A typical segment of the right-of-way for the National Grid’s A1/B2 69-kV line.
A typical segment of the right-of-way for the National Grid’s A1/B2 69-kV line.
A typical segment of the right-of-way for the National Grid’s A1/B2 69-kV line.
A typical segment of the right-of-way for the National Grid’s A1/B2 69-kV line.
A typical segment of the right-of-way for the National Grid’s A1/B2 69-kV line.

To Protect and Deter

Aug. 27, 2014
National Grid USA installs avian retrofits and reduces unknown transmission outages from 28 to 2.  

National Grid USA transformed its lowest-performing 69-kV line into one of its highest-performing lines by implementing an avian-retrofit and re-insulation program on the basis of strong circumstantial evidence. The utility targeted a 54% reduction in all outages and, instead, achieved an 81% reduction.

Outages without a known cause, or an unknown outage, had accounted for nearly two-thirds of all faults and were reduced by 92%. The benefits to National Grid included improved reliability, lower costs and decreased environmental liability.

National Grid’s 69-kV A1/B2 transmission line in Massachusetts, New Hampshire and Vermont.

The Line

National Grid owns and operates electrical infrastructure serving 3.3 million customers in Massachusetts, New Hampshire, New York and Rhode Island. National Grid’s 69-kV A1/B2 line conveys power from the Vernon Hydroelectric Power Station in southeastern Vermont to load centers in central Massachusetts. Built between 1907 and 1910, the 47-mile (76-km) A1/B2 line is the oldest transmission route in New England. As the line approached a century of service, 500 of the 615 structures on the A1/B2 main line were steel lattice towers with a single overhead shield wire and two three-phase circuits supported by 20-inch and 24-inch (508-mm and 610-mm) insulators rated at 69 kV and 88 kV, respectively.

From 1997 through 2003, 298 line faults were recorded on the A1/B2 main line, making it the utility’s worst-performing 69-kV line, with a disturbance rate more than double the utility’s average. National Grid crews inspected the line after each fault to determine the cause, assigned only when physical evidence exists. In the outage management system (OMS), 43 of the A1/B2 faults (14%) were attributed to lightning and 61 faults (21%) were attributed to a variety of other known causes such as weather, line equipment, vegetation and, in one case, an animal. In contrast, 194 faults (65%) were attributed to an unknown cause.

Turkey vulture carcass caught between triangular perch discourager and conductor on the unretrofitted A1/B2 line. (Photo taken between 1996 and 2004.)

Avian Issues

Despite only one animal outage record, avian issues historically were a concern on the A1/B2 line, and bird activity was noted in many of the OMS unknown outage entries. Avian outages and electrocutions can occur when a bird, its feces or nesting material touches differently energized contacts (phase-to-phase) or an energized contact and a ground (phase-to-ground).

The Avian Power Line Interaction Committee (APLIC), an industry group dedicated to addressing avian electrocutions, recommends energized contacts be separated from grounded or other energized contacts by 60 inches or 40 inches (1,524 mm or 1,016 mm) horizontally for eagles and hawks, respectively, and by 40 inches vertically. Because A1/B2 conductors had just 36 inches (914 mm) of horizontal clearance and 20 inches to 24 inches of vertical clearance, the line was recognized as susceptible to avian outages.

Triangular perching discouragers were installed in 1996 on approximately 40% of the structures to improve reliability and address liability concerns related to the Bald and Golden Eagle Protection Act and the Migratory Bird Treaty Act; however, outages persisted.

Lightning outages and unknown outages by month (from 1997-2003) and mean annuals days with lightning in Franklin or Worcester County, Massachusetts. Courtesy of National Lightning Detection Network.
Hourly lightning outages (with a three-hour moving average) and unknown outages (with a five-hour moving average) from 1997-2003.

Analysis

In 2003, National Grid engineers initiated a project to improve reliability on the A1/B2 line. They elected to re-insulate the lattice towers with 32-inch (813-mm), 115-kV post insulators to improve lightning performance. Because of the long-standing avian concerns, EDM International Inc., a consulting firm specializing in avian protection and wildlife outage prevention, was contracted to assess the line’s vulnerability to avian-caused outages. National Grid directed EDM to assess whether the predominant cause for the unknown outages was lightning or birds, and to provide avian-mitigation recommendations, if warranted.

