Although the demand for reliable electric power continues to increase, today's economic and regulatory climate makes it difficult for utilities to even contemplate the construction of new overhead lines. This situation not only places a strain on existing power lines, it also makes de-energizing those lines for maintenance less desirable. As a result, utilities are relying more on the practice of working on energized power lines. By adopting live working, utilities can avoid power outages and the associated revenue loss, minimize the duration of power disruptions, increase system availability and enhance system stability and reliability.

For live work to be effective, it must be performed safely by maintaining proper distances between the worker and the grounded and/or energized parts of the line and the tower. This procedure can be difficult with compact towers, which are designed with smaller spacings between conductors and between conductors and ground than towers of traditional design. Safe live work on compact towers is especially challenging when the operating voltage and overvoltage levels, due to switching impulse levels, are increased as the required air distances must also be increased.

The solution lies in the installation of temporary voltage-limiting devices, known as portable protective gaps (PPGs), for the duration of the work. As tests at the Electric Power Research Institute's (EPRI) Power Delivery Center in Lenox, Massachusetts, U.S., (PDC-L) have shown, if PPGs are installed, utilities can perform safe live work on compact towers. PPGs are typically located on the structure adjacent to the structure on which live work is performed.

How PPGs Work PPGs, which are portable rod-to-rod air gaps, protect workers by ensuring that a flashover will not occur between the employee and the line but across the protective gap instead. The gap is typically installed on the adjacent structure. To be effective, PPGs must not operate _ that is, they must not spark over on ac. They must withstand ac voltage stresses at the work site. Conversely, PPGs must operate _ sparkover _ reliably on transient overvoltages (TOVs). If the above conditions are met, the PPG is "properly coordinated" with the work site voltage stresses and with the work site withstand performance. However, the sparkover performance of the work site is influenced by work site conditions, including the presence of the worker and tools in the air gap.

Compact 500-kV Towers When the Western Area Power Administration (Western), Golden, Colorado, U.S., undertook the California-Oregon Transmission Project (COTP) and upgraded nearly 200 miles (322 km) of line from 230 kV (double-circuit) to 500 kV (single-circuit), it also modified most of the 817 suspension towers along the line from traditional to compact design (Fig. 1). As a result, the minimum approach distance (MAD) _ the shortest permissible distance between any part of a worker's body or tools or the material being handled and the grounded tower _ specified by Western for live working on traditional towers was no longer adequate.

Research Project and Testing To develop criteria under which live work with PPGs could be performed safely on the upgraded transmission line, Western funded a comprehensive research and testing project at PDC-L. Work at PDC-L concentrated on two main issues:

Appropriate distances between energized and earthed parts. The minimum number of undamaged insulators required at the work site.

PDC-L performed TOV tests with a 5.6-MV outdoor impulse generator. Three front times of the TOV test wave shape were used: 100, 200 and 250 sec. The time to half-value was 3500 sec. The 50% switching impulse sparkover voltage, U50, was determined using the up-and-down method and the multilevel method.

In all, PDC-L performed more than 180 ac and switching impulse tests on a full-scale tower mockup for various work site scenarios, including tools and a model of the worker in the air gap (Fig. 4).

The tests involved various work site scenarios for both hot-stick and bare-hand live-working methods because the sparkover performance of the work site is strongly influenced by the details of work site conditions, which include the presence of the worker and the tools in the air gap. Indeed, the performance of the PPG, itself, is also influenced by the presence of the grounded tower structure. Fixed length hot sticks (no metal pins) and a plastic rib cradle were used to eliminate the effects of small floating electrodes.

Figure 2 shows a full-size test setup in which the inboard leg of the V-string (24 bells), with the cradle installed, is ready to be lowered to ground for replacement of damaged insulators. The worker (represented by the mannequin) is on a vertical ladder, and the worker and the detached insulator leg are electrically floating.

Figure 3 shows the worker, positioned on a horizontal ladder, detaching the outboard leg of the V-string. The insulator cradle is installed on the outboard leg, and broken insulator units are visible in both insulator legs. The position of the worker relative to the insulator string is important because the worker's body can short out several bells during work.

Test Results After reviewing and comparing the results of all tests, PDC-L personnel were able to identify the "worst-case" scenarios for live working. Proper coordination of the PPG was confirmed directly for these worst-case scenarios by applying 100 TOV shots at the appropriate voltage level and demonstrating that all sparkovers occurred across the PPG and none at the work site.

Broken Insulators The flashover voltage of an insulator string depends on the number of defective units in the string, their location within the string, and the severity of the damage. Shorted units represent the worst case of insulator damage. Units that are broken but not completely shorted retain some of their dielectric strength (Fig. 5).

