Field Experiences Partial Discharge Diagnosis on MV Cables Philipp Legler Technical Support Applications Engineer (Megger Site Germany) XVI. Szigetelésdiagnosztikai Konferencia (19. - 20.10.2016, Sümeg)
Introduction Reasons of Cable Faults Most of them are Workmanship Failures Predominantly caused by Service Contractors (Man Power & Time issued) Main Reasons are Time Pressure - need to do xx Joints a Day (mostly caused by Purchase Department) Joint Hole not large enough Especially in Cities: to much other Cables / Pipes Climatic Conditions Not having the Skills / Certificates 2
Case Study No. 1 Operator: DREWAG Test Object: 12/20 kv & only XLPE insulated Cable Length: 367 m Joint Type: Tyco MXSU 5131 Measurement History: 10.12.2007 29.11.2012 17.03.2015 25.03.2015 (only Phase L2) PD Diagnosis with OWTS 3
Case Study No. 1 PD Measurement Results of 1 st Diagnosis (2007) L1 L2 L3 Disturbance Level [pc] 140 136 123 PDIV [kv rms ] 15.6 6.0 12.0 PDEV [kv rms ] 18.2 18.9 19.6 PD max [pc] (PDIV) 535 481 381 PD avg [pc] (PDIV) 364 280 260 PD max [pc] (U 0 ) 323 3247 381 PD avg [pc] (U 0 ) 195 2477 260 PD max [pc] (1.7 U 0 ) 1921 13296 6505 PD avg [pc] (1.7 U 0 ) 1121 9155 4970 Frequency [Hz] 588.24 571.43 575.57 4
Case Study No. 1 PD Mappings U U 0 U 1.7 U 0 DREWAG just started with Diagnosis (less Experiences & kind of missing Conviction for PD) No final Decision further Investigation 5
Case Study No. 1 PD Measurement Results of 2 nd Diagnosis (2012) L1 L2 L3 Disturbance Level [pc] 128 121 134 PDIV [kv rms ] 19.2 6.0 14.4 PDEV [kv rms ] - 10.7 10.7 PD max [pc] (PDIV) 767 2430 278 PD avg [pc] (PDIV) 527 1521 149 PD max [pc] (U 0 ) 163 2647 115 PD avg [pc] (U 0 ) 115 1881 100 PD max [pc] (1.7 U 0 ) 767 4389 1010 PD avg [pc] (1.7 U 0 ) 527 3235 672 Frequency [Hz] 355.88 356.09 357.14 6
Case Study No. 1 PD Mappings U U 0 U 1.7 U 0 Even more clear Weak Spot Location, but still: NO final Decision further Investigation 7
Case Study No. 1 PD Measurement Results of 3 rd Diagnosis (2015) L1 L2 L3 Disturbance Level [pc] 64 70 69 PDIV [kv rms ] 13.2 6.0 13.2 PDEV [kv rms ] - 8.4 - PD max [pc] (PDIV) 201 778 263 PD avg [pc] (PDIV) 116 624 121 PD max [pc] (U 0 ) 201 3170 144 PD avg [pc] (U 0 ) 129 1285 92 PD max [pc] (1.7 U 0 ) 958 4935 337 PD avg [pc] (1.7 U 0 ) 862 3797 303 Frequency [Hz] 355.04 356.09 355.56 8
Case Study No. 1 PD Mappings U U 0 U 1.7 U 0 NO Final Decision further Investigation 9
Case Study No. 1 www.powertrans.com.au
Case Study No. 1 In the Meantime: Mainly Diagnosing the Cable Network Baseline Study with mostly no final Conclusion Only with Replacements on very important Cables BUT: PD Diagnosis performed on another Cable on the 7 th of May 2015 Clear concentration of PDs detected at several Location NO Action Breakdown on the 11 th of August 2015!!! Phase L1 & L2 (460 m) & Phase L3 (550 m) Turnaround inside DREWAG regarding Necessity of PD Diagnosis 11
Case Study No. 1 DECISION MAKING: Replacement of Joint at Phase L2 (210 m) 12
Case Study No. 1 13
Case Study No. 1 PD Measurement Results of 4 th Diagnosis (2015) PD-free after Replacement 14
Case Study No. 2 PD Testing done by ENSO Equipment: TDM-4540 Wave Shape Comparison Mixed MV Power Cable Length: 1335 m
