Westinghouse Presentation Template Class 3
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
35th EPRI NDE Conference Foreign Object Wear Detection Capabilities Using the Bobbin Probe Bill Cullen Fellow Engineer; Westinghouse Electric Company LLC Utility Sponsors: Allison Bassett, Doug Hansen, Warren Leaverton, APS
1
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Topics • • • • • •
Background Description of Development Program Bobbin Detection Summary Auto History Compare Results Summary Tube Integrity Analysis Impact Conclusions
2
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Background • With RSG inspection cycling (skipping), improved foreign object
(FO) wear detection with eddy current has increased importance • Top-of-tubesheet (TTS) vs Upper Bundle • Historical: bobbin inspection combined with foreign object search and retrieval (FOSAR) of periphery • Recent: +PointTM1 probe or X-ProbeTM1 inspection of periphery • No official examination technique specification sheet (ETSSs) available but detection expected to be improved over bobbin
• In either case, detection capabilities at the expansion transition are not documented
1: +Point and X-Probe are trademarks of Zetec, Inc., in the United States and/or other countries. Other names may be trademarks of their respective owners.
3
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Description of the Development Program • To support a spring 2016 outage, the Westinghouse Data Union Software (DUS) was used to document bobbin coil detection capability • Donor (flaw) signals selected based on a realistic assessment of the likelihood of morphology • Donors taken from ETSS 27091
• Circumferential Groove (wire, weld rod) • Axial Groove (bounding with regard to axially oriented degradation) • Round Tapered Hole (bounding detection case – not realistic)
4
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Description of the Development Program • Three-frequency mix (P5) applied for FO wear at APS • Develop P5 noise distributions at TTS
• Four of six SGs examined
• Unit 2 (different expansion process) and Unit 3 (U1 and U3 use the same expansion process)
• 95th percentile P5 noise values selected to define low, mid,
and high noise condition hosts • Noise data easily extracted using Westinghouse Real Time Auto Analysis (RTAA) software • Select low (1.18V), mid (1.75V), and high (2.64V) P5 noise conditions as hosts
5
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
P5 Noise Distributions
6
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Bobbin Detection Summary – Manual Analysis • Bobbin data collected for the three donor morphologies • Injections performed locating flaws just above TTS (0.08 inch) and at ~0.3 inch above TTS
• For high noise tube at TTS, reliable manual detection judged at:
• 60%TW: Circumferential Groove • 20%TW: Axial Groove
• Tapered wear similar to axial groove performance
• 75%TW: Round Tapered Hole
Improved detection performance in mid and low noise tubes
7
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Example Bobbin Graphic: 60% Circ Groove, High Noise Tube, Injection at TTS
8
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Tube Integrity Impact – Manual Analysis • Volumes defined for the various flaw shapes • “Must detect” depths identified to support two operating
cycles between eddy current inspections based on additional volume growth • Identified reliable detection depths support the integrity analysis with postulated growth of two additional cycles up to the next scheduled eddy current inspection • But can we do better? Yes – with Auto History Compare (AHC)
9
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
EADS/AHC Setup • Field application included auto-auto analysis with AHC used as a tertiary analysis • EADS used as primary analysis pass
– EADS extraction algorithms applied to achieve equal detection as manual analysis
• AHC extraction algorithms provide improved detection (lesser depth) compared to EADS/manual
AHC reliable detection depths were about 10 to 15%TW less than manual analysis of ALFS mix
10
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Auto History Compare Results Summary • AHC reliably extracted flaw-like signals at less than manual • • • •
detection thresholds AHC currently only considers raw differential channels Mix channels showed similar performance but extraction algorithm is not verified • Future development to validate mix channels Application of AHC improved overall detection capabilities far beyond manual analysis, thus adding margin Conclusion: For this combination program, tube integrity is maintained up to the next eddy current inspection
11
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
MAPOD Investigation • Manual analysis
results show ARP of 1 is appropriate
12
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
POD Curve Comparison
13
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Field Implementation – Spring 2016 Outage • • • • •
EADS/AIDA primary/secondary analysis Tertiary AHC analysis of outer 5 peripheral tubes RTAA to screen noise at TTS (2.5V threshold) X-Probe inspection at locations with >2.5V of P5 noise For the spring 2016 outage, only 4 locations required testing with X-Probe (SG11: 95th percentile noise = 1.71V, SG12, 1.45V) • Thus, • Applied auto detection capabilities exceed manual analysis capabilities • Multiple methods/techniques produce a robust FO wear inspection •
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
35th EPRI NDE Conference Foreign Object Wear Detection Capabilities Using the Bobbin Probe Bill Cullen Fellow Engineer; Westinghouse Electric Company LLC Utility Sponsors: Allison Bassett, Doug Hansen, Warren Leaverton, APS
1
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Topics • • • • • •
Background Description of Development Program Bobbin Detection Summary Auto History Compare Results Summary Tube Integrity Analysis Impact Conclusions
2
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Background • With RSG inspection cycling (skipping), improved foreign object
(FO) wear detection with eddy current has increased importance • Top-of-tubesheet (TTS) vs Upper Bundle • Historical: bobbin inspection combined with foreign object search and retrieval (FOSAR) of periphery • Recent: +PointTM1 probe or X-ProbeTM1 inspection of periphery • No official examination technique specification sheet (ETSSs) available but detection expected to be improved over bobbin
• In either case, detection capabilities at the expansion transition are not documented
1: +Point and X-Probe are trademarks of Zetec, Inc., in the United States and/or other countries. Other names may be trademarks of their respective owners.
