Withstand Tests More than Meets the Eye Nigel Hampton, Joshua Perkel, JC Hernandez, Miroslav Begovic, John Hans, Ron Riley, Pete Tyschenko, Christian Linte, Frank Doherty, George Murray, and Leeman Hong NEETRAC Georgia Institute of Technology, ECE Commonwealth Edison PEPCO Consolidated Edison Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 1
Acknowledgements • Thanks to CDFI Participants for the data presented here. • This material is based upon work supported by the Department of Energy under Award No DE-FC02- 04CH1237 and CDFI. • Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Department of Energy. Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 2
Outline • Introduction/Motivation • Length Issues • Withstand Testing Process – Ramp Up – Hold • “Ramp Up” Diagnostic Features • “Hold” Diagnostic Features • Recommendations • Conclusions Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 3
Introduction • Withstand tests are frequently used by utilities who employ diagnostic tests. – As of 2006, approx. 33% of CDFI member utilities employing diagnostic tests use withstand techniques. • Withstand tests have been defined as “Pass/Fail” only. • Utilities maintain records that are much more detailed. • Cable Tested in the last five years: > 4495 miles. Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 4
Motivation Utility records from withstand tests contain much more information than the result of the test. – Test Voltage (including voltage at failure) – Time on Test – Segment Length – Segment Insulation – Segment Location – Failed Equipment Type How to use this information in a diagnostic manner? Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 5
Results of Withstand Tests 618 Conductor Miles Tested (one utility feeder cable system) Percentage of Tests [%] 100 No evidence of cascading failures for IEEE 400.2 testing practices. Retests 0 1 2 3 Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 6
Length Effects • Withstand tests can be used to test long lengths of cable. • Comparison of withstand failure on test rates must include length adjustments. 2000 ft. Failure 500 ft. 500 ft. 500 ft. 500 ft. Censored Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 7
Withstand Test Process Voltage RAMP UP HOLD t Test t = 0 Time Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 8
Withstand Test Process Voltage RAMP UP HOLD t Test t = 0 Time Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 9
Why “Hold” and “Ramp Up” Phases? Focus is generally on the “Hold” phase but “Ramp Up” is important too. 100 Survivors [% of Total Tested] 80 60 40 Large percentage of failures Voltage occur during Ramp Up portion. Insulation [kV] 20 Extruded 16 PILC 16 PILC 22 XLPE 22 0 0 1 2 3 4 5 Time on Test [Minutes] Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 10
“Ramp Up” Phase Diagnostic Features Getting up to test voltage is half the battle Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 11
“Ramp Up” Data Voltage RAMP UP HOLD V f t Test Time Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 12
Ramp Up Failures • Failures during the “Ramp Up” phase have accounted for as much as 70% of the total failures on test. • The failure mechanism during “Ramp Up” phase is different from the “Hold” phase mechanism. Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 13
“Ramp Up” and “Hold” Failure Mechanisms 20 Failures on Test [% of Total Tested] 15 10 “Hold” Failures 5 Two different failure mechanisms “Ramp Up” Failures 1 0.1 0.5 1.0 5.0 10.0 50.0 100.0 Time on Test [Minutes] Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 14
Weibull Curve – “Ramp Up” Failures (VLF) 10 Failures on Test [% of Total Tested] Evidence of two failure modes. 5 3 2 IEEE 400.2 Recommended Test Voltage 1 1 5 10 20 30 40 50 Voltage [kV] Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 15
Weibull Curve – “Ramp Up” Failures (DC) 80 70 Two failure modes present during “Ramp 60 Failures on Test [% of Total Test] Up” portion that are voltage dependent. 50 40 30 20 10 In this case, 60% of the tests produced a failure before reaching the target test voltage. 0.1 1.0 10.0 Voltage [U0] Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 16
“Hold” Phase Diagnostic Features Time is everything Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 17
“Hold” Data – Failure During Test Voltage RAMP UP HOLD Time on Test t Test Time t f Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 18
“Hold” Data – Test Passes Voltage RAMP UP HOLD Test Passed t Test Time t f Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 19
Analysis of Times on Test PASS NOT PASS Segments Segments Censored Failure Times Times on Test Use data to construct Weibull curves for different areas These curves represent a diagnostic feature. Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 20
Length Effect on Failures on Test ≈ 900 Conductor Miles Length Length Length 20 20 20 Adjustment Adjustment Adjustment 17.2% 1000 Feet 1000 Feet 1000 Feet Failures on Test [% of Total Tests] Failures on Test [% of Total Tests] Failures on Test [% of Total Tests] 500 Feet 500 Feet 500 Feet NONE NONE NONE 10 10 10 5 5 5 4.1% 3 3 3 2.4% 2 2 2 1 1 1 30 0.1 0.1 1.0 1.0 10.0 10.0 100.0 100.0 0.1 1.0 10.0 100.0 Time on Test [Minutes] Time on Test [Minutes] Time on Test [Minutes] Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 21
System Subset (Length Adjusted) 30 Curve includes tests from four areas of a single utility. 20 Failures on Test [% of Total Tested] 13.5% 10 5 4.6% 4.6% of tests led to failure during 3 “Ramp Up.” 2 30 1 1 0.5 1.0 5.0 10.0 50.0 Time on Test [Minutes] Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 22
Weibull Curves by Area (Length Adjusted) Area 40 1 Failures on Test [% of Total Tested] 2 3 30 29.0% 4 22.9% 20 17.8% 10 10.0% 30 1 10 100 Time on Test [Minutes] Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 23
Separation of Regions 35 Failures on Test [% of Tested] 30 25 Area 1 is clearly different 20 from the others. 15 10 Overall 1 2 3 4 Area Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 24
Recommendations • Defined “Ramp Up” procedure should be employed with each test. • Detailed records should be maintained. Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 25
Conclusions • Withstand tests consist of two phases: – “Ramp Up” – “Hold” • A significant percentage of failures occur during “Ramp Up” phase. • Useful diagnostic features can be derived from withstand data. – Voltage at failure (“Ramp Up”) – Time on test at final test voltage (“Hold”) • Results can be used to prioritize areas for action. Fall 2008 ICC Meeting - Subcommittee F: Field Testing and Diagnostics 26
Recommend
More recommend
