pressure activated leaf seal technology readiness testing

Cross Manufacturing Company (1938) Limited Brush Seals and Aerospace Products, South Site, Hopton Road, Devizes, Wiltshire, SN10 2EU

Pressure Activated Leaf Seal Technology Readiness Testing

GT2014-27046

Presented by Peter Crudgington - Director Cross Manufacturing Co. (1938) Ltd

in conjunction with CMG TECH


Acknowledgements

Co-authors: Aaron Bowsher, Engineering Manager, Cross Mnfg Clayton Grondahl, Director, CMG TECH James Dudley, Aircraft Engine Design Consultant Tracey Kirk, Test & Development Engineer, Cross Mnfg Andrew Pawlak, Senior Technician, Cross Mnfg The authors would like to thank the directors of Cross Manufacturing for the funding of this work and allowing publication. They would also like to thank the dedicated team of engineers and manufacturing staff at Cross who work together to produce products of outstanding quality

2


Content

• Pressure Activated Leaf Seal – PALS

• Smooth Rotor Testing

– Static

– Dynamic

• Acoustic Noise Investigation

• Simulated Shrouded Turbine Blade Testing

• Summary

• Future Work

3


Introduction PALS Design

4

Center of rotation.+

Stator - cut away to show leaf seal assembly.Backing ring - keyed in stator

Leaf Seal elements: - slotted ‘shim’ stock- multiple frusto-conical layers - wear resistant alloy

Support member - assembled withseal members and backing ring

Rotor / shaft - seal surface

Highpressure

side.

Larger diameter

& Segmented design

Concept Welded Design

Bolted Design


Introduction Pressure Activated Leaf Seal (PALS) Concept

5

Large startup & shut down clearance: Rub avoidance.

Minimum operating clearance: Performance gain.

Non-contacting operation: Long seal life. CLICK IMAGE


Smooth Rotor Testing Overview

6

Conditions:

• Speed – Typically 300ft/s (up to 345ft/s)

• Pressure Drop – Up to 120psi

• Seal Diameter pre Wear-in – 5.105” deflected (5.185” un-deflected)

AIR FLOW

• Rotor Diameter – 5.080” to 5.133”

• Fluid – Compressed air

• Rotor material – Aubert & Duval 819B (Ni-Cr-Mo Alloy) uncoated


Smooth Rotor Testing Static Leakage with Various Disk Sizes

7

0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

0.018

5.070 5.080 5.090 5.100 5.110 5.120 5.130

Effe

ctiv

e Cl

eara

nce

(inch

es),

[mm

]

Disk Diameter (inches), [mm]

30 psi [207 kPa]

40 psi [276 kPa]

50 psi [345 kPa]

60 psi [414 kPa]

70 psi [483 kPa]

80 psi [552kPa]

5.105" PALS ID,leaves deflected

Leaf vibration

[130.0] [130.3] [128.8] [129.0] [129.5] [129.8]

[0.051]

[0.102]

[0.152]

[0.203]

[0.254]

[0.305]

[0.356]

[0.406]

[0.457]

[129.3]


Smooth Rotor Testing Dynamic Offset Test

8

• 13500rpm • 50psi • 5.105” Seal Bore Size • 5.115” Disk Size • 0.010” Radial Offset to create a 0.005” rub. • Effective Clearance 0.007” prior to rub • Effective Clearance 0.0077” after rub

•After test, bore of seal re-edm’d to 5.114”

CLICK IMAGE


Smooth Rotor Testing - Wear-in Results Rotor Ø5.133”, Seal (deflected) Ø 5.114”

9

0.000

0.002

0.004

0.006

0.008

0.010

0.012

20 30 40 50 60 70 80 90 100

Effe

ctiv

e Cl

eara

nce

(inch

es),

[mm

]

Pressure Drop (psi), [kPa]