3
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Description of the Development Program • To support a spring 2016 outage, the Westinghouse Data Union Software (DUS) was used to document bobbin coil detection capability • Donor (flaw) signals selected based on a realistic assessment of the likelihood of morphology • Donors taken from ETSS 27091
• Circumferential Groove (wire, weld rod) • Axial Groove (bounding with regard to axially oriented degradation) • Round Tapered Hole (bounding detection case – not realistic)
4
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Description of the Development Program • Three-frequency mix (P5) applied for FO wear at APS • Develop P5 noise distributions at TTS
• Four of six SGs examined
• Unit 2 (different expansion process) and Unit 3 (U1 and U3 use the same expansion process)
• 95th percentile P5 noise values selected to define low, mid,
and high noise condition hosts • Noise data easily extracted using Westinghouse Real Time Auto Analysis (RTAA) software • Select low (1.18V), mid (1.75V), and high (2.64V) P5 noise conditions as hosts
5
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
P5 Noise Distributions
6
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Bobbin Detection Summary – Manual Analysis • Bobbin data collected for the three donor morphologies • Injections performed locating flaws just above TTS (0.08 inch) and at ~0.3 inch above TTS
• For high noise tube at TTS, reliable manual detection judged at:
• 60%TW: Circumferential Groove • 20%TW: Axial Groove
• Tapered wear similar to axial groove performance
• 75%TW: Round Tapered Hole
Improved detection performance in mid and low noise tubes
7
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Example Bobbin Graphic: 60% Circ Groove, High Noise Tube, Injection at TTS
8
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Tube Integrity Impact – Manual Analysis • Volumes defined for the various flaw shapes • “Must detect” depths identified to support two operating
cycles between eddy current inspections based on additional volume growth • Identified reliable detection depths support the integrity analysis with postulated growth of two additional cycles up to the next scheduled eddy current inspection • But can we do better? Yes – with Auto History Compare (AHC)
9
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
EADS/AHC Setup • Field application included auto-auto analysis with AHC used as a tertiary analysis • EADS used as primary analysis pass
– EADS extraction algorithms applied to achieve equal detection as manual analysis
• AHC extraction algorithms provide improved detection (lesser depth) compared to EADS/manual
AHC reliable detection depths were about 10 to 15%TW less than manual analysis of ALFS mix
10
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Auto History Compare Results Summary • AHC reliably extracted flaw-like signals at less than manual • • • •
detection thresholds AHC currently only considers raw differential channels Mix channels showed similar performance but extraction algorithm is not verified • Future development to validate mix channels Application of AHC improved overall detection capabilities far beyond manual analysis, thus adding margin Conclusion: For this combination program, tube integrity is maintained up to the next eddy current inspection
11
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
MAPOD Investigation • Manual analysis
results show ARP of 1 is appropriate
12
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
POD Curve Comparison
13
Westinghouse Non-Proprietary Class 3
© 2016 Westinghouse Electric Company LLC. All Rights Reserved.
Field Implementation – Spring 2016 Outage • • • • •
EADS/AIDA primary/secondary analysis Tertiary AHC analysis of outer 5 peripheral tubes RTAA to screen noise at TTS (2.5V threshold) X-Probe inspection at locations with >2.5V of P5 noise For the spring 2016 outage, only 4 locations required testing with X-Probe (SG11: 95th percentile noise = 1.71V, SG12, 1.45V) • Thus, • Applied auto detection capabilities exceed manual analysis capabilities • Multiple methods/techniques produce a robust FO wear inspection •
