Polymeric-housed surge arresters have already completely taken over the market at distribution voltages but this is still not the case for HV applications where they share the market about equally with porcelain. Nevertheless, the technology of polymeric HV arresters – first introduced around 1990 – is widely considered mature. In view of the economic and other benefits they offer, it is only reasonable to expect a steady increase in their application in coming years.
This INMR article from a decade ago is based on a contribution by Prof. Volker Hinrichsen of the Technical University of Darmstadt in Germany but still offers a valuable classification of the design concepts and variants for polymeric HV arresters. It also discusses the relative performance characteristics of each with respect to most service requirements.
Though not being the direct focus of much scientific or engineering interest, arresters have nevertheless undergone considerable technical change over the past three decades. Indeed, a modern surge arrester has relatively few things in common with the arrester used for that same application only 35 years ago. The first and perhaps most significant technological change was the relatively quick transition from gapped silicon-carbide to gapless metal-oxide (MO) arresters. This development started at the end of the 1970s and within only a decade MO designs had established themselves as the state-of-the-art due to their technical and commercial benefits.
MO arresters offer low protection levels, high energy absorption capability, stable operation (even under severe pollution) and lifetimes which can easily exceed 30 years. There have been no general indications of any severe degradation of the MO material even after 20 or more years’ service. Moreover, general failure rates for MO distribution arresters are quite low (in the range of 0.1 to 1.0 per cent per annum with some geographic variation) and virtually zero for most HV units.
Commercially-speaking, MO arresters have offered manufacturers a relatively simple design compared with the complex internal structures of the previous generation of gapped arresters. Indeed, continued progress in this technology and the enormous increase in the volumes of units produced every year have together led to significant decreases in production costs and market prices. Equally important, the very simple structure of an MO arrester – the active part consists basically of a stack of cylindrical MO resistor elements – contributed to a related development only few years later, namely the increasing application of polymeric housings in place of porcelain.
Polymeric-housed arresters were first introduced during the mid 1980s at distribution voltages. A major driving force in this regard was the poor performance of many porcelain-housed distribution arrester designs which often suffered from sealing deficiencies, extreme sensitivity to pollution and dangerous overload performance. While sealing problems were not automatically solved when substituting a polymeric housing for porcelain (it took a few more years to achieve this) the new technology still offered immediate advantages with respect to production automation, reduced lead times, lower shipping costs, easier handling and so on. With price being the most important (if not the only) factor when users select a distribution arrester today, these commercial benefits are the main reason why the worldwide market share of polymer-housed distribution arresters is virtually 100 per cent. The market situation, however, is entirely different when it comes to HV arresters. Here, much more than at distribution levels, technical performance plays an important role and user’s attitude toward newer technologies is typically far more conservative.
The first polymer-housed HV arresters appeared on the market around 1990 as niche products. Unlike the case for distribution applications, price was not the key element in initial market interest. Rather, it was the outstanding technical features of these new designs which included superior breaking resistance under overload conditions, excellent mechanical strength, resistance to vandalism and good pollution performance due to the hydrophobicity of the silicone housing. Today, with growing production volumes, price has grown in importance and here too new polymeric designs have shown that they can be very competitive. Indeed, depending on technical requirements and the specific design, polymeric HV arresters are often less costly than a comparable porcelain-housed unit. Given this, the market share of polymer housed HV arresters is likely to increase steadily.
Major Design Principles for Polymer-Housed HV Arresters
The term HV actually covers quite a broad range, from Um = 72.5 kV up to Um = 800 kV (higher levels exist but do not yet play an important role). This range could be sub-divided into two principal segments where distinctly different philosophies govern the decision-making process for purchasing an arrester. At voltage levels up to Um = 300 kV, i.e. the lower transmission and the sub-transmission levels, in most cases, just normal technical requirements apply. There is typically little need for special features such as extra-high mechanical strength or maximum safety considerations. These are the voltage levels of most standard applications where one increasingly sees the same selection criteria applied as in distribution systems. Therefore, comparatively little time or money is spent to optimize arrester layout for any particular location. This voltage range is therefore the domain of the low cost (in the positive sense) arrester.
