Evaluation and Successful Modification of Impeller Using Forced Response Analysis



Similar documents
Technology of EHIS (stamping) applied to the automotive parts production

Repair of damaged rotor journal portion

Joints - Rolls-Royce Perspective

Relevance of Modern Optimization Methods in Turbo Machinery Applications

Finite Element Analysis for Acoustic Behavior of a Refrigeration Compressor

Investigation of Stress Intensity Factor of Axial Compressor Blade of Helicopter

CFturbo Modern turbomachinery design software

Rapid Design of an optimized Radial Compressor using CFturbo and ANSYS

Sulzer Turbo Services

Dynamics of Offshore Wind Turbines

FORENSIC ENGINEERING ANALYSIS OF A COMBUSTION TURBINE COMPRESSOR BLADE FAILURE

Design Validation and Improvement Study of HVAC Plumbing Line Assembly under Random Loading Condition

Design and testing of a high flow coefficient mixed flow impeller

CH 6: Fatigue Failure Resulting from Variable Loading

Systematic Optimisation of Gearboxes for Hybrid and Electric Vehicles in Terms of Efficiency, NVH and Durability

F.I.R.S.T. Gas Turbine End of Life Presentation

Testing Capabilities. Worldwide Designers & Manufacturers of Air Moving Equipment

Chapter 3.5: Fans and Blowers

Pump Vibration Analysis

PULSATION REDUCTION BY ACOUSTIC FILTERS FOR METERING APPLICATIONS. Robert J. McKee Ray G. Durke

Variable Frequency Drives - a Comparison of VSI versus LCI Systems

G1RT-CT D. EXAMPLES F. GUTIÉRREZ-SOLANA S. CICERO J.A. ALVAREZ R. LACALLE W P 6: TRAINING & EDUCATION

User orientated simulation strategy to analyse large drive trains in SIMPACK

Technical Data. 7. Bearing Fits. 7.1 Interference. 7.2 Calculation of interference F B LLLLLLLLL( A-54

export compressor instability detection using system 1* and proficy** smartsignal software

Mechanical &Electrical Research Institute, National Water Research Center, Delta Barrage, Egypt.

Frequecy Comparison and Optimization of Forged Steel and Ductile Cast Iron Crankshafts

Titanium 50 inch and 60 inch Last-stage Blades for Steam Turbines

Initial Design and Optimization of Turbomachinery with CFturbo and optislang

Calibrating a Large Compressor s Rotordynamic Model: Method and Application

AISI O1 Cold work tool steel

7.3 Fit selection. Inner ring: Rotating. Outer ring: Stationary. Inner ring: Stationary. Outer ring: Rotating. Inner ring: Stationary

Introduction of the Solar Turbines Titan 250 Gas Turbine System

X15TN TM. A high hardness, corrosion and fatigue resistance martensitic grade CONTINUOUS INNOVATION RESEARCH SERVICE.

COMPUTATIONAL FLUID DYNAMICS (CFD) ANALYSIS OF INTERMEDIATE PRESSURE STEAM TURBINE

Mechanical Design of Turbojet Engines. An Introduction

DET SIGMA PUMPY HRANICE PUMP TURBINES

Vibration Analysis Services. FES Systems Inc. 1

Fluid structure interaction of a vibrating circular plate in a bounded fluid volume: simulation and experiment

Nonlinear Modal Analysis of Mechanical Systems with Frictionless Contact Interfaces

TYPE APPROVAL CERTIFICATION SCHEME MASS PRODUCED DIESEL ENGINES

1.3 Turbochargers are categorised in three groups depending on served power by cylinder groups with:

Practice Problems on Pumps. Answer(s): Q 2 = 1850 gpm H 2 = 41.7 ft W = 24.1 hp. C. Wassgren, Purdue University Page 1 of 16 Last Updated: 2010 Oct 29

Master of Simulation Techniques. Lecture No.5. Blanking. Blanking. Fine

Surface Technology. Centrifugal Fan GMD

Fatigue. 3. Final fracture (rough zone) 1. Fatigue origin. 2. Beach marks (velvety zone)

Performance Improvement of Pump-turbine for Large Capacity Pumped Storage Power Plant in USA

Vibration analysis: what does it mean?