Lightning was ruled out as the primary cause of unknown outages because the temporal patterns did not match at either seasonal or daily time scales. Whereas regional lightning data and lightning outages peaked in the summer months, unknown outages peaked in May and October, months with comparatively little lightning. Whereas lightning outages occurred in the afternoon and early evening, unknown outages were common at 5 a.m., peaked mid-morning and gradually declined until 9 p.m. Lightning may have contributed to the unknown outages, but it clearly was not the predominant cause.

Further Avian Considerations

APLIC’s avian-friendly spacing recommendations are based on the dimensions of large birds likely to perch on utility structures. The species of greatest concern in the A1/B2 area were large birds such as the bald eagle, red-tailed hawk, turkey vulture and osprey. These species range in height from 21 inches to 28 inches (533 mm to 711 mm), whereas the existing insulators provided 20 inches to 24 inches of separation. These birds also have wingspans of 56 inches to 96 inches (1,422 mm to 2,438 mm). Though great blue herons are less likely than other species to perch on utility structures, they are highly vulnerable when they do because of their maximum height of nearly 4 ft (1.2 m). Massachusetts breeding bird surveys also indicated populations of all these species were increasing.

The seasonal spring increase in unknown line faults beginning in May was consistent with nesting, breeding and young-rearing for raptors and other large birds. These activities heighten birds’ exposure to electrocution. The September and October increase in unknown outages generally correlated with fall migration along the Atlantic Flyway and was consistent with the specific migration timing for at-risk raptor species. In the fall, area raptor populations — composed mostly of year-round residents — are augmented by northern populations migrating south.

Turkey vultures were a species of particular interest because the species caused a previous outage on the A1/B2 main line. Turkey vultures establish communal roost sites in large trees and transmission structures, and ride thermal air currents as they search for carrion. They typically roost until late morning and spread their wings to warm as they wait for thermals, increasing their exposure to electrocution. Because the late-morning unknown outages peak was compatible with turkey vulture activity, avian electrocutions could not be ruled out as a cause of the unknown outages.

Streamers (long feces released by large birds that can bridge the air gap between conductors and grounds) were another potential outage cause. A streamer outage can be difficult to detect, especially if it does not leave burn marks and the bird is not electrocuted. Streamer outages typically peak in late evening and early morning, consistent with the large number of outages around 5 a.m. The A1/B2 line was particularly vulnerable to streamer faults. Even a moderate-sized streamer released from atop the 20-inch or 24-inch insulator could cause a flashover between the conductor and grounded tower. The outage timing and structure vulnerability indicated avian streamers were a potentially important outage cause.

In April 2003, after flying the line, EDM conducted carcass searches at 15 A1/B2 structures selected based on observed bird use, habitat setting or previous unknown outages. The findings indicated avian electrocutions were a much larger issue than the OMS had suggested. Eight carcasses were discovered in the vicinity of seven structures: four hawks of indeterminate species, two turkey vultures, one unidentified raptor and one great blue heron. One hawk and a turkey vulture had burn marks characteristic of electrocution. The other turkey vulture was electrocuted adjacent to a triangular perching discourager installed in 1996.

An electrocution rate could not be estimated because bird carcasses may either be rapidly scavenged or persist for years. The prevalence of mortalities could not be extrapolated to the rest of the line because only high-risk structures had been inspected.

Partially retrofitted A1/B2 structure. Note the 115-kV insulators, perch discouragers and conductor covers on the retrofitted circuit behind the linemen.

Mitigation

Avian concerns centered on the proximity of the crossarm-mounted conductors to the grounded lattice structure because of the following:

•A large bird perched on the crossarm was at risk of phase-to-ground electrocution

•Triangular perch discouragers were likely to shift birds closer to conductors

•Streamers released from the short insulators easily could cause flashover.

EDM recommended two avian-specific retrofits be implemented during the re-insulation project:

1. Any existing triangular perch discouragers should be removed and replaced with Mission Environmental Products’ Bird Guard spiked perch discouragers, which would more effectively minimize perching on the hazardous crossarm.

