Surge arresters were installed at the Kingston substation in 1928.
Almost 100 years ago, sets of electrodes (rod, sphere, or pipe) were used to limit surges on equipment (Sakshaug, 1991). Some of these systems, particularly the broken piping, may still be in service today. However, the voltage thinning characteristic of the gap versus time before surge does not match well with the resistance characteristics versus the front of most insulators; that is, it is difficult to coordinate.
The next evolutionary step was to add a resistive element in series with the gap, in order to limit the following current after a stop discharge operation. Hopefully, the current limiting would allow the stopper to release that current, instead of relying on a circuit-breaker or fuse nearby. At the same time, the voltage of the resistor during discharge should be low enough that no excessive voltage can appear on the protected equipment.
These competing requirements have led to the use of expensive and complicated nonlinear resistive elements, some involving both solid and liquid materials with high float loads.
From around 1930, silicon carbide (SiC) was used for non-linear resistive elements, leading to much better protection characteristics. Since SiC would conduct significant current at rated voltage, it was necessary to provide an ignition gap that prevents conduction at rated voltage. After discharge of the arrester, these spaces must be sealed against the current, otherwise, the arrester will have thermal failure. In the mid-1950s, active vacancies were created for SiC surge arresters.
These active lacunae contain auxiliary elements which:
Preionize the sparkover space to achieve better levels of surge protection and
Extend the power arc and move its attachment points for better interrupt performance.
SiC surge arresters have been successfully applied on transmission systems up to 345 kV, but some limitations have arisen with respect to surge protection, energy discharge capability, and decompression capability.
Having both vacancies and silicon SiC blocks, the height of the stopper increased until it reached the point where it was difficult to release the built-up pressure in the event of failure, which limited the discharge value pressure of the shut-off device.
Due to their discharge characteristics vs frequency or time to time, SiC surge arresters have been optimized for lightning. They were less effective for steep edge surges and slow switching surges.
In the mid-1970s, metal oxide arresters were developed into commercial products (Sakshaug et al., 1977). Metal oxide blocks are much more nonlinear than silicon carbide, so they only conduct a few milliamps at rated AC voltage. It eventually became possible to do without empty spaces altogether, although older models made use of them
Metal oxide arresters have several major advantages over previous silicon carbide arresters:
Active deviations are not necessary, which improves reliability
Metal oxide can discharge much more energy per unit volume than silicon carbide
Metal oxide provides better protection than silicon carbide over the surge wavefront range
The decrease in the height of the limiter, caused by the elimination of empty spaces, leads to higher pressure relief rates
Virtually all new applications will use metal oxide arresters. The metal oxide enabled new applications, such as series capacitor protection and overhead line switching surge control, which was not possible with silicon carbide.
However, many silicon carbide arresters are still in use. Some researchers have found high failure rates of the silicon carbide arrester, due to moisture penetration, after several years of service on medium voltage distribution systems. This experience does not necessarily apply to surge arresters in substations. If such problems arise, it would make sense to routinely replace silicon carbide arresters on a system. Otherwise, assuming the initial request is correct, the old silicon carbide arresters could remain in service.
The general use of surge voltage surge arresters. The dotted design (1a) applies to silicon carbide, while the continuous design (1d) applies to the latest generation of metal oxide. A manufacturer used the shunt gap (1b) in the first metal oxide arresters. In steady-state, the two non-linear elements withstand the nominal voltage, thus reducing the current slightly. During a surge discharge, the shunt gap would produce a bypass effect to bypass the smaller section of metal oxide, thereby reducing the discharge voltage and providing slightly better protection Another manufacturer used the gap in series with capacitive dimming (1c) in the first metal oxide arresters. In steady-state,
During a sudden discharge, the gap forms “Immediately” due to capacitive gradation. The latest generations of metal oxides do not need these shortcomings, although it is planned to use them to achieve specific purposes (eg coordination of surge arresters, resistance to temporary overvoltages). The surge protector must be installed on something, such as a transformer tank or a pedestal. It must also be connected to the protected system, usually via wire or cable. We will see later that these connections have important effects on overall protection, especially for strong peaks.
The pedestal and lead, length and location, should be considered part of the overall installation of the surge arrester. On distribution systems, a ground cable disconnector is often used with the surge arrester. If the arrester fails and conducts current stably, the disconnect switch will detonate and disconnect the base of the arrester from the ground. It should happen in about 1 second or faster.
The surge arrester can then remain connected to the system until maintenance personnel has the opportunity to replace it. No circuit breaker or fuse should be used to isolate the faulty surge arrester; if the surge arrester is the only thing that has failed, no customer should lose their electrical service. Of course, the surge arrester does not provide any surge protection during this period with its ground wire disconnected. There would be a clear visual indication that the ground wire has been disconnected; it will be “suspended” below the stopper. Regular visual inspections are necessary to maintain surge protection whenever earthing switches are used.
Many surge-protectors installed in substations or industrial facilities have been fitted with “‘surge meters”. “
These are accessories to be installed in the connection of the earth wire of the surge arrester. Two functions can be provided:
A steady-state current meter calibrated in mA. If this current increases over time, it may indicate thermal damage to the surge arrester. However, the presence of harmonics or external leakage currents would complicate the assessment.
A counter indicates the number of surge current discharges above a certain threshold, which may depend on the frequency or the time before. Even if the count is accurate, it does not mean that the discharge voltage has reached a particular level during these events.
To effectively use surge counters, it is important to follow the readings regularly, beginning with commissioning.