System. Stability. Security. Integrity. 150 Helical Anchor

Geotechnical Design Monopile Foundations for Offshore Wind Turbines

Vehicle Design Summit Electric Hub Motor (V2) Eric Conner Harvey Tang Matthew Peddie

Vertical Multistage Centrifugal Electric Pumps SV Series NEW LASER TECHNOLOGY

SJT / SJM / SJP Large Vertical Pumps

The Design & Analysis of a Low NPSH Centrifugal Pump Featuring a Radial Inlet and Axial Inducer Using STAR-CCM+

V PREFAB WELDING PREFAB WELDING

Structural Integrity Analysis of the Charging Air Tube Support for a Diesel Engine of Commercial Vehicle

FAN & BLOWER CENTRIFUGAL FANS

THE ROLE OF ONLINE AERODYNAMIC PERFORMANCE ANALYSIS

The use of Operating Deflection Shapes (ODS) to model the vibration of sanders and polishers HSL/2006/104. Project Leader: Author(s): Science Group:

Operataing Manual. Single stage radial Fan. Model MKV/TKV

Design and Analysis of Cross Flow Turbine for Micro Hydro Power Application using Sewerage Water

INJECTION MOLDING COOLING TIME REDUCTION AND THERMAL STRESS ANALYSIS

Highly flexible couplings

Mechanical Property Changes in Steel during the Pipe Making Process Brent Keil 1

Mechanical Installation

HIGH SPEED PERMANENT MAGNET SYNCHRONOUS MOTOR / GENERATOR DESIGN FOR FLYWHEEL APPLICATIONS

WATERJET VISUAL TECHNICAL INSPECTION REPORT

Evaporation Technology Using Mechanical Vapour Recompression

SUNON SPECIFICATION FOR APPROVAL DESCRIPTION : DC BRUSHLESS FAN. DIMENSIONS : 60X60X15 mm M O D E L : KD0506PHS2

Frequency domain application of the Hot-Spot method for the fatigue assessment of the weld seams

Fatigue Performance Evaluation of Forged Steel versus Ductile Cast Iron Crankshaft: A Comparative Study (EXECUTIVE SUMMARY)

THE COMPOSITE DISC - A NEW JOINT FOR HIGH POWER DRIVESHAFTS

HORIZONTAL PUMPING SYSTEM

Engine bearing failures and how to avoid them

Case Studies on Paper Machine Vibration Problems

FATIGUE TESTS AND STRESS-LIFE (S-N) APPROACH

HOBBING MACHINE TYPE ZFWZ 8000x40

AISI CHEMICAL COMPOSITION LIMITS: Nonresulphurized Carbon Steels

Begin creating the geometry by defining two Circles for the spherical endcap, and Subtract Areas to create the vessel wall.

1150 hp motor design, electromagnetic and thermal analysis

Energy Saving ECM Motors and Fans Ventilation Catalogue

Repair of Hyd-ro-ac Actuators

BSM MOTOR DRIVEN CENTRIFUGAL PUMPS

SUNON SPECIFICATION FOR APPROVAL

Rapid Modal Testing System for Live Load Rating of Highway Bridges

Whirling Machine. New Dimension of Whirling Technology. Thread Whirling Machines. Big Workpiece Diameter. High Pitch Thread

IMPELLER FATIGUE ASSESSMENT USING AN S-N APPROACH

Stress Analysis of a Chicago Electric 4½ Angle Grinder

GMC 2013: Piping Misalignment and Vibration Related Fatigue Failures

Transcription:

