Sampling Based Maintenance: Assessing Condition of Composite Insulators After Years of Service

Insulators, Maintenance

With their growing popularity as the insulator of choice for many line projects throughout the world, the ideal inspection program for silicone insulators has become a hot topic. This is particularly the case since, even after years of investigating alternative proposed techniques and equipment, there still seems no consensus on how to conduct such work in the most effective and cost-efficient manner. This past INMR article, contributed by Prof. Liang Xidong and Wang Jiafu of Tsinghua University in Beijing, Shen Qinghe of the Shandong Electric Power Research Institute and Zhang Yibo and Li Yan of the China Southern Power Grid’s Technology Research Center, presented a different approach to those discussed within CIGRE and IEEE – namely a system of maintenance based only on results of sampling and testing insulators removed from the line.

Maintenance of line insulators is a key issue for power utilities and, in the case of China, removal of pollution and detection of punctured cap & pin porcelain discs have typically been the two main tasks involved.

According to Chinese regulations, porcelain and glass insulators must be cleaned once a year in medium pollution areas and twice a year in areas suffering heavy pollution. In most places within the country, rain occurs mainly during the summer and the pollution layer begins to build up from late autumn through to the beginning of spring. Pollution flashovers typically occur on winter days having dense fog (such as the serious flashovers affecting the north-east in February 2001) and the most effective time for pollution cleaning is therefore ideally also during the winter.

In the case of transmission lines, cleaning insulators has usually been carried out manually using rags in what is a difficult and demanding job. The quality and consistency of cleaning is often not satisfactory while in most places it is now difficult and sometimes even impossible to arrange for the outage needed for this work.

Chinese regulations also mandate inspections to detect punctured cap & pin porcelain insulators at least once every two years. In the case of glass insulators, shattered units are easily detected from the ground and therefore the application of these is often preferred over porcelain. However, it does not solve the problems historically associated with periodic manual cleaning.

Given this situation, silicone insulator technology has made rapid strides within China. However, this has also raised the important question of when and how these insulators should be maintained


Maintenance Requirements for Silicone Insulators

Silicone insulators are now widely used throughout China on both existing and new transmission lines that pass through areas with pollution. Due to the hydrophobicity transfer property of silicone rubber, pollution flashover accidents such as have frequently occurred with porcelain and glass insulators have been virtually eliminated and, because of this, there has been no need for pollution cleaning for many years now.

As for detection of punctured insulators, this has also no longer been necessary in the case of long rod composite insulators. This is because risk of puncture of the external housing-to-rod interface can be virtually eliminated by increasing bonding quality through wellcontrolled manufacturing practice. Moreover, most Chinese utilities prefer not to search for such punctures in the field, even though devices such as the electric field tester could in principle detect these.

Because their introduction helped eliminate costly and labor-intensive pollution cleaning and detection of punctured insulators, silicone composite insulators were originally promoted as being ‘maintenance-free’. Of course, such a claim was misleading, especially since experience with these insulators is still relatively limited. That means that some form of inspection is needed, if nothing else than to monitor them for possible changes in performance over time and to accumulate service experience.


Sampling Result-Based Maintenance

A proposed maintenance method for silicone insulators could be called “sampling based maintenance” and this concept is illustrated in Fig. 1.

Sampling Based Maintenance: Assessing Condition of Composite Insulators After Years of Service
*From the second maintenance-free period on, the length of each subsequent maintenance-free period would depend on the sampling result at the end of the previous period.
Fig. 1: Illustration of sampling based maintenance.

According to this maintenance philosophy, for the first 5 ~ 8 years after silicone insulators are put into service, there would be no inspection at all and therefore this could be referred to as the first maintenancefree period. The relative length of this first period (i.e. whether closer to 5 or to 8 years) would depend on the severity of the service environment as well as the manufacturer’s product quality, based on experience.

At the end of this first period, several insulators would be sampled from the line. These would then be tested in the laboratory for hydrophobicity, pollution level and electrical as well as mechanical performance.

If these test results are positive, another 5 ~ 8 year period can go by without need for inspection, i.e. a second maintenance-free period. After this second period, similar tests would be conducted on insulators sampled from the line. The relative length of the third and fourth maintenance free periods would ideally be shorter than the first two periods because operating experience for this total time frame is still relatively limited.

However, if test results on the sample insulators after the first maintenance-free period prove unsatisfactory (e.g. there is a significant decrease in hydrophobicity, obvious surface ageing or a decrease in mechanical strength), the second maintenance-free period would be shortened and additional inspections, such as with infrared and ultraviolet equipment, could be considered for any insulators whose behavior seems uncertain. Such non-destructive in-service monitoring methods could then be regarded as a valuable supplement to sampling result-based maintenance.

In some provinces of China where local utilities have sufficient experience, the first maintenance-free period has actually been extended to 8 ~ 10 years while the second period has been increased to 6 ~ 8 years. The key factors to be emphasized in this regard are that the manufacturer’s quality control program must be strictly observed and verified and also that damage during storage, transport or installation must be carefully avoided.


1. Relative Lengths of Maintenance-Free Periods

Deciding on the appropriate length of the maintenance-free period in each case should ideally depend on the line’s environmental conditions. For example, while for most polluted areas the first maintenance-free period could be set as 5 ~ 8 years, in those Chinese provinces having relatively light pollution (classes from A to C), this has now been extended to 8 ~ 10 years, based on past experience.

By contrast, for service areas suffering severe pollution (class E as a result of cement, lime, aluminum, alkali, metal smelting, etc.), the first maintenance-free period could be set at less than 5 years. Similarly, 5 years could also be the maintenance-free period for lines passing within 1 km of a coastline. The relative lengths of the second and subsequent maintenance-free periods should then be selected based mainly on results of the sampling tests of the previous period.