PPG Coordination Sparkover data generated by the PDC-L tests demonstrates the reliable work site overvoltage-control capabilities of the PPG and the proper coordination of the PPG sparkover characteristics with the worst-case work site overvoltage stress. Using the probability scale on the abscissa (Fig. 6), the cumulative distribution of work site electrical strength, which is assumed to follow the Gaussian curve, is drawn as a straight line with slope equal to the standard deviation of 5% (s = 0.05). The lowest curve, marked "PPG," shows the experimentally determined average PPG sparkover data when installed in the COTP tower (point B, U50 = 641 kVo - g peak). Point A, the low probability, (0.14% or 3 below U50) shows that the PPG would spark over 0.14% of the time when it is subjected to a TOV of 545 kV. Point C shows that a TOV of 758 kV will cause a sparkover of the PPG 99.86% of the time.

The curve labeled "Work Site with Broken Units" gives the probability of flashover of the insulator string with 16 broken units in each leg, and it shows that the strength of the string is more than 300 kV higher than that of the PPG over the entire range of voltages. The withstand (the 0.14% point) of the string is 135 kV higher than the sparkover point (99.86%) of the PPG.

The tests results lead to the conclusion that, for this tower design and configuration operating at 550 kVo - o rms, the minimum electrical coordinating distance for bare-hand work is 60 inches (152 cm). This is consistent with the requirements in the National Electrical Safety Code (NESC). At this distance, the withstand (0.14% probability) of the work site with 16 broken units is equal to the sparkover voltage (99.86% point) of the PPG in the tower, which includes the worker "electrically floating" in the conductor-to-tower horizontal air gap with the ladder and tools in place. The tests also have shown that the distance of 60 inches (152 cm) is not adequate if the worker is grounded.

Improvement with the PPG The 41-inch (104 cm) rod-rod PPG used in these tests was designed for an operating voltage of 550 kV and would coordinate with a 60-inch (152 cm) electrical approach distance; however, no inadvertent movement factor is included in this figure. Therefore, to ensure the safety of workers moving onto the conductor, and to provide the inadvertent movement factor, Western added a distance of 30 inches (76 cm) to the MAD for bare-hand work. Although the NESC requires just 12 inches (30.5 cm), Western was able to add the longer distance for worker confidence because the geometry of the tower allowed it. Thus, the total MAD adopted by Western is 90 inches (229 cm).

The PDC-L tests show that the PPG will coordinate with up to 16 broken units. For the same MAD but without the PPG, the maximum number of broken units would be only nine.

Conclusions Data generated by the PDC-L test program enabled the participants to compare criteria for safe live-working conditions and to formulate specific, practical conclusions and recommendations regarding live work on the compact 500-kV COTP tower operating at 550 kV. The specific recommendations for live work on the tower include the MAD with and without the PPG and the maximum allowable number of broken insulators. Additional recommendations refer to live-working practices for the tower and provide guidelines for installation of the PPG.

These recommendations are applicable to live work on the specific COTP tower studied in this project and should not be generalized to other situations without a thorough re-evaluation.

The Known TOV Method (No PPG) The MAD adopted by Western for this method of live work is 135 inches (343 cm). Other considerations include:

The electrical component of the MAD must be adjusted for increases in elevation in accordance with the 1993 edition of the NESC and the U.S. Department of Labor, Occupational Safety and Health Administration Standard, 29 CFR 1910.269 - Electric Power Generation, Transmission and Distribution.

The maximum number of defective insulator units is nine

A hot line order (HLO), which prevents the circuit breaker on the line undergoing live work from reclosing in case of a trip signal, is required to be in effect to reduce the TOV on the line.

The Controlled TOV Method (PPG Installed) The MAD adopted by Western for this method of live work is 90 inches (229 cm). Other considerations include:

Altitude correction is not required.

The maximum number of broken insulator units adopted by Western for this COTP tower is 12 (this is 25% less than the number of broken units that still coordinates with the PPG).

The PPG should be placed horizontally on the top conductor of the bundle, conveniently away from the attachment hardware, and with the gap facing away from the nearest hardware and vibration damper.

The PPG needs to be installed only on the phase(s) to be worked.

An HLO is required to be in effect as an additional precaution.

Additional recommendations include:

Existing line terminal rod-ring gaps provide a degree of TOV control, but it is Western's recommendation that they should not be used without comprehensive engineering case-by-case analysis.

In the event that a sparkover of the PPG occurs during use, the gap rod tips should be examined for pits and polished if necessary. If polishing increases the gap spacing, the PPG should be replaced.

