The use of fiberglass standoffs has grown in popularity as an alternative to standard wood crossarm construction in distribution systems. As the number of units and the amount of time in the field has increased, several utilities have documented cases in which a fiberglass standoff has unexpectedly flashed over. Line crews at National Grid USA initially witnessed the phenomenon after making repairs following a storm. When the distribution circuit was re-energized, a standoff assembly flashover was observed. The assembly was replaced and the failed standoff was sent for evaluation. The standoff was intact other than evidence of arc damage and weathering. Since reported flashovers of this type have been rare, there has been little research of these occurrences to date. The following discussion presents the most plausible theory to explain the standoff failures based on the preliminary analysis of several failed standoff units.

Historically, distribution line construction utilized ceramic insulators mounted on metal pins and wooden crossarms to support the phase conductors as shown in Fig. 1. Although wooden crossarms have performed well, being both strong and durable, they also are heavy and labor intensive to assemble. As an alternative, the insulator industry began producing polymer-based replacements for the wooden crossarms in the form of fiberglass reinforced polymer (FRP) standoff brackets. A typical polymer crossarm installation is shown in Fig. 2. The main advantage of the FRP standoff over traditional wooden crossarms lies in the speed and ease of installation, resulting in lower labor costs. Therefore, even with a slightly higher materials cost, the FRP standoff still has an economic advantage.

A polymer standoff consists of an FRP core with metal end fittings bonded or clamped to each end. One end fitting provides a flat base where the standoff can be mounted to the utility pole, while the other end fitting provides a threaded connection point for a pin-type insulator (Fig. 3). FRP is used for the core because it is relatively inexpensive, easy to fabricate and mechanically strong. Under dry conditions FRP is nonconductive. The FRP core is coated with a UV-resistant paint to give the standoff a weather-resistant finish.

Mechanics of the Flashover

The failure of the fiberglass standoff is the result of a combination of factors, each of which by themselves would probably not lead to a flashover. The first cause to be considered is the poor weathering characteristics of the epoxy coating on the standoff. The epoxy coating is seen to degrade rather rapidly when exposed to the UV present in natural sunlight. As the coating degrades, it exposes the underlying FRP core as shown in Figs. 3 and 4. In fact, many field installations have been noted where half the length of the core was exposed after just a few years in the field. When the core is exposed, precipitation falling on the standoff is wicked into the core by the glass fibers, thus making sections of the core conductive and reducing the overall creepage distance of the standoff/insulator assembly. This alone does not pose a great threat because the pin-type insulator, which separates the standoff from the conductor, should provide an adequate level of insulation. However, the dielectric relationship between the ceramic pin-type insulator and the fiberglass standoff is such that they form a capacitive voltage divider as shown in Fig. 5. The divider is characterized by the following equation:

The capacitance of the ceramic insulator is much larger than the capacitance of the fiberglass standoff, thus V_STANDOFF is nearly equal to V_{LINE-TO-GROUND}. This means that most of the line voltage is held off by the fiberglass standoff and not the ceramic insulator. This does not normally pose a threat under dry conditions. However, as mentioned previously, when precipitation is wicked into the fiberglass core, it becomes conductive thereby reducing the creepage distance. If the reduced creepage distance is not sufficient to hold off the voltage, and the line potential does not shift back to the insulator when the standoff becomes conductive, then the standoff flashes over.

Are Flashovers More Likely After Thunderstorms?

The common impression within the utility industry is that the fiberglass standoffs are more likely to flash over shortly after thunderstorms. However, there is no specific evidence to either prove or disprove this. It is possible that the standoff failures appear to be more frequent following thunderstorms because line crews are out repairing storm damage, and thus are more frequently in a position to observe the failures during this time period. Nevertheless, there are some interesting characteristics of the post-storm time period that lend credibility to the idea that the standoff flashovers are more common in the post-storm time frame.

Two factors may cause increased flashover activity after thunderstorms. The first deals with the wetting and drying characteristics of the standoff and the impact on its conduction properties. The second deals with high-frequency transients, which may occur when re-energizing a line after a storm-related outage or line repair. During the storm, precipitation is wicked into the core through areas of exposed fiberglass. The wetted core provides a conductive path, causing the line potential to shift from the standoff to the insulator. Once the storm passes and the standoff begins to dry, dry bands begin to form, which disrupt the conductive path and cause the potential to shift back to the standoff. If the potential across the standoff is great enough, an arc will form over the surface to bridge these high-resistance dry bands. This dry-band arcing process leads to carbonization of the polymer, creating permanently damaged conducting sites that can grow over the surface of the rod. This process can then lengthen to eventually short the entire standoff, resulting in a flashover. Coincidentally, the time frame in which the candlestick becomes partially dry is also when line crews are likely to be re-energizing circuits that had tripped due to storm activity. When a circuit is re-energized, a transient is created and the line can experience an increased voltage and a high-frequency ringing on top of the 60-Hz power frequency. As if the overvoltage wasn't bad enough, the voltage divider between the ceramic insulator and the fiberglass standoff is also more likely to be dominated by their relative capacitances, as opposed to resistances, under the higher frequency conditions, thus furthering the extent to which the line voltage must be held off by the fiberglass standoff.

