SBS 15.0 15.0

15.0 HBS 10.0 10.0 10.0 5.0 5.0 5.0 0.0 0.0 0.0 1.0 1.0 1.0 1.5 1.5 1.5 2.0 2.0 2.0 2.5 2.5 2.5 Pressureratio, Ratio,Pr Pr Pressure 3.0 3.03.0 3.5 3.5 3.5 Measured HBS leakage is ~ 36% less than that for a 1st generation SBS over the test supply pressure range ASME Journal of Engineering for Gas Turbines and Power, 2009, 131(1), pp. 012505 GT2009-59072 Hybrid Brush Seal Force Coefficients Clearance [mm] Effective Clearance [mm] Calculated Effective Clearance Supply Pressure = 5 to 30 psig 0.08 0.08 0.08 0.07 0.07 0.07 0.06 0.06 0.06 0.05 0.05 0.05 Effective clearance (CE) as if one tooth laby seal SBS

0.04 0.04 0.04 HBS 0.03 0.03 0.03 0.02 0.02 0.02 0.01 0.01 0.01 0.00 0.00 1.0 1.0 1.0 . 1.5 1.5 1.5 2.0 2.5 2.0 2.5 2.0 2.5 Pressure Ratio, Pr Pressure ratio, Pr 3.0 3.0 3.0 3.5 3.5 3.5 m (T 273.15) cE Pu D ASME Journal of Engineering for Gas Turbines and Power, 2009, 131(1), pp. 012505 GT2009-59072 Hybrid Brush Seal Force Coefficients HBS Leakage vs. Pressure Drop 30.0 Static Condition 600 RPM 1300 RPM Leakage [g/s] Mass Flow Rate [g/s] 25.0 20.0 15.0 10.0 5.0 0.0 1.0 1.5 2.0 2.5 3.0 3.5 Supply Pressure = 5 to 30 psig Seal leakage is

proportional to pressure ratio (discharge/exit ). Little dependency on rotor speed. Pressure Ratio, Pr Pressure ratio, Pr ASME Journal of Engineering for Gas Turbines and Power, 2009, 131(1), pp. 012505 GT2009-59072 Hybrid Brush Seal Force Coefficients HBS Break-Away Torque vs. Pressure Ratio Seal Drag Torque [N-m] HBS Drag Torque [N-m] 10.0 Break-away Torque 8.0 As pressure increases from Pr = 1.0 to 1.7, the break-away torque (i.e. no rotation) 6.0 4.0 2.0 drops by ~ 75% 0.0 1 1.5 2 2.5 3 Pressure Ratio, Pr Pressure ratio, Pr Pads lift-off prior to shaft rotation due to the generation of a hydrostatic gas film as the pressure difference across the seal increases ASME Journal of Engineering for Gas Turbines and Power, 2009, 131(1), pp. 012505 GT2009-59072 Hybrid Brush Seal Force Coefficients HBS Power Loss vs. Speed & Feed Pressure 600 600 600 HBSHBS Power Loss Power Loss [W] [W] With external 500 500 500 Pr = 1.0 (No external pressurization Pr = 1.0 (No external pressurization) pressurization) 400 400 400 (Pr = 300

300 1.0 to 1.7), 300 Pr = 1.7 HBS Power 200 200 200 Pr = 1.7 Pr = 2.4 P + losses Pr = 2.4 100 100 100 decrease by ~ 0 0 0 500 1000 1500 90% over test 500 1000 1500 00 0 500 1000 1500 Rotational Speed [RPM] Rotor Speed [RPM] rotor speed range Pads lift-off due to the generation of hydrodynamic gas-film eliminating contact forces at the rotor/pads interface ASME Journal of Engineering for Gas Turbines and Power, 2009, 131(1), pp. 012505 GT2009-59072 Hybrid Brush Seal Force Coefficients Dynamic Load Tests (no shaft rotation) Fext x Ks ss L Lf Ls z mm f =244mm LLf =244 mm f =221mm LLf =221 L=248 248mm mm L= FFext ext Keq Meq Ceq x Equivalent Test System Meqx Keqx Ceqx Fext ASME Journal of Vibrations & Acoustics, 2007, 129, pp. 648-655 GT2009-59072 Hybrid Brush Seal Force Coefficients Parameter Identification (no shaft rotation) ASME Journal of Vibrations & Acoustics, 2007, 129, pp. 648-655 x xe it F Fext e

