Lightning arresters

Construction and Condition Monitoring

Lightning Arresters

Selection of Surge Arrester

Voltage 400/√3 = 230kVDuring single phase to ground fault, voltage on healthy phase may go upto 1.4 to 1.5 times230 x 1.4/1.5 = 323 – 346Temporary O/V= 1.5 pu = 336kVLAs are available at 336kV, 360kV, 372kV and 390kV Higher ratings are selected taking into consideration of ageing of LA elements

LA Characteristics

Construction of disc

These are made by mixing ZnO with small amount of additives such as Bi2O3, CoO, Cr2O3, MnO and Sb2O3

ZnO grains (about 10μm dia) have low resistivity and surrounded by a granular layer which is a high resistive Oxide layer(0.1 μm thick). The two are strongly bonded.

LA

AS PER IEC-60099(5), TECHNIQUES FOR THE HEALTH MONITORING OF SURGE ARRESTERS IN SERVICE

1. TOTAL LEAKAGE CURRENT MEASUREMENT2. WATT LOSS MEASUREMENT3. INSULATION RESISTANCE MEASUREMENT4. THIRD HARMONIC RESISTIVE CURRENT

MONITORING.5. THIRD HARMONIC RESISTIVE CURRENT (THRC)

MONITORING WITH COMPENSATION FOR THIRD HARMONICS IN SYSTEM VOLTAGE.

THE MEASUREMENT PRINCIPLE

- HARMONICS ARE CREATED IN LEAKAGE CURRENT ON APPLICATION OF FUNDAMENTAL FREQUENCY, DUE TO NON LINEAR VOLTAGE-CURRENT CHARACTERSTIC OF SURGE ARRESTERS.

- THIRD HARMONIC IS THE LARGEST HARMONIC COMPONENT OF THE RESISTIVE CURRENT

EFFECT OF 3RD HARMONICS IN SYSTEM VOLTAGE

THIRD HARMONIC IN SYSTEM VOLTAGE CREATES CAPACITIVE HARMONIC CURRENTS WHICH AFFECT THE MEASURED VALUE

ERROR IN THE MEASURED VALUES MAY BE CONSIDERABLE

AS REPORTED, 1% THIRD HARMONIC IN SYSTEM VOLTAGE MAY INTRODUCE ERROR UPTO 100% IN THE MEASURED VALUE

Basic circuit for LA testing

ZnO type Surge Arrester

Equivalent Circuit

Equivalent Circuit

Rp = Non-linear resistance of the granular layer(108 Ώm for low electric field stress and 10-2

Ώm for high electric field stresses)C= Granular layer has a relative dielectric

constant between 500 and 1200 depending upon the manufacturing process.

Rs= Resistance of the ZnO grains with resistivity of 10-2 Ώm

Ageing of Metal Oxide Surge Arresters Normal Operating Voltage causes ageing of ZnO

Blocks Temporary O/V, Switching O/V and Lightning O/V

may cause overloading of all or some of the ZnO blocks

External Pollution may cause non-linear voltage distribution. Accelerated ageing caused by internal PDs

Moisture Entry through sealing gaskets, may lead to shorting of ZnO discs and overstressing of healthy ZnO blocks.

The degree of ageing depends on the nature/ quality of the granular layer.

The increase in Resistive Leakage Current may bring the arrester to Thermal instability and complete Arrester Breakdown

Failure of 400kV LA

Failed LA hanging with Bus pipe

Shattered pieces of LA stacks

Close view of shattered pieces

Damaged Surge Monitor and shattered pieces of LA stacks

Another failed LA

FAILURE OF LAs

FAILURE OF LAs

Failure of LA

Failure of Surge Monitor after failure of LA

Preventive Action taken by POWERGRID

Condition Monitoring of Surge Arresters to avoid sudden failures/ blasting leading to unplanned outages.

Failure investigations with manufacturers and improvements in manufacturing quality

POWERGRID practice for Surge Arrester Monitoring

Third Harmonic Resistive Current- 1 Yearly

Measurement as a routine test

Capacitance and Tan of each stack - SOS

Insulation Resistance of each stack - SOS

Third Harmonic Resistive Current

Make LA(0-2Yrs.)

LA (2-5Yrs.)

LA (5-10Yrs.)

LA (10-15Yrs.)

LA >15Yrs.

Alstom 10-25μA

10-40μA

20-100μA

40-125μA

50-150μA

Elpro 10-30μA

10-40μA

20-100μA

40-125μA

50-150μA

CGL 10-20μA

10-40μA

20-50μA

- -

Oblum 10-20μA

10-40μA

20-50μA

- -

Third Harmonic Resistive Current Measurement –contd.-

The limit for Third Harmonic resistive current has been fixed as 500 micro-amp.

About 35 nos. Surge Arresters have been removed from service based on Third Harmonic Resistive current.

