Paralleling CTs

Copyright © SEL 2015

Paralleling CTs for Line Current Differential Applications: Problems and Solutions

David Costello, Jason Young, and Jonas Traphoner

Schweitzer Engineering Laboratories, Inc.

Outline

• Review two fundamental principles

♦ Tapping multiratio CT derates CT

♦ Paralleling CTs to 87L inputs decreases security

• Compare 87L designs

• Discuss two case studies

• Present solutions

Equivalent Circuit for Paralleled CTsExternal Fault

I1I2 XMXM

RL RLRCT RCT

IF IF

CT BCT A

Criteria to Avoid SaturationModified for 87L Applications

F B STDX

I • Z 1 7.5VR

Fault Current ComponentsSymmetrical (AC) and Asymmetrical (DC)

R

tM M LV V

i t sin t sin eZ Z

Effect of Tapping 1200:5 Multiratio CTC800 at 300:5 Versus 1200:5

Time to Saturate

SAT

F S BSAT

VX 1I R RRT ln 1

XR

External CT Summation ConnectionNumber of CTs Is Greater Than Relay Inputs

1 2

IR IL1 + IR

IL1

Line

••••••

Modern Relay With Multiple CT Inputs

1 2

IR IL1 + IR

IL1

Line

••••••

1 2

Traditional Slope-Based 87L

IOP = KIRT

Restraining Region

Operating Region

IRT

K0

IOP

K2K1

Alpha Plane Characteristic

Re(k)

Im(k)

Operating Region

RestrainingRegion

Rad

ius

–1

Angle

External Fault Detector

–

+

–

+

DIOPP

DIRTP

DIRTR

DIOPRDIOP

DIRT

∆IOPR

∆IRTR

Advanced Relay Characteristics

Re(k)

Im(k

)

–4

–3

–2

–1

0

1

2

3

4

–4 –3 –2 –1 0 1 2

Case Study 1 One-Line Diagram

Substation Alpha

Substation Charlie

262524

21 22 23

Bus A Bus B

16

33

Line 1

Line 2

CGSubstation Bravo

•••

•••

Raw Event Data From Breaker 51Asymmetrical CG Fault

Relay and CT Connection87L With Tapped and Paralleled CTs

21 22

Bus A

Line 2

Primary 87L / 21 / 67

Secondary21 / 67

2000:5 2000:5800:5 800:5

2000:5800:5

2000:5800:5

Filtered 87L Event DataICL 1.5 Cycles After Fault Inception

Case Study 1 Summary

• CTs are paralleled to 87L relay – for external fault, relay measures only error or difference between CTs

• CTs are tapped down – this derates CT performance

• External asymmetrical fault (reclose) causes saturation and dissimilar CT performance

Case Study 2 One-Line Diagram

4 7

5 8

6 9

13 32

14 33

15 34

•••••• •••

TCC-6TCC-7

Substation PO-1

TCC-8TCC-9

Substation L

Substation SK

Raw Event Data From Terminal 14-15

Breaker Failure Relay DataIndividual CT 14-15 Signals

Difference or Error Between CTsBreaker 14 and 15 Data

Filtered Data for TCC-8BG Fault Evolves to ABG, IA Spikes

Breaker Failure Relay DataIndividual Breaker 32 and 33 Signals

Difference or Error Between CTsBreaker 32 and 33 Data

Difference Between CTsAnother, Less Obvious Case

Case Study 2 Summary

• CTs are paralleled to 87L relays

• CTs are tapped down

• External BG asymmetrical fault inception causes TCC-9 CT error and misoperation

• BG to ABG fault transition causes TCC-8 CT error and misoperation

Solution 1 – Add Short DelayAlso Delays Internal Fault Trips

Solution 2 – Use Full CT WindingCase Study 2 CTs at 300:5 and 1200:5 Taps

Solution 3 – Expand Restraining RegionNot Effective, Reduces Dependability

Im

Re

8.06.04.02.0

Solution 4 – Use Advanced Relay

Conclusions

• CTs paralleled to 87L relays

♦ Breaker-and-a-half and ring-bus applications

♦ Relays with one three-phase current input

♦ Measurement of only error or difference between CTs for external fault

• CTs tapped down

♦ Is sometimes done to meter low load currents better

♦ Derates CT performance during faults

Conclusions

• Two case studies

♦ Both show 87L misoperations for external faults

♦ Both use externally paralleled and tapped CTs

• Solutions involving setting changes

♦ Empirically derived settings may not work for all cases

♦ Delay also slows tripping for internal faults

♦ Increase in slope or Alpha Plane restraint decreases dependability

Conclusions

• Solution – use maximum CT winding

♦ Increase tap to improve CT performance

♦ Use short delay (1-cycle maximum) with improved CT taps if CT saturation is still possible

• Solution – install modern relay

♦ Relay should have two CT inputs, external fault detector, generalized Alpha Plane, and more

♦ Replay of original data proves secure

♦ Consider using full winding CTs

Questions?