The electric network in the eastern region of Saudi Arabia operates under harsh environments that include desert, marine and industrial pollution, along with high ambient temperatures. Such conditions make insulator-contamination flashovers the main concern for the electric utility.

To control the pollution problem, the utility tried many solutions that significantly reduced flashovers, but those solutions increased acquisition and maintenance costs of the insulators. The utility began searching for an insulator with better pollution performance at a low life-cycle cost (LCC). Silicone rubber (SiR) insulators appeared to be the most promising solution because of their generally acknowledged superior pollution performance. However, their performance had to be evaluated under different pollution conditions prior to application, because actual field conditions cannot be accurately simulated in the laboratory. Therefore, with the main objective being the improvement of the power transmission system reliability at minimum cost, the SiR insulators were installed on a 230-kV line in desert area of Saudi Arabia to test their performance.

Insulator Performance Trial Program

Insulator pollution performance is the most important factor in selecting transmission line insulators in the eastern region of Saudi Arabia. To maintain reliable electric power service under harsh weather conditions, the Saudi Electricity Co. (SEC) performed extensive live-line high-pressure water washing of insulators, increased the specific creepage distance of insulators, and used different insulator designs and materials. These efforts resulted in significantly reduced insulator contamination flashovers, but these measures were expensive, so SEC investigated cheaper solutions.

The performance of ceramic insulators in desert environmental conditions is satisfactory compared to their performance in other areas such as coastal and industrial. However, the expected performance of SiR insulators is not well known because hydrophobic characteristics change with time and fluid conditions. Therefore, in 1995, SiR insulators were installed in the eastern region desert of Saudi Arabia to evaluate their performance over a six-year period. They were compared to ceramic insulators and the former porcelain desert long rod (DLR) insulators installed on the same line. Insulator washing (a major part of LCC) is not required in this desert, thus, the expected costs will be low compared with that for other areas in this region.

Test Site and Test Line

SEC selected the Faras Power Plant — Khurais 230-kV line as a test line for the trial because the complete line is located in desert. The prevalent pollution is characterized by blown salt-laden sand and dust in a dry atmosphere. The temperature varies from 10°C to 22°C (50°F to 72°F) between day and night during this hot-weather season extends more than six months each year.

The performance of insulators installed on this 131-km (81-mile) line, erected on 310 towers some 100 km (62 miles) inland from the Arabian Gulf, was considered to represent their performance in desert conditions. This 230-kV line was insulated with DLR insulators when constructed in 1981 and re-insulated with SiR insulators in February 1995. DLR insulators were installed on many 230-kV transmission lines during the early 1980s in inland and coastal areas. They performed well in inland areas where humidity is low without being washed, and they were cleaned by prevailing winds. For coastal applications, however, they are being regularly washed to control the contamination-related flashovers.

The test line experienced five failures of the DLR insulators between 1981 and 1995. Most were not pollution related, but each time this line tripped, more than 200 MW of industrial load was lost from the power system. The lost load mainly affected water pumping stations and some industrial load from oil and gas production facilities. Reliable electrical power is critical because oil is the main source of income for the country, and its production continuity is vital for the world.

The performance evaluation study of SiR insulators that continued until May 2001 was conducted in two parts: a field trial on an in-service transmission line, followed by laboratory testing.

Field Trials

Prior to installation, all composite insulators were inspected for any damage to the sheath or to the end seals where the rod enters the end fitting. The damage could result in moisture ingress to the housing, causing the insulator to fail electrically. Any composite insulator with even minor damage, cuts or indentations on the composite material surfaces (shed or sheath), or with its fiberglass rod exposed, was discarded and replaced. The close proximity of the insulators elevated patrol inspections that were conducted via bucket truck or climbing towers. The line was energized and all damaged SiR insulators were replaced.

Day and night inspections of all the test insulators were made monthly. The insulators were observed for audible noise from corona discharges, partial breakdowns, discoloration and mechanical abnormalities. Binoculars were used as required, and a night viewer aided nighttime corona inspections.

The results of the field trials confirmed that the performance of SiR insulators was satisfactory. No electrical and mechanical failures occurred, and detailed and precise inspection did not reveal any mechanical damage, cracks or erosion in the rubber housing. Careful handling during installation of SiR insulators contributes to longer service life.

Laboratory Investigations

Test insulator strings were removed from the line and sent to King Fahd University of Petroleum and Minerals (KFUPM; Dhahran, Saudi Arabia) in addition to the manufacturer facilities and universities abroad to perform chemical, electrical, mechanical and physical tests.