For EHV levels (Um = 360 kV and higher), special requirements in regard mechanical characteristics play an increasingly important role and one which cannot easily be fulfilled by the low cost designs. Furthermore, users are less willing to take any risk whatsoever of possible arrester failure. The electrical and mechanical requirements for arresters in this category are therefore evaluated by careful system studies. Moreover, in many cases, the user has detailed knowledge and information about the system configuration as well as clear ideas about the optimal arrester for the particular application. As a result, this segment is the major domain of the special feature arrester.
Both low cost and special feature type HV arresters are today available in polymer-housed designs.
• Mechanical Supporting Structure
Fig. 1 provides a schematic classification of the various mechanical design principles of HV arresters, whether polymer or porcelain-housed. An important differentiation is made between designs employing a hollow core insulator with an intentionally enclosed gas volume and alternative designs where the housing is applied directly onto the MO column. In the case of polymer-housed arresters, the basic mechanical designs can be classified as follows:
Type A Design
This design concept – the tube design – is the more conventional approach, looking quite similar to a porcelain-housed arrester. The stack of MO resistor elements is mechanically supported by an internal cage structure made for example using FRP rods. This insert is then clamped between the end flanges with the help of compression springs. Additional supporting elements may be necessary to fix the insert in the radial direction. What is important and sometimes criticized in this concept is the fact that, because of the enclosed gas volume, this type of arrester needs a sealing and pressure relief system. In fact, for polymeric HV arresters, this system must be designed and manufactured with the same care as for porcelain-housed arresters. However, this does not represent a real problem for an experienced manufacturer.
There are numerous models of polymeric HV arresters which have an excellent service record with respect to their sealing systems. Therefore, whether or not this type of arrester design has problems with moisture ingress is mostly a matter of manufacturer know-how and production quality control (much as with porcelain-housed arresters).
Another issue with the Type A design relates to whether vapour can permeate directly through the polymeric sheds and walls of the housing or through the bonding area between the flanges and the FRP tube. Research studies as well as service experience over the past 15 years have demonstrated that this is not the case. Rather, the amount of moisture ingress due to these mechanisms is even below that which can pass through a good sealing system. Therefore, this is not really an issue. This small amount of moisture can easily be controlled by internal desiccants, much as with nearly every HV device in electric power systems. Apparently, more research is now being done in order to better understand these mechanisms and to derive minimum design requirements for composite hollow core insulators used as arrester housings.
The Type A arrester is more expensive to produce than a comparable porcelain-housed arrester since composite hollow core insulators are generally more costly than porcelain, at least for voltages up to about 400 kV. This arrester is therefore the typical special feature unit which, in most cases, offers several technical advantages over both the low cost polymer-housed and the conventional porcelain-housed HV arrester.
Some of these potential benefits of the Type A design are extremely high mechanical strength, the safest possible short-circuit performance and the possibility of having a single-unit arrester even up to 300 kV system voltage. Future developments with respect to the cost of composite hollow core insulators will determine whether or not this type of arrester design will be limited mostly to EHV applications.
Type B1a Design
This type, often referred to as the wrapped design, was actually the very first design principle for polymer-housed arresters when they where first introduced at distribution voltages in the mid 1980s. The concept has been extended to HV arresters since one possible solution for building a HV arrester is to connect a large number of distribution arresters in parallel and in series.