Case Study No. 2 PD Measurement Results 16
PD Mapping Comparison Case Study No. 2 VLF Sinus DAC - 50 Hz Slope 0 U 1.7 U 0 0 U 1.0 U 0 VLF Sinus DAC - 50 Hz Slope
Case Study No. 2 TDR Diagrams L1 L2 L3 18
Case Study No. 3 985 m XLPE Cable Drum Operation Voltage: 5.8 kv 3 Customer made Weak Spots Competitor Comparison at Vienna Harness Racing Track with Centrix 2.0 & with integrated TDM-4540 Competitor could not find any Fault even not at higher Test Voltage 19
Case Study No. 3 20
Case Study No. 3 PD Measurement Results 21
Case Study No. 3 Voltage Wave Shape Comparison (PDIV & Weak Spot Location) PD Mapping for U 1.0 U 0 L1 DAC - L2 50 Hz Slope L3 VLF Sinus 22
Case Study No. 3 Voltage Wave Shape Comparison (PDIV & Weak Spot Location) PD Mapping for U 1.3 U 0 L1 DAC - L2 50 Hz Slope L3 VLF Sinus 23
Case Study No. 3 Voltage Wave Shape Comparison (PDIV & Weak Spot Location) PD Mapping for U 2.0 U 0 L1 DAC - L2 50 Hz Slope L3 VLF Sinus 24
VLF 0.1 Hz Sinusoidal vs. 50 Hz AC Case Study No. 4 Cable Information: 12/20 kv XLPE Cable (1992) Length: 1605 m PD Defect in Joint at 850 m Breakdown Channel starts at inner Field Grading Degrading of cable insulation 50 Hz AC PDIV = U 0 = 12 kv RMS VLF 0.1 Hz Sinusoidal PDIV > 2.5 U 0 = 30 kv RMS
VLF 0.1 Hz Sinusoidal vs. 50 Hz AC A few more detailed Pictures Case Study No. 4 Treeing at the inner field control section of the splice insulator Channel growth due to surface discharges at the splice insulation material Channel growth due to surface discharge at the cable insulation material Source: VWEW Infotag 2004, contribution E-On 26
VLF 0.1 Hz Sinusoidal vs. DAC Comparison Measurement 0.1 Hz & DAC Case Study No. 5 DAC Results PD detected & located L1 @ 4 km L2 @ 3.4 km & 3.6 km L3 @ 3.4 km 27
VLF 0.1 Hz Sinusoidal vs. DAC Case Study No. 5 L1 L2 L3 VLF 0.1 Hz Sinusoidal Results PD Mapping up to 2 U0 PD located at L1: PD-free L2 @ 3.6 km L3: PD-free
Case Study No. 5 Comparability & Reproducibility of PD Characteristics A Take Out from a Paper published in 2009 Source: Applied Voltage Frequency Dependence of Partial Discharges in Electrical Trees, N. Jäverberg, H. Edin, Proc. IR-EE-ETK, Stockholm, Sweden, 2009 29
Diagnostic Test: New laid Cable 12 kv 3-single Phase Cable XLPE MV Cabel (NA2XS(F)2Y) Length: 2,308 km (7 Joints) Switch Gear: Driescher (air-insulated) Case Study No. 6 Cable Length: 2,3 km 30
TanDelta Measurement on new laid Cable Case Study No. 6 31 Source: IEEE 400.2 (2013) Table G.1
Case Study No. 6 32
Case Study No. 6 33
Diagnostic Test: Service aged Cable 12/20 kv 3-single Phase Cable Mixed Cable (NEKBA & NA2XS(F)2Y) Length: 653 m (7 Joints) Switch Gear: Air-insulated Case Study No. 7 Cable Length: 2,3 km 34
TanDelta Measurement on service aged Cable Case Study No. 7 35
Case Study No. 7 PD Mapping Comparison 1.0 U 0 1.3 U 0 2.3 U 0 50 Hz Slope VLF Sinus 1.0 U 0 1.3 U 0 2.0 U 0
Summary Summary: PDIV of DAC & 50 Hz Slope Technology (VLF CR) with much lower compared to VLF Sin PD depends on Voltage Gradient over Time Lower the Change of Voltage Lower the PD Activity Results of VLF Sin difficult to interprete with PD Activities at Operation Frequency 37
