Flashover tests under ice conditions need to be performed indoors to secure a reasonable repeatability needed to extract dimensioning rules from the test results. However, test chambers are often relatively small, which allows only small scale testing. The described test procedure allows testing of full scale insulators up to and including 800kV AC and DC. This study shows the non-linearity of the flashover voltage per meter insulator length. One should therefore be careful to extrapolate dimensioning rules obtained from small scale ice tests.

The present preliminary results indicate a non-linearity for suspension insulators of about 15% in the range of insulators lengths from 1 to 4 meters. More work is needed on this area to quantify the nonlinear behaviour more exactly for different types of insulators and applications

1. INTRODUCTION

Flashovers on high voltage outdoor insulators caused by ice have been reported in many countries. To combat this problem, the insulators have to be selected and dimensioned to withstand the icing conditions. Many contributions have been made in this area using different icing test methods [1]-[9]. Icing tests are complicated because of the sensitivity of the test results on the test conditions. The variations need to be minimal and several tests under identical conditions need to be performed to increase the accuracy of the outcome of the tests. It is therefore desirable to perform the tests in a controlled climatic chamber instead of an outdoor test yard. However, limited by the size of most such climatic chambers, most of the test results were obtained for insulator lengths shorter than 1 meter. However, many of the practical problems in service are related to larger insulators. If the relation between the flashover voltage and the insulator length is non-linear, one should be careful to use the results obtained on short insulators for the dimensioning of insulators in distribution and transmission networks. In this investigation, a first attempt has been made to examine the correlations between the AC withstand voltage under ice conditions and the insulator length.

2. GENERAL LABORATORY DATA

The STRI of climate hall in which the ice tests are performed is characterized by the following:

Diameter 18 meters Free height 20 meters

Voltage 860 kV AC, 1200kV DC, 1600kV

SI, 2200kV LI

Temperature -10 to +40oC controllable

The size of the climate hall allows large test objects to be tested under their operation voltage and allows more than one test object to be tested at the same time. The wall of the cylindrical test hall is well isolated, which allows ice and snow tests to be performed independent of the season of the year.

Water cooling tanks, water spraying systems and a measuring system are installed in the climate hall. By cooling down the wall, the temperature of the air can reach -10 ^oC. By adjusting the temperature of the water, pressure of the water spray and the distance between the nozzle and the test object both ice and snow conditions can be simulated [9].

3. TEST DESCRIPTION

To date, no standard methods for evaluating the withstand voltage of insulators under ice and snow conditions is devised [10]. Therefore, a careful analysis of the test parameters is needed before performing such tests. Icing tests are normally time consuming, and energy consuming why it is necessary to minimise the total test time by minimizing the number of tests.

To investigate the linearity of the flashover voltage as a function of the insulator string length, three lengths were tested. This gives 3 points (plus the origin) on the curve U(FO) as a function of the length.

4. CONCLUSIONS

1. Flashover tests under ice conditions need to be performed indoors to secure a reasonable repeatability needed to extract dimensioning rules from the test results.

2. The described test procedure allows testing of full scale insulators up to and including 800kV AC and DC.

3. The flashover voltage per insulator length is a non-linear function of the insulator length. The present results on cap & pin suspension insulators indicate a non-linearity of about 15% in the range of insulator lengths from 1 to 4 meters.

4. Different insulator types and applications are expected to have a different non-linear performance under ice conditions

5. One should be careful to extrapolate dimensioning rules obtained from small scale ice tests.

6. More work is needed to establish internationally accepted test methods and dimensioning rules for insulators under ice conditions

From: internet