Common to all Type B1a arresters is the fact that the mechanical structure is wrapped directly onto the MO resistor elements (in some cases applying a thin intermediate foil between the MO stack and the wrap). This could be done in a wide variety of ways and indeed many sub-solutions for this basic concept have been realized for technical, commercial and patent reasons. Possibly the most economic variant is shown in Fig. 2. Fiberglass rovings soaked in uncured epoxy resin or pre-impregnated ribbons, are wound crosswise around the MO stack. The module is then cured in an oven and the resulting rigid ribbons provide the required mechanical strength. Since they do not fully overlap, these ribbons form rhombic ‘windows’ which are important technically for short-circuit performance and commercially in regard to minimizing the amount of material used (which is also a concern in order to limit the amount of flammable material present). If these windows are too large, however, the mechanical strength of the module could become insufficient.￼
At the center in Fig. 2 is a design variant where no windows remain open due to full overlap of the ribbons or by using pre-impregnated FRP mats with appropriate orientation of the glass fibers. This provides high mechanical strength but forms a closed tube in which internal pressure can build up in the event of overload, possibly leading to violent breaking of the housing. In order to improve short-circuit performance in this case, slots can be provided on the surface which function as pre-determined breaking points. The variant on the right in Fig. 2 shows a design which is also completely closed, realized by a pre-impregnated mat wound around the MO stack. However, in this case the glass fibers are almost exclusively arranged in an axial direction. This is also a possible means to improve the short-circuit performance since, if carefully designed, the tube will easily tear open in case of internal pressure build-up.
Arrester Type B1b is quite similar to the closed tube variants of Fig. 2, however there is a basic difference in the manufacturing process. Instead of wrapping FRP material onto the MO column, a pre-fabricated FRP tube is used which has a diameter larger than that of the MO column in order to be pushed over it. The resulting gap between MO and FRP material is then filled by solid or semi-solid material. Again, slots can be provided on the surface in order to improve short-circuit performance.
Type B2 Design
This is a completely different design concept usually referred to as the cage design. While the mechanical strength for Type A and Type B1 arresters is provided solely by the FRP structure (whether a closed or partly open tube) this is accomplished by the MO resistors themselves with the Type B2 concept. For this purpose, they are clamped between the metal end fittings by FRP loops or rods which apply enormous axial pre-stress in the range of 100 kN. The basic mechanical design is shown in Fig. 3. The variant at the left uses loops which are fixed to notches in the end flanges. This design was first introduced for distribution arresters and later extended to HV units. A sub-variant (middle) uses an additional bondage from polymeric material in order to achieve the mechanical and short-circuit characteristics required for HV and EHV applications. A third variant for Type B2 arresters (right) was also first realized for distribution and then developed further for HV. Here, FRP rods are applied which are mechanically fixed to holes in the end flanges employing a proprietary clamping system.
Fig. 4 shows actual arrester modules produced by two manufacturers. The main technical advantages of these cage designs are that they offer comparatively high mechanical strength combined with an inherently good short-circuit performance. It is impossible to make any generalization on comparative manufacturing costs of these designs since much will depend on factors such as: production volumes, degree of automation, extent of process optimization, quality of raw materials and so on. However, it can be stated that, in general, the Type B approach constitutes the most economic way to produce an arrester. At the same time it offers a level of technical performance which ranks somewhere between comparable porcelain or polymer-housed Type A arresters. The Type B arrester is therefore the typical low cost arrester referred to earlier and this is one of the reasons for the design’s success on the market.
• Outer Housing & Sheds
In regard to mechanical characteristics, there are a number of different design possibilities for the outer housing and sheds of polymeric HV arresters. However, as far as material is concerned, there has been a clear tendency worldwide towards silicone rubber (SR). Other polymeric materials – such as ethylene propylene diene copolymer (EPDM), EPDM/SR blends and ethylene vinyl acetate (EVA) – all of which are widely used in distribution and which may perform quite well in these applications, are generally not accepted for HV or EHV use. The reason for this preference for SR relates to its long-term hydrophobicity as well as its chemical structure which is inherently less sensitive to solar radiation due to the high bonding energy of its basic component. This is, however, a benefit which must be paid for since market prices for SR are typically in the range of two times that of EPDM.