Three insulator samples (A, B and C) installed on the Faras Power Plant — Khurais 230-kV line were considered adequate and representative for each laboratory test. Samples A and B were removed in March 1998 and May 2001, respectively, and Sample C in May 2001. Two samples were sent abroad for the electrical, physical and chemical tests. The third sample was subjected to electrical and chemical tests at the KFUPM test facility.

A new insulator sample, designated as Sample R, also was tested at manufacturers' facilities for physical and chemical properties only for comparative purposes. For electrical and mechanical strength tests, the insulator's rated values were used for comparison with the test results from the other samples.

• Electrical Tests were performed on the naturally polluted insulator Samples A, B and C, in accordance with the IEC 507 Clean Fog Method. To fulfill the requirements of IEC 507 (minimum short-circuit current), and because of the limitation of the test plant, the insulator Samples A, B and C were partially short-circuited to obtain flashover and withstand level.

  The wet power frequency flashover voltage (U_50%) for Sample C is lower than that of other samples (Fig. 1). This is attributed to the higher equivalent salt deposit density (ESDD) level of Sample C (Fig. 2), which is almost twice that of Samples A and B. As a rule, the higher the insulator contamination, the lower the flashover voltage. However, the measured flashover voltage values still demonstrated good pollution performance of all three test samples. They are well above the 825-kV-rated wet power frequency flashover voltage, and the risk of pollution flashover is still low.

• Pollution Measurement. The ESDD and non-soluble deposit density (NSDD) were measured. Figure 2 shows that the pollution level had increased with the time, as the pollution level measured at KFUPM for Sample C is higher than that measured abroad for Samples A and B. The lower levels could be because of the loss of some of the deposits during transportation of the insulator abroad, as both Sample B and Sample C were removed from the line on the same day.

  Measurements of insulator pollution levels are vital because pollution could affect insulator performance.

• Mechanical Test. The aim of the mechanical time-load test was to check the mechanical strength of the insulator for any reduction from the specified mechanical load (SML) at different in-service years. Insulator Samples A and B were subjected to a failing load test per IEC 1109, and there was no failure at 100% SML. The socket failed in both cases at load, but the value of failing load in each case was well above the rated 90 kN SML value (Fig. 3).

• Physical Properties were evaluated by measuring hardness, roughness and water drop receding contact angle of SiR material of test Samples R, A and B.

  The hardness was monitored to evaluate the possibility of the rubber cracking. Measurement results showed an increase in hardness that could be attributed to oxidation or aging by ultraviolet (UV) radiation, or the result of mineral-filler migration.

  Surface hydrophobicity of the shed material of Samples R, A and B was classified as HC1, in accordance with the STRI Classification Guide 92/1. Figure 4 shows water drop receding contact angles as measured for each sample. The measurement results proved that hydrophobicity of the tested samples was intact.

  Generally, the surface roughness of an insulator will influence the hydrophobicity in a different manner depending on the materials. If the receding contact angle is below 70 degrees, a change in surface condition is possible. This is because the low molecular silicone will migrate to the surface and cover even the microscopic surface peaks of the pollution layer, making them hydrophobic.

• Chemical Tests were performed to diagnose the rate and extent of aging of SiR material. A Fourier Transform Infra-Red Spectroscopy (FTIR) test was performed to see if decomposition of alumina tri-hydrate (ATH) occurred as a result of aging. The Electron Spectroscopy for Chemical Analysis (ESCA) technique was used to analyze all the elements on the uppermost layers of the insulator surface. The Scanning Electron Microscopy (SEM) technique was used to study the surface structure and its roughness. SEM coupled with Energy Dispersive X-ray (EDX) was used to measure the chemical composition of the surface area to check for any decrease in the low molecular weight silicone (LMWS) and to recognize contamination. These tests did not show any appreciable degradation of SiR material.

Table 1. Reliability Indices of Reference and Test Insulators

Insulator	Number of Strings	Years in Service	Number of Failures	Failure Rate	Reliability Index
Reference insulator	1134	14	5	3	0.25
Test insulator	1134	6	Nil	0	1

Table 2. Reference and Test Line Data

Item	Reference Line	Test Line
Length	131 km (81 miles)	131 km (81 miles)
Voltage	230 kV	230 kV
Structure type	Latticed steel	Latticed steel
Insulator material	Porcelain DLR	Silicone rubber
Total structures	310	310
Total strings	1134	1134
Replacement work cost per string	*SR273 (US$72.8)	SR205 (US$54.7)
Insulator life	30 years	25 years
Date commissioned	1981	1995
* US $1 = SR3.75

Power Transmission System Reliability Index

The power transmission line faults are caused by many reasons like lightning, contamination or mechanical failures. However, it can be safely assumed that more than 90% of the line interruptions are because of insulator failures. Therefore, the reliability of the transmission line is directly proportional to the reliability of the insulators.