The possible presence of electrically floating electrodes at the work site should be kept in mind because the resulting dielectric strength of the air gap tends to be reduced. TDW

Hank J. Kientz enrolled in the School of Electrical Engineering at the University of Colorado while working full-time as regional maintenance manager for a national trucking company. He began his career as an electrical engineer in substation and transmission line design for Western Area Power Maintenance. Kientz is a senior member of IEEE and is active on the IEEE/ ESMOL Subcommittee, is chairman of the NESC Subcommittee 8 on Work Rules, and is a U.S. delegate to the IEC Technical Committee 78 on Live Working.

Harold J. Fox, Jr., received the BS degree in management from Rutgers-University College. He has been employed at Public Service Electric and Gas Co. for 36 years, during which he has been involved in transmission and distribution construction and maintenance. Presently, he is manager of transmission construction and maintenance. He serves as an alternate member on ANSI-C2, SC-8, is active in the IEEE and the Engineering in the Safety, Maintenance and Operation of Lines (ESMOL) Subcommittee and chairs several task force groups.

J. David Mitchell received the BSEE degree from the University of Alabama in 1971. Since graduation, he has worked for Alabama Power Co. in various areas, including distribution, transmission line construction, and transmission line design. Mitchell is presently employed as a senior engineer in the Power Delivery Transmission Line Department. He has held various posts at the local IEEE chapter and section levels and is secretary of IEEE/PES ESMOL Subcommittee, a member of Eta Kappa Nu, and a registered professional engineer in Alabama.

George Gela obtained the BASc, MASc and PhD degrees in electrical engineering from the University of Toronto. He has been project leader, High Voltage Studies at Trench Electric in Toronto, Canada, and a faculty member of the Power Group, Electrical Engineering Department, at Ohio State University. In 1990, he moved to EPRI High Voltage Transmission Research Center (HVTRC), now the J.A. Jones Power Delivery Co.-Lenox Center (PDC-L), where he is in charge of the research programs in live working, underground manhole events, underground secondary distribution cable faults; is responsible for development and support of electrical modules of the EPRI Workstations; heads other projects; and lectures at PDC-L seminars. Gela is the international chairman of the IEC Technical Committee 78 Live Working, a member of CIGRE and the U.S. delegate to WG 33.07 "Electrical Insulation in Live Working and Other Special Conditions in Electrical Systems." He is a senior member of IEEE.

Paul Lyons is a project manager in the Overhead Transmission Target in the Power Delivery Group of the Electric Power Research Institute (EPRI), Palo Alto, California, U.S. He is located at EPRI's Power Delivery Center in Haslet, Texas, U.S. As an EPRI project manager, Lyons is responsible for structure related research projects such as the conductor wind loads project and the failure containment project. He is also responsible for line maintenance related research projects such as reliability centered maintenance (RCM) for transmission assessment and inspection methods (AIM) project, transmission inspection and maintenance system and live working methods project. Lyons is the responsible EPRI project manager for the operations contracts for both the Power Delivery Center-Haslet and the Power Delivery Center-Lenox.

History of the PPG Substation entrance rod gaps, which have been used for years to prevent substation equipment flashovers caused by lightning, were the precursors of today's PPGs. The experience and data gained through their use proved helpful in estimating the performance of PPGs and developing confidence in their use for personal protection during live work.

The first PPGs were conceived and used in the late 1960s, but the first PPGs intended for personal protection during live work in the United States at 500 kV were developed in 1971. One of them, developed by A.B. Chance, Ohio Brass and several U.S. utilities, consisted of horn gap electrodes separated by 41 inches (104 cm) and attached to a live-line tool. Hydro-Quebec, Montreal, Quebec, Canada, developed and used a similar gap at the same time.

In 1981, Ontario Hydro, Toronto, Ontario, Canada, developed portable sphere gaps for use at both 500 kV and 230 kV. Since then a number of countries, including China, Russia and South Africa, have reported on the use of protective gaps.

Evaluating Sparkover Risk during Live Work A number of considerations have been taken into account both in the development of live-work procedures and in the establishment of minimum approach distance (MAD) values for portable protective gaps (PPGs). For live work the issue is evaluating overall risk of sparkover. For PPGs the issue is ensuring that they will always spark over at a voltage that is lower than that needed to cause sparkover of the air gap between the worker and grounded parts. These considerations are listed below.

The probability of each of these events occurring at the same time has been accepted as being statistically less than 1/10,000,000 for normal live-working procedures. The addition of the PPG significantly decreases the risk of sparkover at the work site.