For Type A arresters there has never really been an alternative to SR for both production and performance reasons. This type of design belongs mainly to the family of special feature arresters, where traditionally high-end raw materials have been used. From a production point of view, the only way to cover a hollow core tube is by using SR.
Although the very first such designs used an insulator with sheds made from high temperature vulcanizing (HTV) SR individually slipped over the FRP tube, contemporary composite hollow core insulators are mainly covered by a direct molding process, using either room temperature (RTV) SR or liquid silicone rubber (LSR). The latter may well become the material of choice for the future offering important production benefits, e.g. a reasonable compromise between ease of handling, process temperatures and pressures as well as vulcanizing time. The same arguments favouring SR apply for the Type B2 arresters, which can only be covered by a direct molding process. All common types of SR can be found with arresters of this specific design.
Most alternatives in respect to type of polymeric material exist for the Type B1 design, where the external sheds can be produced either by direct molding or by slipping pre-fabricated housings over the internal modules. The latter concept therefore offers the highest flexibility in production, but special care is required for its implementation. Since Type B1 arrester designs usually do not have a smooth internal surface, a sealing material (e.g. a silicone compound) must be inserted between the internal parts and the outer housing. This has to be done in such a way as to prevent formation of any internal voids which could cause internal partial discharges and affect long-term performance. Furthermore, an appropriate sealing system at the end fittings must be provided. Fig. 5 shows two possible ways of sealing. The alternative shown on the right inherently offers better reliability due to the chemical bonding of the housing to the end fittings. Since this solution can only be achieved by direct molding, it is restricted to those designs employing SR.
It should be emphasized that while it is virtually impossible to create a good housing using poor materials, it is certainly possible to make a poor housing even when employing very high quality materials. In other words, apart from a discussion of materials, it should never be forgotten that the final performance of an arrester is greatly influenced by its design as well as by quality control during production. While only the first aspect is generally evaluated during accelerated ageing tests, it is the latter factor which contributes most to long-term risk and therefore should also be carefully considered.
Performance Aspects of Polymeric HV Arresters
• Short-Circuit Performance
The failure rate of MO HV arresters is typically very close to zero. The main reasons for failure could be unforeseen temporary overvoltages or nearby lightning strikes due to shielding failures of the overhead lines close to a substation. However, in this extremely rare case, an arrester is expected to fail in a manner which does not endanger persons or nearby equipment.
Traditionally, users have accepted that, even in the event of a well-performing pressure relief system, there would always remain some risk of the arrester’s porcelain housing shattering due to the thermal stress imposed by the burning arc. Indeed, this fact has been taken into account by the applicable arrester standard (and also by its predecessor) by giving guidance on the definition of ‘non-violent shattering’: namely that all potentially ejected parts shall reach the ground within a circle around the arrester having the same radius as the arrester’s height.
Both of the arresters shown in Fig. 6 have successfully passed the short-circuit test (referred to in the old standards as the pressure relief test). Above a certain unit length, the pressure relief performance of porcelain housings cannot be controlled any more than this. They are therefore limited to a length of about 2 meters and a single-unit porcelain-housed arrester can thus be realized only up to 245 kV system voltage. Polymer-housed arresters (if not made from cast resin) have an inherently good short-circuit performance since they do not rely on brittle materials such as porcelain. In addition, the results of short-circuit tests involving these types of arresters are generally more reproducible. There are, however, important differences between the alternative design concepts.
Type B arresters do not have an enclosed gas volume and therefore do not need any specific pressure relief devices or vents. The intention is that their housing tears open at any location (wherever the puncture or flashover of the MO column happens to occur) and the full arc develops outside the housing. This is in fact the only possible failure mode for the Type B1a design of Fig. 2 (left) or for the Type B2 arrester. For other Type B arrester variants, however, mode of failure will depend on the particular design of the tube formed by the FRP wrap and how reliably this concept works. If the tube is too strong, it can tear open quite violently. Theoretically, it is possible that in the case of all Type B arrester designs the mechanical structure will break down completely under short circuit and that various internal parts, e.g. MO resistor elements, will be ejected. The extent to which this occurs or does not occur will be determined by the specific design.