2. Test Items

At the end of each maintenance-free period, insulators need to be selected for sample tests, ideally conducted in the laboratory and not the field. These would include visual examination, hydrophobicity and pollution measurements as well as testing for electrical and mechanical performance, among others.

According to national regulations, porcelain insulators must be cleaned and checked for punctures on a regular basis.

2a. Visual Examination

Much information can be obtained through visual inspection of the insulator surface, including condition of the pollution layer, possible damage due to erosion or ageing on the sheath and sheds (e.g. chalking, hardening, discoloration). In addition, a careful check should be made for signs of possible slippage of the end fitting and sealing system. Finally, the housing could be manually folded.

Damage detected on either the housing or the seals will mean there is an increased possibility of brittle fracture. Moreover, evidence of slippage of the end fitting not only risks severe damage to the sealing system but also a reduction in mechanical performance. Attention should also be paid to the sheds and locking devices, the pollution layer and the condition of the corona ring.

2b. Hydrophobicity

The hydrophobicity of the housing and sheds should be examined in a laboratory and measured at three places – the high voltage end, the ground end and at the mid point. When hydrophobicity falls in the range of HC1 ~ HC4, this would be judged good because the flashover voltage would then be much higher than the operational voltage.

An HC of 5 would be considered as medium performance while HC6 or HC7 would mean that all hydrophobicity had been lost and the insulator was now at increased risk.


2c. Measurement of Pollution

Pollution measurements on insulators sampled from the line would serve to evaluate their pollution condition and measurement points to collect ESDD and NSDD would be the same as for the tests on hydrophobicity.

2d. Electrical Performance

In some parts of China, electrical performance is tested on sampled composite insulators having natural pollution. In these cases, the difference between the results of dry and wet power-frequency flashover voltage must not be greater than 30%.

A steep front impulse voltage test is sometimes also performed to check the quality of the sheath-rod interface.


2e. Mechanical Performance

The mechanical performance of silicone insulators can decrease over years of service. Therefore shorttime mechanical failure loads should be tested on sampled insulators to determine the average residual mechanical strength of insulators still in service. Special attention should be paid to any insulators where mechanical performance has fallen to the specified mechanical load (SML) value.

According to experience from China, crimped composite insulators and those with inner-wedge designs have shown excellent mechanical performance over time, i.e. their remaining mechanical strength will generally be higher than SML even after 10 years’ service. The mechanical performance of outerwedged insulators, however, is usually not so stable and deviations in mechanical failure loads can prove unacceptable. As such, no outerwedged insulators have been used on Chinese transmission lines for the past ten years.

Depending on region, dynamic tensile and rotational mechanical performance is sometimes also tested to ensure the complete safety of insulators in service.

2f. Additional Inspections

Infrared (IR) and ultraviolet (UV) inspection are usually conducted as an additional test for any insulators with damaged housings or where there is doubt about their condition. Greater attention should then be given to testing any such insulators where either an elevated temperature profile or corona discharge is observed.

Steps Following Sample Testing

Maintenance decisions should be taken after sampling inspection and these can be classified into one the following typical situations:

• If test results on sample insulators are found to be good, there could be no deviation from the planned duration of the subsequent maintenance-free period.

• If only a few of the insulators sampled are sufficiently good or are found to contain slight visual flaws, the next maintenance-free period could be maintained but with special attention to any aspects of possible concern. The number of units sampled can also be increased to gain more information.

• If the appearance of sampled insulators is unsatisfactory or results are not as expected in many of the units, more tests should ideally be carried out to find the reasons. The subsequent maintenance-free period should therefore be shortened and close attention paid to the insulators.

• If the mechanical or electrical performance of sampled units has decreased significantly (in relation to the insulators when new), careful attention should be paid to the insulators and a program set-up to replace them in step-by-step manner.

• If the sample insulators are found to be seriously damaged, the ongoing safe operation of the line can no longer be guaranteed. In this case, all the insulators should be changed out as soon as possible. At the same time, various measures should be taken to maximize safety around that line.

Other Aspects of Sample-Based Maintenance

1. Inspection Season

Because of more favorable weather, the collection of samples for inspection should ideally be done in spring or autumn together with scheduled routine line inspection. In China, pollution accumulation by autumn is much less than that in spring due to washing by summer rains. Attention must therefore be given to this expected difference in pollution data for the two seasons.

2. Measuring Sites

In theory, the location of measurements should be selected at random. Still, many utilities tend to select measuring spots in the worst environments along a line to ensure that units in service in these sections are not somehow overlooked.

3. Selection of Sample Insulators

Ideally, three sample insulators should be selected for testing from each measuring site.


Sampling result-based maintenance is one of the most suitable and costeffective methods to monitor the condition of silicone insulators in service and has been used in China for years now with satisfactory results.

According to this concept, the duration of each subsequent maintenance-free period should be selected based on the results of testing samples taken down at the end of the former maintenance-free period. The first maintenance-free period could be set at 5 ~ 8 years, as now used in most of China.

This method is dynamic, meaning that it is based on accumulated experience and all sample tests should be carried out in laboratories and not in the field. Visual appearance, mechanical performance, hydrophobicity and electrical characteristics are the typical parameters tested and additional IR or UV inspections may be required if the findings prove not to be conclusive.

Corrective measures may have to be taken after sampling-based inspections. Seriously damaged insulators should be changed out as quickly as possible while close attention should be paid to insulators where tests from samples indicate slight damage.




Global Insulator Group manufactures glass, composite, porcelain insulators and line fittings. Design and excellent performance of our products are proven by more than 60 years of operation experience. Global Insulator Group has extensive sales network and supply insulators to more than 100 countries in the world. The company is aimed at the development of high-tech production, innovative technical solutions and enforcement of trust for business partners in the global market.