Turbomachinery Symposium 2008 Case study Evaluation and Successful Modification of Impeller Using Forced Response Analysis Bhabesh Thakur Hui Kuang Robert Huffman Murari Singh GE Oil & Gas, Bethlehem, PA 1

Background History A compressor impeller failed in the field with multiple cracks and missing chunks of material from the blades were found at the inlet after 1,000~1,200 hours of service. It appeared that a foreign object caused the damage (FOD). The impeller was taken off the line for repair and a spare unit installed. 2

Background History (Continued) The spare unit tripped off line following high compressor vibration after 1,700 hours of service. Unit was disassembled and it was discovered that the impeller blades were damaged similar to the first event. Inlet screen and upstream components were investigated to locate signs of FOD. No signs of FOD were found. Pictures of failed blades with missing material are shown on Figure 1. 3

Pictures of Failed Blades Figure 1: Pictures Showing Failed Blades with Missing Material 4

Evaluation of Original Design (1) The unit has 12 Inlet Guide Vanes (IGV). Modal Analysis indicated possibility of resonance condition with 2x(IGV) forcing frequency and impeller natural frequencies at, 1) 1937 Hz, 5-ND (1C) (5 Nodal Diameter, 1st Order Mode) 2) 2006 Hz, 5-ND (2C) (5 Nodal Diameter, 2nd Order Mode) The FEA Model and Modal Analysis results are shown in Figure 2 through Figure 9. 5

Evaluation of Original Design (2) Figure 2: FEA Model of the Original Design 6

Evaluation of Original Design (3) Figure 3: Radial Component, Mode Shape 1937 Hz, 5-ND (1C), Original Design 7

I Evaluation of Original Design (4) I 1 ANSYS 10.0A1 MAR 6 2006 16:23:03 NODAL SOLUTION STEP=9999 UY (AVG) RS':S=l PowerGraphics EFACET=l AVRES=Mat DMX =. 014626 SMN =-.008352 SMX =.009659 -.008352 r-1 -. 00635 ~ - -.004349 ~ -.002348 ~ -.347E-03 ~.001654 ~. 003656 L j 005657 - CJ. 007658. 009659 Figure 4: Tangential Component, Mode Shape 1937 Hz, 5-ND (1C), Original Design 8

Evaluation of Original Design (5) Mode Animation Figure 5: Axial Component, Mode Shape 1937 Hz, 5-ND (1C), Original Design 9

I Evaluation of Original Design (6) 1 ANSYS 10. 0A1 MAR 13 2006 09:58:45 NODAL SOLUTION STEP=9999 ux (AVG) RSYS=1 PowerGraphics EFACET=1 AVRES=Mat DMX =. 007169 SMN =-. 0 0 4678 SMX =.004304 --.004678 -.00368 o -.002682 D -.001684 D D. 312E-03 D.00131 D.002308 D.003306 -.004304 -.686E-03 I Figure 6: Radial Component, Mode Shape 2006 Hz, 5-ND (2C), Original Design 10

I 1 Evaluation of Original Design (7) ANSYS 10.0A1 ~AR 13 2006 09:59:10 ~ODAL SOLUTION STEP=9999 UY (AVG) RSYS=1 PowerGraphics E:FACET=1.li.VRE S =Mat DMX =. 007169 SMN =-.003738 SMX =.005058 -.003738 1--J -.002761 -~ -. 001783 A -.806E-03 ~.172E-03 L._J 001149 ~. 002126 ~. 003104 L._J 004081.005058 - I Figure 7: Tangential Component, Mode Shape 2006 Hz, 5-ND (2C), Original Design 11

I Evaluation of Original Design (8) I NODAL SOLUTION STEP=1 SUB =37 FREQ=2006 UZ ( AVG) RSYS=1 DMX =9.52 SMN =-6.219 SMX =6. 255 J\N J AN 1 9 2008 19:41 : 48-6. 219-3.447 -. 67 4865 2.097-4. 833-2.061. 711144 3. 483 4. 869 6. 255 Figure 8: Axial Component, Mode Shape 2006 Hz, 5-ND (2C), Original Design 12