In general, inflammability is another area of concern especially in the case of polymer-housed arresters. Current standards require that any open flame following a failure will self-extinguish within a maximum of 2 minutes (this value is apparently still under consideration). Arrester designs such as the Type B2 having a minimum content of epoxy resin are advantageous in this respect.
In summary, the main advantage of the Type B design in relation to porcelain-housed arresters is that the risk of violent shattering is extremely low or even zero (with differences between the sub-variants of this concept as explained above). The possibility of ejected parts or a complete loss of mechanical integrity must also be considered in the layout of a substation. Fig. 7 illustrates a Type B2 arrester after a short-circuit test at its rated short-circuit current. By contrast, Type A arresters, being based on a totally different design concept, can virtually guarantee that no internal parts are ejected and that they will remain mechanically intact after failure. For example, Fig. 8 shows the typical result of a short-circuit test at Is = 63 kA on a Type A arrester unit of 2.5 m height. Its mechanical integrity remained unaffected and, according to the literature, its mechanical strength after failure was 75 per cent of the original value.
This performance actually allows this type of arrester to serve additionally as a post insulator. In such an application, the higher price of the arrester (compared with a porcelain or polymer-housed Type B arrester) can still be justified since the overall expense of a substation can be reduced if this benefit is taken into consideration during the planning phase (the need for separate post insulators including their pedestals and foundations can be eliminated). Of course, a failed arrester will constitute a short-circuit and will have to be replaced. However, this is an extremely rare event (failure rate close to zero) and therefore this can be considered as an acceptable level of risk. Examples of such applications are provided in Fig. 9.
As an additional advantage in the event of failure, by appropriate orientation of the venting outlets, the arc can be controlled in any favoured direction. This is particularly important when arresters are installed close to transformer bushings. Risk of fire is usually not an issue here since the heated MO resistors do not come into direct contact with the housing. Rather, the arc commutates to the outside within only a few milliseconds, burning there at some distance from any flammable material. With all these benefits, the Type A arrester really represents the high performance type, perhaps not always needed yet preferred for certain special applications.
• Pollution Performance and Radial Electric Field Strength
Problems due to pollution were one of the original reasons why polymer-housed arresters were considered at their very early beginnings. Indeed, many users had become quite concerned about the sometimes poor pollution performance of the gapped arrester design. As it has turned out, MO type arresters are in general not very sensitive to pollution. Nevertheless, the hydrophobic surface of a polymeric housing offers an additional safety margin in this respect.
Fig. 10 illustrates three possible mechanisms which can affect a multi-unit MO HV arrester operating in a polluted environment. Theoretically, there is a risk of external flashover of the housing (1), but this is not a real concern if the environmental conditions are not too extreme. The power frequency voltage is nearly perfectly graded axially by the internal MO column and transient overvoltages are therefore limited more effectively than for any other equipment in the system.
Similarly, the possibility of excessive heating of the MO columns in a multi-unit arrester (2) is mostly theoretical. This could occur when the conductivities of the surface layers on the individual housings are extremely different. In such a case, the surface current along one unit of high external conductivity may commutate to the MO column of the next unit if its housing has a lower surface conductivity. These currents may then heat the MO resistors of this unit.
This phenomenon was first observed during field tests supervised by Cigré in the late 1980s but there is really no indication that this will constitute a severe problem in service. When there is reason for concern at locations of extreme pollution severity, the best way to avoid partial heating of individual units is to apply single-unit arresters whenever possible. Depending on the arrester design principle, this solution is limited to system voltages in the range of 170 kV to 300 kV.
