High-Impedance Differential Applications With Mismatched CTs Russ Franklin, Alliant Energy Hossein Nabi-Bidhendi, ABB Inc. Michael J. Thompson and Héctor J. Altuve, Schweitzer Engineering Laboratories, Inc. 2017 Alliant, ABB, and SEL Introduction High-impedance differential (87Z) is desirable for many reasons It requires matched ratio CTs Often, this is a limitation Options for using mismatched CTs have compromises Alliant set out to validate easiest option 87Z 2,000 W MOV 87 50 1
Outline Why mismatched CTs? 87Z fundamentals Summary of options High-current lab testing Conclusions Regulated Utility Profile 2016 Data IPL WPL Total Electric customers 489,867 468,451 958,318 Wisconsin WPL Electric sales (000s MWh) 15,970 13,868 29,838 Summer peak demand (MW) 2,996 2,681 Iowa IPL Gas customers 224,420 187,338 411,758 Gas sales (000s Dths) 63,388 61,840 125,228 Operating revenues (millions) $1,820 $1,459 $3,320 2
10 Years Estimated Capital Expenditures Updated November 2016 2016 2020 $6.8 billion 2021 2025 $4.5 billion Other Other Gas Generation Gas Distribution Renewable Generation Gas Distribution Other Generation Gas Generation Electric Distribution Other Generation Electric Distribution Electric Distribution Strategy Grid Enhancements and Customer Need Reliability Distributed generation/renewable integration Integrate planning with generation and transmission Increase remote monitoring and control Eliminate lower voltage systems Increase life extension and rebuild rate Resiliency improvements including underground/hardening Customer Focused Customer data availability Utility as trusted advisor for technical solutions Advanced metering infrastructure in Iowa 3
Why 87Z? High performance Virtually no limit to number of branch circuits Simple CT wiring Extremely simple settings calculations Multiple owners Standardization Outage constraints Budget constraints Equipment failure Upgrades Why Mismatched CTs? 4
87Z Fundamentals Summing Junction 87Z 2,000 W 86a MOV 87 CT1 CT2 V S CT3 CT4 1 2 3 4 50 Saturated CT Path Differential Path Simple Settings Calculations Simplifying Assumption Complete Saturation I CT123 I CT4 I CT123 I CT4 CT 123 V CT123 V JCT V CT4 CT 4 V CT123 V JCT V CT4 CT 123 CT 4 V+ V+ Set point 0 VCT123 V JCT V CT4 0 V CT123 V JCT V CT4 V Summing Junction Voltage No Saturation V Improved margin Summing Junction Voltage Complete Saturation 5
Internal Fault Relay sees same signal regardless of fault current Rectangular waveform Height = MOV clamping voltage Width C Class (Iron) Volts 2000 1500 1000 500 0 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 Cycles Operate Quantity (Voltage) for Internal Fault Rectangular pulse ½ cycle COS filter Magnitude C100 = Pulse too narrow C 4 V C4 8 V C800 > 800 V 6
Tapping a CT to Match Ratios X1 X2 X3 X4 X5 240T 160T R W GND Turn-to-turn and turn-to-core voltage stress Lead-to-lead and lead-to-ground voltage stress Conductors in conduit Terminal block in breaker cabinet Wiring below this is protected by MOV To summing cabinet and relay panel 400T CT tapped at 240T 1.5 kv clamping voltage Insulation stress 2.5 kv Peak across winding 1.5 kv Peak to ground Consequences Damage Failure to trip Options to Mitigate Risk of Insulation Failure Apply MOV across the full winding Use tapped CT or auxiliary CT as autotransformer Change the relay Lower MOV clamping voltage Percentage restrained All options have drawbacks 7
Assessing Risk of Simply Tapping the CT Examine standard IEEE C57.13 Applied voltage routine test (2.5 kv Peak ) Induced voltage routine test (2.263 kv Peak ) Interturn overvoltage type test (3.5 kv Peak ) Check routine HiPot test levels on breaker Check with manufacturer Test CT under realistic conditions Utility Source Standard CT Test CT LeCroy Oscilloscope X1 X2 X1 X2 X3 X4 X5 GND High-Current Laboratory Tests Meter 87Z 8
High-Current Laboratory Tests 3 new CTs 5 shots minimum for each 12-cycle duration Retest afterward Voltage Waveforms 9
Voltage Waveforms X5 X5 I Dischg. CT X3-X5 X3 V X3-X5 I Sec X3 I Sec Equivalent Circuit CT X1-X3 V X1-X3 X1 X1 10
Conclusions 87Z is a popular scheme It requires matched ratio CTs Methods to accommodate mismatch have drawbacks Simplest method is to tap the CT No complex calculations Risk of insulation failure Manufacturers unwilling to certify their CTs Conclusions Alliant decided to assess risk in high-current laboratory Three C400 400T CTs One 87Z relay with 1 kv clamping voltage At least five shots each CTs showed no sign of degradation after the tests Mixing 240T and 400T CTs has acceptable risk This combination is not necessarily worst case 11