2,500 Evaluation of Original Design (9) Original Design Impeller Interference Diagram 2,000 24E (2xIGV) 1,500 Frequency, Hz 12E (1xIGV) 1,000 Nmax Speed 5,335 rpm 500 Min Speed 4,000 rpm 0 0 1 2 3 4 5 6 7 8 9 Nodal Diameter Family Figure 9: Interference Diagram, Original Design 13

Forced Response Analysis of Failed Impeller (1) The Results of the Forced Response Analysis conducted at the possible resonance frequencies are as follows, 1) 1937 Hz, 5-ND (1C) Goodman Factor of Safety 0.9 (Figure 10 ). 2) 2006 Hz, 5-ND (2C) - Goodman Factor of Safety 1.3 (Figure 11 ). 3) Goodman Factor of Safety 0.9 and 1.3 for the above cases are less than the acceptable Margin of Safety 1.5. 14

Forced Response Analysis of Failed Impeller (2) Goodman Diagram Harmonic Response Stresses, 1937 Hz, 5-ND (1C), Original Design 60 50 Fatigue Strength Material = AISI 4320 Steel Steady Stress Based on: Rotor Speed = 5,100 rpm & Steady Gas Load 40 Fatigue Strength Adjusted for Surface Finish Alternating Stress Based on: Harmonic Response of 5% (peak to peak) of the Gas load Alternating Stress, ksi 30 F.S. = 0.9 20 10 0 0 10 20 30 40 50 60 70 80 90 100 110 120 Steady Stress, ksi UTS Figure 10: Goodman Diagram, Response Stress, 1937 Hz, 5-ND (1C) 15

Forced Response Analysis of Failed Impeller (3) Goodman Diagram Harmonic Response Stresses, 2006 Hz, 5-ND (2C), Original Design 60 Fatigue Strength Material = AISI 4320 Steel 50 40 Fatigue Strength Adjusted for Surface Finish Steady Stress Based on: Rotor Speed = 5,100 rpm & Steady Gas Load Alternating Stress Based on: Harmonic Response of 5% (peak to peak) of the Gas load Alternating Stress, ksi 30 20 F.S. = 1.3 10 UTS 0 0 10 20 30 40 50 60 70 80 90 100 110 120 Steady Stress, ksi Figure 11: Goodman Diagram, Response Stress, 2006 Hz, 5-ND (2C) 16

Reason For Failure Possible Resonance condition for 2x(IGV) Forcing frequency with Impeller natural frequencies at 1937 Hz, 5-ND (1C) and/or 2006 Hz, 5-ND (2C) may have caused failure. Considering Forced Response Stress, Goodman Factor of Safety 0.9 and 1.3 for the above cases are less than the acceptable Margin of Safety 1.5, which eventually may have resulted in the failure. 17

Design Modifications Design Modification: 1) Weld repair current damaged impeller. 2) Machine back blade Leading Edge 0.75 in. (Re-radius Leading Edge to maintain 0.17 in. normal thickness). 3) Perform Modal Analysis and Forced Response Analysis of new assembly. Affect on Impeller Performance: 1) Decreases head approximately 0.65% 2) Decreases flow approximately 0.12% 18

Modified Design (1) The FEA Model with modified design, blade trimmed at leading edge by 0.75 inch, was created. The Frequency variation of the Tested Blades with the FEA model was 3.32-6.38 % (lower). The FEA Model and Modal Analysis results are shown in Figure 12 to Figure 16. The Goodman Diagram Based on the Results of Forced Response Analysis is shown in Figure 17. 19

I Modified Design (2) I 1 Figure 12: FEA Model Modified Design, Blade Trimmed at Inlet 0.75 in. 20