Finally, the possibility of internal partial discharges also exists in the case of both porcelain and polymer-housed arresters of the Type A design. Internal partial discharges can be initiated by radial electric field stress due to different voltage distributions along the internal MO column and the outer surface of the housing.
The internal axial voltage distribution of an MO arrester is nearly even due to its high self capacitance as well as the presence of the grading ring(s). The external distribution, however, shows statistical behaviour in case of surface conductivity due to pollution. The electric potential of an intermediate fitting or flange can be transferred along the insulator and thus a high electric field stress could occur between the outer surface of the housing and the MO column (see Fig. 11).
In the case of a Type A arrester with a gas-filled gap between the housing and the MO column, the highest fraction of the radial voltage drops across this gap due to the geometry and the ratios of dielectric constants of the different materials. The electric field stress at the MO column can therefore reach values above the initiation level of partial discharges.
There are two methods to make Type A arresters resistant to this effect. One is to apply a sufficiently large gap. Investigations under salt fog stress have shown that a gap of 25 to 30 mm for porcelain and of 10 to 20 mm for polymer housings with SR sheds is sufficient to effectively avoid any internal partial discharges. As a gap distance of less than 10 mm cannot be applied for other reasons (e.g. short-circuit performance), the risk of internal discharges in Type A arresters is quite low. Secondly, only materials and MO resistors which are not sensitive to possible partial discharges should be employed. In the case of FRP elements, for example, they should have high tracking resistance while MO resistor elements should be properly coated and resistant to ageing even in oxygen-free atmospheres.
Radial field stress is also a concern for Type B arresters. Here the radial voltage drops across a distance of only few millimeters between the outer surface and the MO column. Therefore, a weak design could risk being punctured by this stress. As the possible radial field stress increases with the distance between two metal fittings (where the potential difference is equalized), the maximum unit length is limited. For this reason, the present arrester standard requires a weather ageing test (including salt fog stress under applied voltage) for the longest unit in the case of all polymer-housed arrester types. Poorly-designed shed profiles will fail this test by puncture.
For solutions optimized with respect to material consumption (a concern with respect to housing wall thickness and shed profile) maximum lengths of about 1.2 m per inidividual arrester unit have emerged. Single unit arresters of Type B designs are therefore limited to system voltages up to 170 kV. Above that, they must be constructed as multi-unit arresters. Any posible risk of partial heating would in this case be reduced by the better heat dissipation capability of a Type B arrester compared with the Type A design.
Type A arresters – if the design rules addressed above are followed – can be manufactured in lengths of more than 2.5 meters. Above this value, other restrictions apply, e.g. from short-circuit performance or from handling aspects during production and transport. Nevertheless, with this length single-unit arresters are possible up to 300 kV system voltage.
• Mechanical Aspects
One of the major differences between alternative design concepts for polymer-housed HV arresters relates to their mechanical strength, defined in terms of applicable head loads, particularly force or bending moment (cantilever strength). Normally, one should distinguish between static and dynamic loads, but unlike porcelain-housed arresters where a lot of experience is available, general rules for specification of these two values do not yet exist and are still under consideration.
According to arrester standards, the manufacturer has to specify an MPSL (maximum possible service load) and laboratory bending tests must demonstrate that there is no damage to the arrester at this value. Definition of a damage limit is, however, more difficult because a polymer-housed arrester does not break in the manner of a porcelain-housed type. When a certain load value is exceeded, the degree of damage increases permanently starting from zero.
Actually, the standard refers to visual inspection and to a remaining permanent deflection after the bending test of less than 5 per cent of the arrester’s height. There is agreement, however, that this definition has to be reconsidered. New proposals which are now being discussed within IEC suggest 1 per cent of the arrester height or 5 per cent of the maximum deflection. Whatever the future definition of MPSL, this value must be applicable in service without any damage to the arrester. Reasonably, it must be interpreted as a dynamic value but, due to lack of general experience, it is really up to each manufacturer how to specify the permissible static loads for their particular arrester design. A reasonable approach is to specify 50 per cent to 70 per cent of MPSL as a static load value depending on specific design variant.
