# Thermal Characteristics of a 330 KV Surge Arrester

Surge Arresters with ElecNet

Surge arresters are used in a number of industries as protective devices against a variety of electrical events such as high voltage lightning impulses. The design choices of various materials and components for lightning arresters may depend on their thermal characteristics among many other factors. To demonstrate this, the thermal analysis of a 330 KV metal oxide arrester is considered in this example. The adiabatic heating of the varistors due to a lightning strike and the subsequent temperature distribution in the device as a function of time are presented here.

## SURGE ARRESTER MODEL

A simple surge arrester comprising various essential components is shown in this figure. The metal Oxide varistors, spacers, electrodes, base and the porcelain casing are clearly identified in the model.

View larger image

## MODELING a LIGHTNING IMPULSE

Typical thermal tests of surge arresters comprises the application of (multiple) voltage impulses followed by cool down period(s). In this study, as an example, a 330 KV impulse is applied to the top electrode of the model over 50 milliseconds. This could represent a typical lightning strike. The voltage profile representing such an event is shown in this figure.

View larger image

## ELECTRIC FIELD DISTRIBUTION

The electric field distribution at the end of the impulse at 5.0E-5 seconds is shown here.

View larger image

## TRANSIENT FIELD DISTRIBUTION

The equipotential lines as well as the electric field magnitude distributions as a function of time are shown in this video.

Watch Video

## COUPLING for THERMAL ANALYSIS

Assuming the process is adiabatic, the ElecNet/ThermNet coupling of the model is established to study the thermal characteristics of the device. The coupling is seamless and the solver interface between ElecNet and ThermNet is shown here. Adiabatic processes are defined as isolated events the duration of which is specified at the interface. Also, a delay of 400 seconds is applied before the application of the lightning impulse. As such, in subsequent results, the temperature changes are tracked from 400 seconds onwards.

View larger image

## ELECTRIC THERMAL SIMULATION RESULTS

Thermal processes including conduction, convection and radiation can be modeled in ThermNet. This is done by defining the thermal conductivity and specific heat capacity of the materials and by setting up convective link boundary conditions on model surfaces. Note that in ThermNet a number of boundary conditions can be applied such as fixed temperature, heat sources etc.

The key questions for coupled electric-thermal analysis is to determine the maximum temperatures reached in various parts of the model as well as post event cool down modeling. In the model setup section, the setting of the convective boundary condition was described. This figure shows the model temperatures ate times 400, 1000, 20000 and 30000 seconds. The temperature distribution as well as thermal variations give expected results.

View larger image

## TEMPERATURE vs. TIME

The results plotted here show the temperature variation in the middle of one of the varistors as a function of time. Expected cooling rates are verified by this plot.

View larger image

## SURGE ARRESTER TEMPERATURE DISTRIBUTION

An animation of the temperature distribution in the model as a function temperature is shown here.

Watch Video