Traditionally, insulator failure rate (FR) is defined as the number of insulator failures per 10,000 insulators per year. In accordance with international practice, the failure rate acceptable to the electric utilities worldwide is less than one. Failure is defined as the loss of ability of a device to perform any of its intended functions. Therefore, actual insulator failures or flashovers, even if the insulation is self restored, will be counted as insulator failures.

The following equation was developed to calculate the reliability index (RI) from the failure rate of used SiR insulators as well as other used insulators:

RI = 1 / (1 + FR) where RI = 1

FR = Number of failed insulators per 10,000 insulators per year

The minimum acceptable value of reliability index is 0.5, corresponding to a failure rate of one as mentioned above. It is best if the failure rate is zero, which corresponds to the reliability index of unity. Therefore, it can be said that the reliability index closer to unity means better power-system reliability.

Five DLR insulator strings failed in 14 years of service, while SiR insulators did not fail in about six years (Table 1).

The reliability index of 0.25 for DLR insulators is low compared to what is generally acceptable to electric utilities worldwide, which is above 0.5. However, the unity reliability index for SiR insulators reveals their excellent performance.

Cost Effectiveness of SiRs

The cost-effectiveness of the test insulators was determined by comparing the LCC of these insulators with the LCC of the reference insulators. In general, the LCC of an insulator of an overhead transmission line is defined as the sum of the purchase cost, initial installation cost and lifetime maintenance cost (line-wash cost), over the expected life of the insulators. Insulator maintenance costs other than wash costs are negligible. Reference and test lines data are shown in Table 2.

The LCCs of the insulators are shown in Table 3 and calculated as follows using a 30-year life for DLR insulators and a predicted life for SiR insulators of 25 year:

LCC = Purchase Cost + Replacement Cost + Lifetime Wash Cost

Also, each cost component as a percentage of LCC and LCC saving given by use of SiR insulators, are shown along with LCC comparison in Table 3.

The above calculations show that the life-cycle cost of DLR insulators is about 1.5 times that of SiR insulators. This means that the saving is the result of the difference in purchase cost (89% of LCC for DLR insulators and 85% of LCC for SiR insulator). Nevertheless, other significant advantages of SiR insulators include easier installation, transportation and improved power-system reliability.

Because of the technology of SiR insulators is relatively new, this development could still have greater advantages over porcelain insulators as the benefits from improvements in materials and design and the economies of mass production materialize. These potential benefits seem likely to make SiR insulators the most cost-effective type of insulator in future, even for applications in desert environments.

More Studies Needed

There was not a single incident of mechanical or electrical failure of the SiR insulators. Equally important, the LCC was lower than that of the reference insulators.

SiR insulators may not yet be the best choice for inland desert areas. More field studies are necessary to confirm that their inherent service life is equal to, or longer than, that of ceramic insulators.

Table 3. Life-Cycle Costs Percentage Distribution and Comparison

Item	DLR Insulators	SiR Insulators
Life-cycle cost	SR2,824,600 [US$753,230]	SR1, 578,130 [US$420,840]
LCC per string/year	SR83 [US$22.13] (R1)	SR55.7 [US$14.85](R2)
Purchase cost	89% of LCC	85% of LCC
Replacement work cost	11% of LCC	15% of LCC
LCC saving in 25 years	(R1-R2) × 25 × 1134 = SR773 955 [US$206,388]
R1/R2	1.5

Dr. Ibrahim Yousif Al-Hamoudi received a BSEE degree from Basrah University in 1976 and was awarded a MS degree by Bath University (UK) in 1999. Al-Hamoudi joined Saudi Electricity Co. (Eastern Region Branch) and has held several managerial positions in generation, system operations and power transmission. Currently, he is vice president, Consolidated Transmission Area, Eastern/Central Saudi Electricity Co. Al-Hamoudi has contributed technical papers to several international professional conferences.

hamoudi.iy@se.com.sa

Dr. Zakariya Mahmoud Al-Homouz was awarded a BSEE degree from Yarmouk University (1986), an MS degree from Jordan University of Science & Technology (1989), and a Ph.D. in electric power engineering from King Fahd University of Petroleum & Minerals (KFUPM) in July 1994. Al-Homouz started his teaching career at the Jordan University of Science & Technology in 1986 and moved to KFUPM in 1989. In 2000, he was appointed associate professor in the electrical engineering department. Al-Homouz performs research, contributes to technical papers, and is a member of professional engineering institutions and societies.