The highest mechanical strength can be achieved by Type A arresters. Since the manufacturer can apply any housing diameter and wall thickness of the FRP tube, independent of the MO column diameter, virtually any mechanical strength can be reached. Of course, there are economic restrictions which apply.
The highest static cantilever strength value offered on the market would be necessary, for example, to fulfill seismic requirements for a 550 kV arrester according to the US standard IEEE 693 (acceleration of 1 g at the arrester base; 50 per cent mechanical utilization). These requirements could also be met by Type B arresters, however only if some additional reinforcement is introduced such as bracing elements attached to an intermediate flange of the arrester. The principal arrangements for both these solutions are shown in Fig. 12. Seismic requirements impose the highest possible requirements on the mechanical performance of a HV arrester. Standard applications for systems up to Um = 300 kV can usually be served by arresters of considerably lower mechanical strength values.
The minimum required mechanical strength for standard applications, if no additional mechanical stress has to be considered, is 400 N for Um £ 420 kV, 600 N for Um = 550 kV and 800 N for Um = 800 kV. These values take into account normal service conditions (e.g. wind speed of 34 m/s) and connection to the overhead line and bus bar respectively by flexible wires. In a 420 kV system, for example, where the arrester height a value which is offered by most (but not all) Type B arresters on the market.
Table 1 provides an overview of static cantilever strengths based on data made available by various suppliers. From this data it could be concluded that Type B1 arresters can be applied mainly for system voltages up to 170 kV. Type B2 arresters are mechanically stronger and can also be used in EHV systems if there are no additional special requirements regarding mechanical strength. In many cases, however, the user specifies mechanical load values which by far exceed the minimum values mentioned above.
Polymer-housed arresters differ from porcelain-housed types in that they are deflected under continuous static and far more under dynamic mechanical loads. This is not a real problem but must be taken into consideration in the process of station layout, particularly if the distances between the phases and between phases and ground are chosen at the lower limits. Bending plays less a role for Type A than for Type B arresters.
Manufacturers of Type B arresters like to point out that there are alternatives to install an arrester other than the classical standing upright, base-mounted position. Indeed, suspended mounting of the arrester, for example, requires only a minimum of mechanical strength for the unit. However, in this case the whole substation construction must be designed to carry the weight of the arrester and this is often not the case for existing installations. Such mounting alternatives should therefore ideally be considered early on in the planning phase.
Transmission line surge arresters (TLSA), which are now increasingly installed to protect overhead line insulators from back-flashover, are a good example of an application where all the major benefits of Type B arresters can be utilized without any mechanical restrictions. Their low weight, ease of handling, short-circuit performance and comparatively low price all play an important role in this regard. It should be pointed out, however, that for those locations which have traffic by the public, the totally safe overload performance of Type A arresters might make them a better choice. Fig. 13 illustrates two innovative applications where Type B2 arresters have been installed as TLSAs on EHV lines.
The continued growth of the polymer-housed segment of the HV arrester market will of course ultimately depend on key factors such as technical performance and market prices (both for final product and individual components). However, an equally important consideration will be the ability and flexibility among users to recognize and utilize the potential for optimization offered by this comparatively new technology. For example, as has already been pointed out, station layout with polymer-housed HV arresters offers an excellent possibility to optimize costs – not widely utilized as of yet – if utility engineers look beyond the traditional ways of arrester installation. Suspended mounting of arresters, for example, allows installation of comparatively weak Type B designs even for those applications where high mechanical requirements would normally exist for base mounting. In much the same way, the higher costs of the Type A arrester design can easily be justified if its inherent ability to also serve as a support device is utilized. But to take full advantage of such benefits, the new design possibilities must be recognized and considered almost from the start when planning new installations.