Following tests were conducted on the removed LAs:

1. Insulation Resistance test at 5.0kV2. Dissipation factor at 10.0kV

Third Harmonic Resistive Current Measurement –contd.-

Insulation Resistance measurement gives good indication of moisture entry in the stack.

Capacitance and Tan Delta measurement indicates degradation of ZnO blocks.

In 90% of the cases, moisture entry has caused the failures of LAs in POWERGRID whereas in about 10% cases, it was because of degradation of ZnO discs due to various stresses.

Capacitance and Tan Delta Measurement

Capacitance and Tan Delta Measurement on LA having 560 micro-Amp

Stack Capacitance Tan Delta Remarks

A 85pF 0.06

B 89pF 0.065

C 91pF 0.05

D 270pF 1.12

Capacitance and Tan Delta Measurement on LA, 360kV having 595 micro-Amp

Stack Capacitance Tan Delta Remarks

A 80pF 1.0

B 40pF 0.2

C 40pF 0.2

D 40pF 0.2

Third Harmonic Resistive Current -765mic.amp.

IR value of Top Stack - 50 G Ohm

IR value of Middle Stack - 0 GOhm

IR value of Bottom stack - 100 GOhm

Moisture was found inside the middle stack.

Failures of LAs in POWERGRID since 1999

Make LAs failed/ blasted

LAs removed from service

Total % age failure

A 15 19 34 3.79%

B 12 16 28 2.8%

C 5 0 5 1.1%

D 2 0 2 0.5%

Failure investigations Investigations carried out involving all

the manufacturers viz M/s CGL, M/s Alstom, M/s Oblum and M/s Elpro.

Displaced Gasket leading to moisture entry

Conduction marks on packing rubber

Cracked packing material

Rusted Spring

Conduction of ZnO discs

Corrosion on the Gasket Area

Failures of LAs

Failures of LA stacks Failures of Surge Monitors Failures of LA stacks and LA monitors more in

rainy seasons. Many LAs removed based on Third Harmonic

Resistive Current Measurements whereas leakage current as indicated by Meter was still within limits.

Investigation on Surge Arrester Stacks

Defective LAs (based on THRC), were tested. Resistive leakage current measured to be high. Only 90kV could be applied and leakage current was 1400 micro-Amp. Following observations were made:

Rusting of various components due to moisture entry. Sealing gasket de-shaped. Conduction marks on the surface of ZnO discs and

heating of wedges was also observed.

Increase of third harmonic current during service was mainly due to moisture entry and then conduction over ZnO discs leading to overstressing of the healthy stacks.

Investigation on Surge Arrester Stacks-contd.-

Manufacturer-A

5 Units of LA stacks tested for moisture entry at 1.0m, 1.5m and 2.0m. Following tests conducted before and after dip test:

1. Partial Discharge Measurement2. Resistive Leakage Current Measurement3. Reference Voltage4. Megger Test

All tests passed successfully indicating no moisture entry in the stack. Based on this, it was agreed to carry out dip test at 1.5 m as a routine test in future.

Investigation on Surge Arrester Stacks-contd.-

Manufacturer - B All the stacks had shown very high value of

resistive current upto 2000micro-amp. All the stacks except bottom stacks were

defective due to entry of moisture through cracked copper tube used for dry air filling and also from flat gaskets used in these stacks.

All the inner MS components were found rusted. Gaskets were also found cracked. Slippage of gasket in one case was also

observed.

Improvements – Surge Arrester Stack

“O” rings/ elliptical cross sections (neoprene, butyl or equivalent) is a better sealing option in comparison to flat gasket with no groove.

Water dip test to be conducted at a minimum depth of 1.5m from top of the Arrester for 30 minutes to be followed by routine electrical tests PD, Reference Voltage, Residual Voltage and IR measurements(5kV Megger). For IR acceptance criteria to be within ±10% of the pre-dip values.

Improvements/ Modifications – Surge Monitors

The Terminal for connection to be provided at the bottom. This will help in avoiding the moisture entry.

Surge Monitor enclosure to be tested for IP-66 for proving efficacy of sealing arrangement. Manufacturers are being insisted for IP-67.

Surge Monitors be subjected to dip test as a Routine test. Dip test at 1.5 meters for 30 minutes.

Outcome of Joint Failure Investigations

1. Moisture entry through sealing system/gaskets has led to degradation of discs and consequent increase in THRC.

2. Accelerated degradation of the ZnO discs due to manufacturing defects/process problems

3. Third Harmonic Resistive Current Measurement is technique for precise monitoring of health of the Surge Arresters.

Conclusion Failures are mostly due to moisture entry. Third Harmonic Resistive Current Measurement

technique is very effective in detecting defective/ aged Surge Arresters.

Moisture entry in the stacks can be detected by IR measurement.

Degradation of ZnO blocks can be detected by Capacitance and Tan Delta measurement

Dip test at manufacturers works shall help in identifying the defective LA stacks.

O rings are better than flat gaskets for sealing.

Thank You for your kind attention please