I Modified Design (3) I 1 ANSYS 10.0A1 MAR 10 2006 14:10:27 NODAL SOLUTION STEP=9999 ux (AVG) RSYS=l PowerGraphics EFACET=l AVRES=Mat DMX =.542E-03 SMN =-.117E-03 SMX =.227E-03 --.117E-03 -.786E-04 Cl -.404E-04 D -.227E-05 D.359E-04 D.740E-04 D. 112E-03 D. lsoe-03 D.lBBE-03 -.227E-03 GE-Conmec 1973.7Hz (SND) Figure 13: Radial Component, Mode Shape 1974 Hz, 5-ND (1C), Modified Design 21

I Modified Design (4) I 1 ANSYS 10.0A1 MAR 10 2006 14:10:27 NODAL SOLUTION STEP=9999 UY (AVG) RSYS=l PowerGraphics EFACET=l AVRES=Mat DMX =.542E-03 SMN =-.113E-03 SMX =.107E-03 -.113E-03 --.886E-04 Cl -.641E-04 D -.396E-04 D -.lsle-04 D.935E-05 D.338E-04 D.583E-04 D.828E-04 -. 107E-03 GE-Conmec 1973.7Hz (SND) Figure 14: Tangential Component, Mode Shape 1974 Hz, 5-ND (1C), Modified Design 22

I Modified Design (5) I 1 NODAL STEP=1 SUB =18 FREQ=1974 UZ (AVG) RSYS=O DMX =6.344 SMN =-5.709 SMX =5.768 J\N JAN 19 2008 14:42:14-5.709-3.158 -.607973 1.942-4.434-1.883.66726 3.218 Model with 0.75 shorter Blade, Trimmed at Inlet 4.493 5.768 Figure 15: Axial Component, Mode Shape 1974 Hz, 5-ND (1C), Modified Design 23

Modified Design (6) 2,500 Modified Design Impeller, Blade leading Edge Cut (0.75 inch) 2,000 24 (2 X IGV) Frequency, Hz 1,500 1,000 12 (1 X IGV) Max Speed = 5,335 rpm 500 Min Speed = 4,000 rpm 0 0 1 2 3 4 5 6 7 8 9 Nodal Diameter Family Figure 16: Interference Diagram, Modified Design 24

Forced Response Analysis Modified Design (1) The Results of the Forced Response Analysis conducted at the possible resonance frequency are as follows, 1) 1974 Hz, 5-ND (1C) Goodman Factor of Safety 2.4 (Figure 17 ). 2) Goodman Factor of Safety 2.4 for the above case is greater than the acceptable Margin of Safety 1.5. 25

Forced Response Analysis Modified Design (2) Goodman Diagram Harmonic Response Stresses, 1974 Hz, 5-ND (1C), Modified Design 60 Fatigue Strength Material = AISI 4320 Steel 50 Steady Stress Based on: Rotor Speed = 5,100 rpm & Steady Gas Load Fatigue Strength Adjusted for Surface Finish Alternating Stress Based on: Harmonic Response of 5% (peak to peak) of the Gas load 40 Alternating Stress, ksi 30 ` 20 10 0 F.S. = 2.4 UTS 0 10 20 30 40 50 60 70 80 90 100 110 120 Steady Stress, ksi Figure 17: Goodman Diagram, Response Stress, 1974 Hz, 5-ND (1C) 26

27 Conclusion Original design was modified: 1) Blade trimmed at leading edge by 0.75 inch. 2) Damaged blades repaired. The above modifications resulted in increased Impeller frequencies and acceptable Goodman Factor of Safety: 1) 1937 Hz, 5-ND (1C) increased to 1974 Hz, 5-ND (1C) and Goodman Factor of Safety increased from 0.9 to 2.4 2) 2006 Hz, 5-ND (2C) increased to 2357 Hz, 5-ND (2C) Out side range of the Forcing Frequency. The unit with the modified design has been in service during the past last 3 years without a failure.