The porcelain cap & pin suspension insulator with cemented hardware was introduced a century ago. The external details of modern designs of this type of insulator (also commonly referred to as bells, discs or units) look almost identical to the original design – a fact that is no small tribute to the engineers who first proposed it. The bells are mechanically linked to one another to form a string whose length increases with operating voltage. They are offered in a range of mechanical strength ratings, commonly expressed as kN (kilo Newton) in most of the world and kips (kilo pounds) in North America. The most noticeable change in the insulator units designed for higher mechanical strengths is in the dimensions of the hardware. The porcelain composition can also be different, however this is not easily apparent visually.
The focus of this past INMR article by Professor Ravi S. Gorur was on suspension insulators where the electrodes (cap & pin) are separated by about one inch (circa 2.5 cm) of porcelain, thereby subjecting the dielectric to considerable electrical stress. The failure mode of particular interest in this case is puncture and not surface flashover. Punctured bells can lead to dropped conductors as well as decreased life for the rest of the string since the remaining healthy units are then subjected to greater electrical stress than normal. In other insulator configurations, e.g. line posts or long-rods, the electrical stress is fairly low because the voltage is applied across a much bigger body of porcelain dielectric. Puncture is therefore unlikely to occur.
The formulation details of most insulator materials and their processing methods have evolved over time. Typically, these are closely-held secrets by the manufacturers. In the case of composite insulators, users are already aware that variations in these factors can lead to a major difference in performance and useful service life. By contrast, users today tend to think of porcelain insulators as comprising a single homogeneous class. But this is not true. The purpose of this article was to illustrate that, in fact, large variations are possible in porcelain insulators made by different suppliers.
Porcelain insulators today are generally required to pass ANSI C 29.1 or IEC 60383 standards before they can be considered by utilities for installation on their overhead networks. However, the various tests described in these standards are not intended to compare products from different suppliers. Nor do they provide information on how long the insulators can be expected to last in service. This type of knowledge is obtained only through experience. There are many documented instances where porcelain insulators have performed well for over 50 years and other situations where they failed prematurely by puncture in a relatively short space of time (i.e. from 5 to 20 years). As part of this research, new porcelain suspension insulators from different manufacturers were assembled from utility warehouse inventories. In addition, a number of insulators were obtained after removal from lines following 10 to 30 years in service in the United States. Some of these removed insulators had already been punctured. These insulators are listed in Table 1 and those with common letter identification come from the same manufacturer.
Materials & Manufacturing Details of Wet Process Porcelain
To better appreciate the role played by porcelain microstructure on both electrical and mechanical performance, below is a brief synopsis of the materials used and the manufacturing details for such insulators. The basic steps in the manufacture of porcelain insulators are illustrated in Fig. 1. Porcelain is a mixture of clay, quartz and feldspar. In the case of distribution voltage insulators (typically 15,000 lb or lower), it is common to use siliceous (quartz) porcelain, whereas aluminous (corundum) and cristobalite porcelain are used for higher strength (25,000 lb or more) as usually required for most transmission voltages. These minerals are mined from many places all over the world and, since no two locations are alike in all respects, some variations can be expected in raw material quality. These variations must then be considered during subsequent manufacturing processes (e.g. mixing, firing, etc) to ensure consistent quality of the final product. The raw materials are mixed in water to form a slurry. Water facilitates intimate blending of the various constituents and helps shaping the product into its the final form. The moisture is then removed in various stages of forced drying. Particle size of ingredients is one of the parameters that can greatly influence insulator performance. Particle size is commonly expressed as a mesh number such that the higher the mesh number, the smaller is the particle size. High voltage porcelain insulators use much greater mesh numbers (i.e. 325 or higher) than porcelain for non-electrical applications. For example, it is quite common to use mesh numbers of only 150 (106 µm) for applications such as dinnerware, sanitary ware or floor tiles. Firing takes place in shuttle kilns or in long tunnel kilns where temperature cycle is controlled through the different zones with the maximum being in the range of 1150-1250°C. There are variations in the energy source used for heating the kilns, (oil, gas, and electric) with natural gas being perhaps the most common today. The metal hardware (i.e. the cap and the pin) are cemented onto the porcelain shell. There are differences in cement composition used by the various manufacturers with some using Portland cement while others use a mortar mix (Portland cement plus silica filler).
Evaluation methods for test insulators
The insulators being tested were first cut axially using a high pressure water jet in order to expose their cross-section. Details such as the color of the porcelain body, uniformity of cement-porcelain interface and presence of large pores, if any, were noted. It was striking to see the large variations in the porcelain body color as shown in Fig. 2 and listed in Table 2. Deviations from bright white suggest presence of impurities and/or inadequate oxygen during the firing process.
The important material characteristics of the porcelain body are porosity, microstructural defects, crystalline phases and their size. These properties were evaluated in a scanning electron microscope (SEM). Their impact on insulator performance is summarized in Table 3.
Scanning Electron Microscopy
The following steps describe the preparation of samples:
1. Small pieces were obtained from the insulator using a diamond saw.
2. Samples were polished using 800 grid sand paper, 6 µm and 1 µm diamond films.
3. Samples were then cleaned in an ultrasonic bath with acetone.
4. Samples were etched with 20% HF acid for 10 seconds to expose grain boundaries.
5. Gold coating was done on polished and etched samples using gold sputter coater for 3 minutes.
A Philips XL-30 Scanning Electron Microscope was used for obtaining the pictures. The most significant pictures (good as well as bad examples) are shown in Fig, 3 and 4. Fig. 3 shows examples of porosity and microdefects. Fig. 4 shows structural phases in porcelain. Table 3 refers to these figures.
X-ray Diffraction (XRD)
XRD analysis was done to quantify the phases present in different porcelain insulators. The Internal Standard Method was used with calcium fluoride as the standard. Calibration data was first developed using pure quartz, corundum and mullite. A small piece of the insulator was finely powdered in an agate mortar and pestle and mixed in a known ratio with calcium fluoride. X-ray powder diffraction was then done using a Siemens D500 diffractometer. Table 4 shows the different phases present in the insulators evaluated. Low strength siliceous porcelain insulators do not have any corundum. It is interesting to note that the corundum content is quite variable among the high strength aluminous porcelain insulators evaluated.
Microstructure of Insulators
Punctured insulators removed from service had the following features: agglomeration of corundum crystals (Fig. 4, insulator B1), long running micro cracks (as seen in Fig. 3, insulator E1), large quartz crystals (> 60-110 µm) and agglomerates of pores. Those insulators that had not failed had well dispersed and relatively few pores.
This research has demonstrated that there can be quite significant variations in the microstructure of porcelain suspension insulators. Other factors that determine performance in service are cement composition and operating conditions including: line voltage, position of the unit in the string and lightning severity. Existing standards for porcelain insulators help to specify minimum requirements. However, tighter specifications are probably desirable to ensure high reliability and long life of insulators on EHV and UHV transmission lines. The tests referred to in this article can be performed fairly quickly and therefore could be used as the basis for an upgraded technical specification.