Volume 9, Number 1 
Here is an interesting question. What methodology should a utility employ for selecting an insulator - particularly if it is a composite insulator - in order to ensure satisfactory performance under a given type of service condition?
Generally-speaking, several methods are available to help guide the insulator specification decision. Each has its own special characteristics as well as advantages and drawbacks. Ultimately, the final choice depends mainly on the resources available to the utility and the period of time required to obtain the results.
The various available methods for making a selection could be divided into two broad categories: in one category, the performance of the insulators is carefully measured by employing a series of accelerated ageing tests conducted in a laboratory and simulating actual service conditions; in the second, the insulators are evaluated simply by installing them under real service conditions.
The evaluation of insulators through accelerated ageing tests requires that a utility has its own fairly sophisticated laboratory. In the absence of such, the utility will have to rely on testing done by independent laboratories. This evaluation, based on a series of standarized tests developed by experts from within CIGRE and IEC, might also employ additional in-house, state-of-the-art tests. It is not simple to analyze the results of these many tests and therefore this requires experienced researchers.
The complete laboratory evaluation of a composite insulator is fairy expensive and can typically be achieved in about one year or less, depending on the nature of accelerated ageing test which has been chosen. Because such testing can be completed and the findings examined and interpreted in a fairly short period of time, it is quite likely that the insulators tested are identical to those still available on the market.
By contrast, the method which calls for trial installation of a group of insulators on a selected overhead line in a given service environment requires a very long time to complete. There is, by definition, no acceleration of the normal ageing processes and it is not reasonable to expect any meaningful results in less than 5 to 10 years at a minimum.
The method also relies on periodic line inspections in order to properly monitor the ageing process of the insulators - a requirement which incidentally has the benefit of training utility personnel. Reliable results are obtained only after a long period of time, although of course at a far lower cost than accelerated ageing. This procedure also ensures that the test conditions are in fact the real service conditions although test results are then applicable only to lines in similar service conditions.
Because of the long test duration, by the time the results are finally available and analyzed, the tested insulators may or may not even be available on the market. Perhaps they may have been replaced by newer and improved (at least according to the manufacturer!) models.
A variation of the evaluation via trial line installation is for the utility to build an outdoor test station in a specific environment. The site could be chosen to represent service in any given type of pollution. Only the test voltage needs be varied in order to obtain a certain acceleration factor. The period of time required to obtain meaningful results from this process is more than one year but probably less than that needed for the trial installation.
Finally, there is yet another method : i.e. performing no testing at all. The utility simply installs the composite insulators directly onto their network and hopes that they will perform sufficiently well.
Except for this last method, all the above roads require a proper evaluation of the insulators before making any decision for extended use on the network. Each method therefore replies on a profound knowledge of composite insulators because the interpretation of test results is probably the most difficult part of all the methods.
Dr. Claude de Tourreil