38 Question Time
39 Let s have a break?
Contact Philipp Legler Applications Engineer Technical Support E-Mail: Philipp.Legler@megger.com Telephone: +49 (0) 35208 8429140 Mobil: +49 (0) 151 14089240 Köszönöm a figyelmet! 40
18/30 kv Medium Voltage Power Cable OWTS Diagnostic Measurement Joint Type: H&S Sucofit (African Manufacturer) Case Study No. 2 1 st PD Measurement: 13 th of November 2012 2 nd PD Measurement: 12 th of October 2015 3 rd PD Measurement: 23 rd of October 2015 41
Case Study No. 2 PD Measurement Results of 1 st Diagnosis (2012) 42
PD Mapping Comparison: OWTS vs. PDD U U 0, U 0 = 18 kv RMS Case Study No. 2 U U 0, U 0 = 18 kv RMS 43
Case Study No. 2 Final Decision after PD Measurement & Evaluation Excavation of Joint on Phase L2 at 1400 m due to PDIV below U 0 Joints at 2050 m with PDIV above U 0 under Observation New Measurement after 3 Years 44
Further Results of Joint Investigation Inclusion of oily Fluidity Case Study No. 2 45
Bad Shrinkage of Stress Control Tube Case Study No. 2 Assumption: high Humidity Rating inside Joint Body 46
Case Study No. 2 Bad processed Mastic Tape resulting in Air Inclusions Discoloration due to Effect of Heat (bad Connection) 47
Case Study No. 2 Clear visible Creases on Insulating Body due to bad Peeling Partial Discharge Trackings over entire Length 48
Case Study No. 2 PD Measurement Results of 2 nd Diagnosis (2015) 49
PD Mapping Comparison: OWTS vs. PDD U U 0, U 0 = 18 kv RMS Case Study No. 2 U U 0, U 0 = 18 kv RMS 50
Case Study No. 2 Final Decision after PD Measurement & Evaluation Excavation of Joint at 2050 m due to PDIV below U 0 Weak Spots at 720 m & 1400 m with PDIV above U 0 under Observation New Measurement after 3 Years 51
Case Study No. 2 Moisture on Sheath Wires 52
Case Study No. 2 Bad shrinked Field Control 53
Case Study No. 2 Clear visible Creases on Insulating Body due to bad Peeling Partial Discharge Trackings over entire Length 54
PD Mapping Comparison: OWTS vs. PDD (3rd Measurement) U U 0, U 0 = 18 kv RMS Case Study No. 2 U U 0, U 0 = 18 kv RMS 55
Case Study No. 3 How can be a Woudlouse responsible for measuring PD or can cause a Cable Fault? 56
Example from the Cabinet of Curiosities Case Study No. 3 57
Example from the Cabinet of Curiosities Case Study No. 3
Case Study No. 5 VLF 0.1 Hz Sinusoidal vs. 50 Hz AC Research project of Norwegian University of Science & Technology Topic: Condition Assessment of Wind Farm Medium Voltage Cable Joints (by Hans Lavoll Halvorson) a) b) c) a) Dimension Drawing b) Reduced Cross Section c) Insulating Teflon Piece d) Metal Cylinder d)
Measurment Results Case Study No. 5 At VLF 0.1 Hz Sinusoidal PD were only detectable at higher Status of Aging Process All Joints PD-free Below 50 Hz no PD detectable Still Differences regarding PDIV 60