Wear-inRepeat 1Repeat 2Repeat 3Repeat 4Repeat 5

Start of wear-in

End of wear-in,start of

repeat 1

Rotor speed 13,500rpm, 302 ft/sPotential error of up to -15% of repeat

point value on effective clearance

Repeat cycle 1 difference

attributed to leaf tip burrs[0.051]

[0.102]

[0.152]

[0.203]

[0.254]

[0.305]

[138] [207] [276] [552] [621] [690][345] [414] [483]

• Orange line - ‘Wear-in’, the first time the seal is run

• Blue line repeat 1– post wear in burr stabilisations

• Consistent effective clearance throughout repeat 3 to 5

• Leaf tips wear to the rotor to establish good synchronicity

• Burrs form on both top and bottom leaves in both positive and negative axial direction


Smooth Rotor Testing Repeatability – Static Rotor Ø5.133”

10

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0 10 20 30 40 50 60 70 80 90

Effe

ctiv

e Cl

eara

nce

(inch

es),

[mm

]

Pressure drop (psi), [kPa]

Repeat 1 Repeat 2 Repeat 3 Repeat 4 Repeat 5

[138] [207] [276] [552] [621][69] [345] [414] [483]

[0.127]

[0.254]

[0.381]

[0.762]

[0.889]

[1.016]

[0.508]

[0.635]

0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

0 20 40 60 80 100 120 140

Effe

ctiv

e Cl

eara

nce

(inch

es),

[mm

]


Repeat 6 Repeat 7 Repeat 8 Repeat 9 Repeat 10

[0.051]

[0.102]

[0.152]

[0.203]

[0.254]

[0.305]

[0.355]

[138] [276] [552] [690] [966][414] [828]

[0.406]

• Higher pressure static leakage sweeps up to 120 psid post wear-in

• Repeatable closure with pressure

• No hysteresis

• Low pressure static leakage sweeps up to 80 psid post wear-in

• Repeatable closure with pressure

• No hysteresis


Smooth Rotor Testing Interleaf Leakage – Static Rotor Ø5.133”

11

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0 10 20 30 40 50 60 70 80 90Ef

fect

ive

Clea

ranc

e (in

ches

), [m

m]


Normal Leaves taped-up Interleaf Leakage

[138] [207] [276] [552] [621][69] [345] [414] [483]

[0.127]

[0.254]

[0.381]

[0.762]

[0.508]

[0.635]

• Removed front shroud to expose leaves

• Tested statically as standard

• Upstream face of top leaves taped to block air flow between leaves

• Tested statically with taped leaves

• Difference in effective clearance accounted to interleaf leakage


Smooth Rotor Testing Analytical Comparison- Static Rotor Ø5.133”

12

0.000

0.010

0.020

0.030

0.040

0.050

0 20 40 60 80 100 120 140

Effe

ctiv

e Cl

eara

nce

(inch

es),

[mm

]


Pre-Dynamic Test

Post-Dynamic Test

Interleaf leakage adjusted

FEA top leaf

Acousticnoise

[138] [276] [552] [690] [966][414] [828]

[0.254]

[0.762]

[0.508]

[1.270]

[1.016]

CLICK IMAGE

• Post wear-in closure

• Ø5.133” rotor

• Ø5.202” un-deflected PALS bore

• Analytical results follow closely to the measured results adjusted for interleaf leakage

• Acoustic noise was present at low pressures during the post dynamic testing

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 Frequency (Hz)

0.05 [1.27] 0.04 [1.02] 0.03 [0.76] 0.02 [0.51] 0.01 [0.25]

P-P,

mils

[μm

]


Acoustic Noise Investigation Experimental Set-up

13

Air in Kulite port

Down-streampressure tap

Air outUp-stream

pressure tap

Borescope port

Up-stream airthermocouple

Stator, max leaf deflection contact

plane

Rotating leaf assembly

• Leaf vibration resulted in an audible noise at low differential pressures

• Noise stopped at pressures above 30psid

• 2D bespoke rig designed and build to investigate potential causes along with existing static and dynamic set-ups


