C-3 2012 IEEE International Conference on Condition Monitoring and Diagnosis23-27 September 2012, Bali, Indonesia

Investigation Experience Due To High Number ofCVTs Failure (Part: 1)

Rikardo SiregarPenyaluran dan Pusat Pengatur Beban Jawa Bali

PT. PLN (Persero)Jakarta, Indonesia

rikardo_siregar@pln-jawa-bali.co.id

Abstract-Base on data, PLN P3B Jawa Bali has about 4000CVTs. More than 70% are installed in a system of 150 kV, 13%in 500 kV and the rest in 70 kV which are coming from about 30different trademarks. In 2011, at least 36 CVTs failed. Most ofthem reported with low voltage or zero volt on secondary side.There is no any specific information was reported to explain theroot cause. The failures happened from different trademarks andwill need a big efforts to find the main problem. To have a goodresult, RCM concepts will be used to describe the evidences,causes and consequences of the failures and the supporting ofstatistical failure.

Keywords-component; failure rate, root cause, statisticalfailure.

I. INTRODUCTION (HEADING 1)Every utility so does PLN P3B Jawa Bali really concern

about the reliability of all of the apparatus to keep thecontinuity of the electricity steady. Some disturbance could behappened and some time it came from the apparatus, notbecause of external fault. No matter if it is internal or externalfault, the one we need to fmd is the root cause of the fault.

At this moment, fault in capacitive voltage transformer(CVT) will be concerned. Some data have been recorded due tothese failures but less information about the reason. The mostreported cases are zero (0) volt on secondary side.

The first step in revealing the problems is to describe thefailure, by collecting the information before failure, anomalyand others. All the process are taken will be described in thispaper.

II. RCMFoR CAPACITIVE VOLTAGE TRANSFORMER(CVT)

A. RCMBasically, Reliability Centered Maintenance (RCM) is

applied to prevent failures and optimize the maintenanceprocedure. It means, we can use this method to describe andreveal all facts and possibilities of failures besides improvingthe ability to manage the maintenance. All knowledge which iscoming from this process will help the utility to prevent thesame failure in the future.

Identify applicable sponsor/s here. If no sponsors, delete this text box.(sponsors)

978-1-4673-1018-5/12/$31.00 ©2012 IEEE 586

Achmad SetiawanPenyaluran dan Pusat Pengatur Beban Jawa Bali

PT. PLN (Persero)Jakarta, Indonesia

Achmad.setiawan@pln-jawa-bali.co.id

Several basic concepts that could be applied for the CVT[1]:

• What are the function and associated performancestandards of the asset in its present operating context?

• In what ways does it fails to fulfill its function?

• What happens when each failure occurs?

• In what way does each failure matter?

• What can be done to predict or prevent each failure?

• What should be done if a suitable proactive task cannot befound?

The answer of each question above will lead us to fmd thefailure modes of the CVT, the causes of the failure, thecriticality of each failure mode, and the corresponding action toprevent the failure modes. [1].

B. Capacitive Voltage Transformer (CVT)In describing the electrical apparatus, it should be

mentioned the function and the functional failure. How do wejustify the functional failure is depend on the expectation of theuser from the equipment. Then, what is the CVT defmition?

CVT is an instrument transformer that has the mainfunction to convert the transmission class voltage tostandardized low voltage and easily measurable values used formetering, protection and controlling the high voltage system.The CVT functional failure is a condition where the equipmentcould not be able to convert the transmission class voltage levelto the low voltage level as the user want.

This picture below (Figure 1) is an example of CVT'swiring diagram. C1 and C2 are the main capacitive voltagedivider. One of the links (usually C2) is used forcommunication link. In the main medium voltage transformer,some of the CVT are applied with over voltage protector(visually, look like a small fuse). The main function of thisequipment is to protect the medium transformer from overvoltage.

Figure 1. CVT Diagram with C1 and C2

III. FAILURE IDENTIFICATION OF CVTAll failure should be able to be explained, because it's

important to avoid the same failure in the future, especially forCVT ini this case. We need to fmd the root cause for everyfailure. In case of investigation, action will be taken fordifferent trade mark, because base on data, failure is dominatedby certain trade mark. In this chart below, we can see numberof failure for different trade mark.

Figure 2. Percentage of Failure of Different Trade Mark

Based on chart above, it is clearly seen that the trade mark"A" dominated the failure with the average age of theinstallation is 15 years. It is about 67% and the second is trademark "B" about 14%. By showing that data above, it does notmean that the failure happened because of the technology orengineering. There are others possibilities, likes operationalhistory, installation and others. The implementation of RCMmethod hopefully could help the utility to fmd the real causes.

Another thing that put in to consideration about this failureis the number of failures. Since year of 2007 until 2011 thereare more than 100 CVT failures for different voltage level andlooks going up for year after year. Most of the failure happenedin 150 kV. The statistic failure of the CVT can be seen in thispicture below.

587

Figure 3. Number ofCVT Failure by Year

A. FMEA and FMECAIn the above explanation about the basic concept of RCM,

one of the basic question actually has the same meaning asFMEA (Failure Mode Effect Analisys). It means extract allpossible condition that could make the failure.

In PLN P3B Jawa Bali, FMEA and FMECA have alreadydefined. All kind of possible failure already decribed. It couldgive a benefit to the engineer to learn what condition couldmake one failure. In this paper, we are not going discus about itin detail.

B. Investigation for Trade Mark "A"According to RCM method, the failure should be explored

completely to find the root cause. The first step should beoperational histories then location of installation andinvestigation after failure. At this moment, we only reportedinvestigation after failure. Others will be done in the next part.

The steps of the investigation are identifying the technicaldata, visual inspection (out side or in side of the mediumvoltage), electrical measurement (capacitance of the maincapacitor, tangen delta, ratio of the winding and insulationresistance).

The most interest thing of the investigation are the presenceof water at the base box (the function of main seal is fail) andthe over voltage protector burned as shown in this figurebelow.

Figure 4. The Spare Gap Burned

As already explain before that the main function of thespark gap is to protect the main medium voltage of the CVTfrom the over voltage, switching voltage or impuls voltage. We

can assume that the burned spark gap failed due to overvoltage, no matter switching or impulse The question is, howoften the over voltage happened. To answer this question, weneed the over voltage data. At this time that data is notavailable. Besides we need to fmd out if burned spark gapvisually recognized for all failures.

C. Future WorksBased on the above explanation, the CVT failed after the

spark gap faunction fail, but could not be explain more howcome this spark gap fail. It will be fail after so many overvoltages, in the system. It is still questionable and should beproved.

To fmd the answer, measurement sould be done on thesystem, especially substation with high number of failures.

588

One other thing that should be answered is the burned sparkgap. Is it happen for all failures or not.

REFERENCES

[1] Diego R. Morais, Jacqueline G. Rolim, Joni Coser and Hans H. Zurn,"Reliability Centered Maintenance for Capacitor Voltage Transformer",9th International Conference on Probabilistic Methods Applied to PowerSystem, Sweden - June 2006.

[2] S. Coatu, D. Rucinschi, M. Costea, A. Marinescu University'Politehnica" of Bucharest, "Characterization of the Capacitive VoltageTransformers as Coupling paths and Measuring Device for the HighVoltage Transients in a Substatio