Laboratory Demonstration

A laboratory experiment was set up to investigate the discharge behavior of a fiberglass standoff upon energization after the passage of a storm. In this test, an ANSI Standard 29.5 pin-type porcelain insulator, complete with a tied conductor, was placed on top of a standoff. The standoff had previously been in service for a sufficient time for most of the painted surface to have been lost due to weathering, revealing much of the underlying resin and reinforcing glass fibers. A small amount of distilled water was sprayed over the entire assembly to simulate wetting from the passage of a storm. The output of a voltage source was connected to the conductor on top of the insulator, while the bottom of the standoff was grounded. The applied voltage was chosen to be at a high frequency to correspond to the oscillations present in a typical energizing switching operation. Numerous discharges were found to appear along the length of the standoff, occasionally accompanied by discharges on the insulator and conductor assembly. Figure 6 shows an example of the discharge behavior.

The voltage applied in this test was 35 kV at a frequency of 1 kHz. Figure 6 shows that there are several coincident arcs along the length of the fiberglass rod, indicating that most of the applied voltage lies across its surface rather than on the insulator it is supporting.

Degradation of Other FRP Apparatus

Other FRP products often experience the same type of field degradation regardless of whether or not they are under electrical stress. For example, FRP guy strain insulators that are used in place of ceramic guy insulators (commonly referred to as “johnny balls”) often exhibit similar surface degradation from UV exposure and environmental contaminants. They are afforded somewhat better UV resistance compared to the insulator standoffs because they are veiled rather than painted, although they still show similar degradation over time. The veiling technique is used to provide the guy strain insulators with better abrasion resistance. Although they are not normally under electrical stress, it is unclear how the surface degradation affects the mechanical strength of the insulators.

Possible Solutions

There are several possible solutions that can be applied to avoid the standoff flashovers:

• Different insulators

  It may be possible to use insulators with a lower capacitance, thus altering the capacitive divide ratio and placing more of the working voltage across the insulator instead of the standoff. Some polymer insulators may offer a lower capacitance than ceramic insulators. However, capacitance is highly dependent on insulator shape, so long, skinny insulators, much like the candlestick, would most likely offer the best chance for a more favorable divide ratio.

• Different or additional sheath material

  The problem may be remedied by keeping the fiberglass core from being exposed. One possibility would be to jacket the core in several millimeters of silicone rubber. While it would add to the cost of the unit, it would also provide far superior weatherability and keep the inner core from being exposed. Other materials such as EPDM or an improved paint formulation could also be employed.

• Conductive standoffs

  By incorporating conductive materials in the core of the standoff, or even by using a metallic standoff, the voltage distribution would favor a condition where the line voltage would be across the insulator only, at all times. This approach achieves the goals of fast and inexpensive construction while reducing the flashover potential of the standoff/bracket.

• Wood construction

  While the FRP standoffs are lighter and less costly to install than the traditional wood crossarms, which is why they are used, the wood crossarm does not pose the same flashover threat as the FRP standoff.

Further Study Needed

Little information is known within the utility industry regarding standoff failures. One reason for this is that the failures are seldom observed and hard to diagnose. If such a failure causes a protective device to operate, the standoff may function normally once the line protection clears the fault and the circuit is re-energized. One consideration that remains relatively obscure is what exactly the electrical path is when the standoff flashes over. It is not known whether it presents a line-to-ground fault, line-to-line fault or both. The main source of information on the occurrence of these failures has come from line crews who witnessed these failures in the field. There has been little laboratory research done to investigate the cause of these failures and possible remedies. It would be beneficial to the industry for more formal research to be undertaken to study the processes that lead to such failures.

Dave Crudele is a consulting engineer with EPRI Solutions Inc. in Schenectady, New York, U.S. He holds BSEE and MSEE degrees from Clarkson University and is working toward professional licensing in New York state. Crudele currently works in the areas of distribution engineering, power quality and distributed energy resources. dcrudele@eprisolutions.com or dcrudele@epri-peac.com

Paul McGrath holds the position of professor of electrical and computer engineering at Clarkson University, where he performs teaching and research in the areas of power, high voltage, electrical insulation and electromagnetics. mcgrath@clarkson.edu

Clayton Burns is a principal engineer in the meter engineering department at National Grid USA Service Co. in Syracuse, New York, U.S. His professional experience includes industrial electrical systems, power generation and electric utility research. Burns earned BSEE and MSEE degrees in 1977 and 1982, respectively, from Lehigh University. He is a registered professional engineer in the state of New York. clayton.burns@us.ngrid.com

Girard Purdy is manager of OH engineering and operations at National Grid USA in Syracuse New York, U.S. His professional experience includes construction management, operations and work methods. Purdy earned his BSEE degree from Clarkson University in 1980. girard.purdy@us.ngrid.com