Acoustic Noise Investigation Hypothesized Cause and Discreditation

14

Hypothesis Result

Upstream pressure fluctuations Disproved – Unaffected by controlled bypass leakage A cavity in the air supply system resonates at the

frequency of the leaves

After wear-in burrs at the tip of the middle leaf silence the noise by supporting top leaves Disproved – Comparison testing from post wear-

in and re-machined chisel like bore. No noise or vibration Vortex shedding from leaf tips excites leaves at their

natural frequency

Wear-in changes leaf natural frequency Disproved – Shim testing proved interleaf gap controlled noise and vibration

Flow approach relative to leaf angle a factor in vortex shedding

Disproved –Leaf angle ineffective on noise and vibration


Acoustic Noise Investigation Hypothesized Cause and Discreditation

15

Hypothesis Result

HCF cracking Disproved – No evidence of HCF from visual inspections using a microscope and dye penetrant

Cushion of air between the support and the leaves Disproved – No need for testing due to noise elimination before venting test

Insufficient damping Disproved – Manual excitation unsuccessful

Interleaf gaps allow relative motion without damping Proved – Shim testing proved interleaf gap greater then .001” (0.025mm) would instigate noise and vibration

Blunt leaf ends redirect flow up leaves Proved – Re-cut leaves with chisel end ensuring at full pressure only the top leaves were in contact with the rotor. No noise or vibration


Acoustic Noise Investigation Established Cause – Interleaf Separation

16

5.133in PALS bore dia. [130.4mm] No burrs, 5psi, [34.5kPa]

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 Frequency (Hz)

0.05 [1.27]

0.04 [1.02]

0.03 [0.76]

0.02 [0.51]

0.01 [0.25]

P-P,

mils

[μm

]

5.133in PALS bore dia.[130.4mm] No burrs, 10psi, [69kPa]

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 Frequency (Hz)

0.05 [1.27]

0.04 [1.02]

0.03 [0.76]

0.02 [0.51]

0.01 [0.25]

P-P,

mils

[μm

]


Simulated Shrouded Turbine Blade Test Overview – Dynamic

17 [0

.08]

View

B

View B

View A

0.0125in[0.32mm]

[0.64]

[4.6

5]

0.025in[0.64mm]

0.00

3in

[0.0

8mm

]

0.18

3in

[4.6

5mm

]

Conditions: • Speed – 20,000rpm • Slot frequency – 4000Hz • Rotor Diameter – Ø5.120” • Cold build clearance – 0.018” • Fully deflected leaf interference – 0.003”

ROTOR AIR FLOW ΔP = 65psi (448kPa)

SEAL AIR FLOW ΔP = 55psi (379kPa)

Cross Manufacturing Company (1938) Limited Brush Seals and Aerospace Products, South Site, Hopton Road, Devizes, Wiltshire, SN10 2EU 18

Simulated Shrouded Turbine Blade Test 15 Hour Steady State Results - Dynamic

0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0:00:00 3:00:00 6:00:00 9:00:00 12:00:00 15:00:00

Tota

l Effe

ctiv

e Cl

eara

nce

(inch

es),

[mm

]

Time (h:mm:ss)

1 2 3 4

[0.025]

[0.051]

[0.076]

[0.152]

[0.178]

[0.203]

[0.102]

[0.127]

• Rotating down the step

• 15 hrs broken down into 4 stages with inspection

• Total effective clearance reported

• Initial period of wear-in before stabilisation

.045in [1.14]

.0003in [7.6μm]

15hr

.0001in [2.5μm]

.048in [1.22]

1hr


Simulated Shrouded Turbine Blade Test Reverse Rotation Results - Dynamic

0

3000

6000

9000

12000

15000

18000

21000

24000

0

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0:00:00 0:10:00 0:20:00 0:30:00 0:40:00

Roto

r Spe

ed (r

pm)

Tota

l Effe

ctiv

e Cl

eara

nce

(inch

es),

[mm

]