i t F Z ( K eq 2 M eq ) i Ceq x W Fext d x t Edis Ceq x 2 Harmonic force & displacements Impedance Function Work External 2 E dis eqK eq x 4 F x Viscous Dissipation DRY FRICTION & STRUCTURAL DAMPING GT2009-59072 Hybrid Brush Seal Force Coefficients HBS Structural Coefficients Load = 63 N, frequency: 20-100 Hz (no shaft rotation) Hybrid Brush Seal Pressure ratio* Stiffness [kN/m] Dry Friction coefficient, Loss Factor coefficient, Pr = 1.0 93 (5) 0.66 0.42 Pr = 1.7 130 (6) 0.51 0.40 Pr = 2.4 141 (7) 0.64 0.27 Pr = 3.0 141 (7) 0.69 0.22 *:atmospheric discharge pressure HBS stiffness increases slightly as supply pressure increases (~35% for pressure ratios: 1 to 3). Dry friction coefficient increases ~5 % due to increase in contact forces between seal components Loss factor (structural damping) decreases due to stiffening effect of increasing pressure differential across seal ASME Journal of Vibrations & Acoustics, 2007, 129, pp. 648-655 GT2009-59072 Hybrid Brush Seal Force Coefficients HBS Dynamic Stiffness vs. Frequency (no shaft rotation) Load = 63 N, frequency: 20-100Hz 400 2 M K eq K M eq 300

eq 200 eq 2 100 Re(F/X) Re(F/X)[N/m] [N/m] Re(F/X) [N/m] R e (F /X )[N /m ] 0 Model TESTs Re(F/X) [N/m] Model -300 Pr=2.4 Pr=2.4 Re(F/X) [N/m] -400 Pr = 1.7 Pr = 2.4 Pr = 3.0 Frequency [Hz] Frequency [Hz] Pr=1.7 Pr=1.7 Test Data Re(F/X) [N/m] -200 Fre quency [Hz] Pr=1.7 Pr=1.7 Re(F/X) [N/m] -100 Re(F/X) [N/m] (F/X) [kN/m] ReRe(F/x) [kN/m] Model reproduces real part of the impedance under the given supply pressure conditions. Freque nc y [Hz] Pr = Frequency 1.7[Hz] Pr = 2.4

Pr = 3.0 Fre quency [Hz] Pr=3.0 Pr=3.0 0 Pr=2.4 Pr=2.4 Pr=3.0 Pr=3.0 20 40 Frequency Frequency [Hz][Hz] Frequency [Hz] 60 80 100 Frequency [Hz] Frequency [Hz] ASME Journal of Vibrations & Acoustics, 2007, 129, pp. 648-655 120 GT2009-59072 Hybrid Brush Seal Force Coefficients HBS Equivalent Viscous Damping vs. Frequency (no shaft rotation) eq K eq 4 F Ceq x Load = 63 N, frequency 20-110Hz 10000 Equivalent damping increases slightly with pressure differential. Results typical of a system with dry-friction & material damping energy dissipation Test Test Data Data 80 0 0 60 0 0 Log(Ceq) (Ceq), [N-s/m] Log [N-s/m] 40 0 0 20 0 0 0 0 20 40 60

80 10 0 40 60 80 1 00 Pr = 1.7 8 00 0 6 00 0 4 00 0 8000 2 00 0 6000 0 0 20 Pr = 2.4 4000 2000 0 0 20 40 Pr = 3.0 60 80 100 P + 1000 100 0 20 40 60 Frequency, [Hz] Frequency 80 100 120