Steam Turbine Generator Core Isolation Issues at the Redhawk Combined Cycle Plant

Transcription

1 Steam Turbine Generator Core Isolation Issues at the Redhawk Combined Cycle Plant Marques Montes Arizona Public Service Company Phoenix, AZ Includes material provided by John Demcko

2 Redhawk Power Plant

3 Redhawk Power Plant Located west of Phoenix near Palo Verde Two 530 MW Units Each unit consists of 2 combustion turbines and 1 steam turbine Commercial operation began in 2003

4 Part 1 End Winding Issues

5 Early Problems From commissioning onward, the STGs at Redhawk were unusually loud during operation. June STG manufacturer warns APS of possible high end turn vibration Sister unit at another site has a stator winding failure Manufacturer recommends immediate shutdown of both STGs at Redhawk Loss of 400MW for Summer Peak

6 The Plan Not Running STGs Summer 2003 Unacceptable Alternative Sought A Means to Monitor Real-time Stator Required We Obtained The Mfgr s Concurrence to Run with Monitoring Required Acceptable/Unacceptable Vibration Levels Defined

7 The Interim Solution Turn-key End Turn Vibration Monitoring Via Fiber Optics Based Accelerometers Vibration Reads at Frequent Intervals Communication with Mfgr When Alert Levels Reached

8 FOA Installed On NDE

9 2003 Results Temporary Instrumentation Config. STG1 Vibration Levels Acceptable STG2 Levels High & Increasing End Turn Structure Repairs Made In October Permanent FOA Instrumentation Package Installed For 2004 Season

10 Vibration, mils pk-pk 2003 Results - STG Redhawk STG2 End Turn Vibration FOA1 Radial 3:30 FOA3 Radial 7:30 FOA5 Radial 11: /28/03 0:00 8/7/03 0:00 8/17/03 0:00 8/27/03 0:00 9/6/03 0:00 9/16/03 0:00 9/26/03 0:00 10/6/03 0:00 Date & Time

11 2004 Results Units Not Operated Extensively Until June STG1 Vibration Higher Than Before End Turn Modifications STG2 Started Low But Climbed Throughout The Year To Well Above 2003 Alarm Level Additional End Turn/Phase Lug Work Scheduled Spring 2005

12 2004 Results - STG2 (June)

13 2004 Results - STG2 (July - December)

14 2005 & On End Turn Vibration Instrumentation Package Is An Effective & Necessary Tool Excessive Vibration Detected Early & Treated As A Warranty Issue STGs Ran Through 2003 & 2004 Peak End Turn Monitoring was Utilized To Assess Spring 2005 Repairs

15 Driven End Problem Arises We believed that with monitoring, vibration issues were under control. Then came April 3 rd, 2009 STG2 Operating Point was 137MW, +9.6MVAR & Vgen = 18.5kV when.

16 A Phase B to Ground Fault Develops

17 Followed by an evolving 3 phase fault

18 Fault Detail

19 Damage Is Substantial! 3Ph Fault Current from Yard is 75kA 0-pk Gen Contribution Unknown due to CT loc. Burning Last Seconds While Field Current Decays From DFR Record, The Worst Case Scenario Is Feared

20 Driven End Fault Damage - 1

21 Driven End Fault Damage - 2

22 Damage Assessment 3 Burned Series Connections On DE Rotor & Stator Badly Contaminated with Carbon & Melted Copper Driven End Turn Structure Repairable Work is Extensive & Labor Intensive

23 Repair Work Scope Strip Stator Bars Back to Expose Good Copper Perform Series Connection Braze Repairs Validate Repairs by Phase Resistance Test & Thermal Camera Insulate (Caps) Series Connections Remove & Reinstall All Series Blocking & Tie With Glass Roving & Epoxy Resin. Apply Red Eye Epoxy & Flooding Resin to the End Windings Install Nose Ring On ID of Series Connection/Caps Bump Test Verify Frequencies Final Electrical Testing (DC leakage, Hipot & Winding Resistance

24 Driven End Repair - 1

25 Driven End Repair - 2

26 STG2 Repaired & Returned To Service FOAs on both DE & NDE (12 total) Will Report To Plant Computer via Next Generation Monitor 4 Velocimeters Installed on Core In Quadrature (120Hz Audible Noise Very High) VibroSystM PCU-5000 Installation Commissioned 4 th Q 2010

27 Part II Cooler Leak and Core Isolation Inspection

28 March 2013 STG2 Cooler Leak Ongoing high vibration caused an air vane to break loose and fall onto a water to air cooler Many gallons of the water-glycol mix ( antifreeze ) were circulated through the machine As part of the repair process, the coolers were removed

29 Typical air vane.

30 Damaged cooler in the shop.

31 Close up shot of cooler damage

32 The Discovery During the post-incident inspection of the machine, it became apparent that the machine has virtually no core isolation. The support structure consists of the stator key bars welded to an annular support ring, aka the ring with ears Each ring with ears is solidly welded to the generator wrapper at 6 points.

33 Core key bar welded to ring with ears

34 Ring with ears

35 These circular connections are where the rings with ears are welded to the wrapper.

36

37 Vibration in the core is transferred to the wrapper through the inadequate core isolation. The wrapper vibrates at 120 Hz like a speaker cone. Near field acoustic data from Redhawk ST2

38 Proposed Solutions After the cooler leak was repaired, the unit was placed back into service. We explored our options on mitigating the excessive vibration caused by the lack of core isolation: Harmonic dampener Wrapper reinforcement Generator replacement

39 What did we decide?

40 Stator ready to be scrapped.

41 New generator

42 New Generator Acoustic Results 23 db reduction in sound pressure level in the 125 Hz (120 Hz) range!

43 Conclusions We believe that the susceptibility of stator end winding failure on these units is due to a flawed design that inadequately isolates the core from the rest of the machine. This excessive vibration also contributed to the failure of the air vane that led to a massive cooler leak. This design flaw would have been prohibitively expensive to repair. Generator replacement was chosen as the best option to protect the reliability of the system. Monitoring the core and end windings of a machine for vibration can give you insight into a potential problem.

44 Questions?