Time (h:mm:ss)

Total Effective Clearance Rotor Speed (rpm)

[0.025]

[0.051]

[0.076]

[0.152]

[0.178]

[0.203]

[0.102]

[0.127]

• Rotating up the step

• Rotor speeds from 5,000rpm to 20,000rpm

• Slot frequency from 1000Hz to 4000Hz

• Improved effective clearance with speed

• Comparable results to the 15hr steady state test

• Stable running

CLICK IMAGE


Simulated Shrouded Turbine Blade Test Pre and Post Test Comparison - Static

20

0.000

0.004

0.008

0.012

0.016

0.020

0.024

0 10 20 30 40 50 60 70 80

Effe

ctiv

e Cl

eara

nce

(inch

es),

[mm

]


Pre-Testing Post Testing

[0.102]

[0.203]

[0.406]

[0.508]

[0.610]

[0.305]

[138] [207] [276] [552][69] [345] [414] [483]

• Noise and vibration present on the pre testing test at pressure drops under 20psi both increasing and decreasing the pressure

• No noise post dynamic tests

• Improvement on total effective clearance


Summary

21

• The PALS design is technology ready

• Smooth rotor testing – No loss of integrity

– Velocity of 300ft/s

– Pressure drop of 120psi

– Radial offsets and 360° rub

• Wear in process

– Relax manufacturing tolerances


Summary

22

• Acoustic noise accounted to interleaf separation

• Simulated shrouded turbine blade testing - Suitability for turbine blade tip applications

– 15 hours consistent performance

– Seal pressure drop of 55psi

– Rotor pressure drop of 65psi

– Rotor pressure 10psi higher than seal pressure

– Radial step of .003”

– Step frequency of 4000Hz


Future Work

23

• Influences of manufacturing tolerances and preload

• Additional reverse rotation testing on simulated shrouded turbine blade with varying degrees of radial offset, eccentricity and rotor pressure to investigate the effects on stability, wear, excitation and high cycle fatigue

• An analytical approach to predicting wear rate based on operating conditions


• sCO2 turbomachinery small size requires leakage control in limited space.

- PALS effective clearance is < 0.004in (0.1mm) and has small cross-section.

• sCO2 turbomachinery operating speed is typically very high. - PALS performance is independent of speed.

• sCO2 turbomachinery operating pressure is high and can require high DP seals.

- PALS design for high pressure is unrestricted.

• sCO2 turbomachinery rotor stability can be affected by ‘flashing’ within seals. - PALS single pressure drop mitigates this concern.

• sCO2 turbomachinery power loss within seals and from windage.

- PALS non-contacting leaves and low leakage should diminish power loss.

• sCO2 turbomachinery manufacturing cost. - PALS ‘Wear-in’ capability can reduce tolerance requirements and cost.

Assessment of application in sCO2 turbomachinery


PALS Compared to Other Seals

Startup Rub

Vulnerability

Operating Rub

Tolerance

‘Wear-in’ Seal

Compliance

Low Leakage

Long Seal Life

High Seal ΔP

Capability

Reverse Rotation Hazard

Axial Length

Rotor Dynamic Issues

Cost

Labyrinth Seals

Brush Seals

Pressure Actuated

Leaf Seals Limited ~ 0.004in

W/o rub.

Improved to 8+ years.

AbradibleLabyrinth

Seals

Symbol key: Unfavorable comparison

Favorable comparison

Somewhat favorable comparison

400psid capability.


• Pressure Actuated Leaf Seal technology readiness test

results and favorable evaluation of attributes vs other seals

show benefit potential for their application in sCO2

turbomachinery.

• Consideration of their use in sCO2 turbomachinery is

recommended.

Recommendation


Head Quarters - Bath Automotive Division Devizes North Site

Brush Seal & Aerospace Division Devizes South Site

Thank You for Listening

Any Questions?

pressure activated leaf seal technology readiness testing

Documents