2. Kaddas Enterprises’ KE1039 conductor covers should be installed on the four crossarm-mounted post insulators. The conductor covers were intended to minimize the risk of streamer flashovers by shielding the phase wires and to insulate the conductors from birds attempting to defeat the perch discouragers.

By increasing the vertical clearance between the grounded crossarm and the conductors from 20 inches to 24 inches to 32 inches, the re-insulation project also was expected to reduce avian streamer and electrocution outages.

Within National Grid, there were several competing hypotheses to explain the A1/B2 line’s performance. In addition to lightning and possible avian interactions, there was speculation that vegetation or insulator contamination could be major contributing factors. National Grid hoped new perch discouragers and conductor covers might reduce unknown outages by one-third. On that basis, the utility chose to implement the recommended avian retrofits at a projected cost of US$1.6 million.

The combined re-insulation and avian-mitigation project was completed in 2004 at a total cost of around $4 million. Approximately 500 steel lattice structures were retrofitted at a cost of about $800 per structure. Work was carried out from February through October, with a three-month suspension in the height of the summer for peak power demand. Helicopter crews were initially slated only for areas with difficult access but proved so efficient that 80% of the work was done from the air, on a time and materials basis. This approach resulted in savings of about $1.6 million relative to planning estimates and much greater savings relative to actual bids.

Reliability Improvements

EDM compared the performance of the A1/B2 main line prior to retrofitting (1997 through 2003) to the line’s performance after retrofitting (2005 through 2012). T-tests were used to compare numbers of line faults per year. Prior to retrofitting, National Grid recorded 298 faults on the A1/B2 main line; after line modifications, National Grid recorded 65 faults. On an annual basis, all outages decreased from 42.6 to 8.1 per year (P=0.001), unknown outages decreased from 27.7 to 2.1 per year (P<0.0001), lightning outages decreased from 6.1 to 2.8 per year (not statistically significant) and other known outages decreased from 8.7 to 3.3 per year (P=0.01).

Average annual outages before and after retrofitting. Bars represent one standard error.

Perspective

During the project planning, National Grid used a variety of assumptions to target an overall outage reduction of 54% from the combined re-insulation and avian-retrofit project. Instead, the project reduced outages by 81%. Although substantial reductions occurred in lightning outages (55%) and outages from other known causes (63%), the greatest reduction was in unknown outages (92%). The strong circumstantial evidence for avian issues prior to the project, plus the extraordinary reduction in unknown outages in response to the retrofits, suggest avian interactions had been a chief cause of A1/B2 outages.

National Grid’s outage protocols are rigorous and evidence driven. The difficulty of identifying bird outages in the field is highlighted by the existence of just one OMS avian outage record from 1997-2003, a period in which many dozens likely occurred. Although electrocuted birds may hang in place, carcasses are frequently blown far from the structure by the shock, then obscured by vegetation or removed by scavengers. Detection challenges are exacerbated by thick vegetation, weather and nighttime conditions. The only physical evidence of a streamer outage may be a scorched conductor.

The improved performance on the A1/B2 line was likely because of a holistic combination of enhanced avian protection and re-insulation, which improved vertical separation. Multiple line modifications incrementally improved avian safety using strategies of perch management (perch discouragers), insulation (conductor covers) and increased separation (larger insulators). The removal of the pre-2003 triangular perch discouragers likely had a beneficial effect as they had exacerbated an existing electrocution hazard; this unintended effect serves as a reminder avian retrofits should be implemented carefully under expert guidance.

The overall financial benefits of the project are difficult to quantify because lost revenue varies widely depending on both the number of affected customers and time of day. No meter revenue is lost from momentary faults, but power quality and customer satisfaction suffer. Each momentary fault also increases wear and tear on equipment such as breakers and incurs approximately $5,000 in ground and aerial field crew deployment costs. Based on the number of averted outages, the benefits to National Grid were substantial. 

Acknowledgements

Julian Cox, who led National Grid’s reliability and performance measures team’s involvement in the project, also provided invaluable data and input in support of this article. Mike DiCecco supervised the helicopter-based component of the project implementation. EDM’s involvement in the project was led by Rick Harness, Dr. Greg Phillips and Joel Hurmence. Dr. James Dwyer and Lisa Nelson provided data analysis and review comments for this article, and Daryl Austin developed the project map.

Brian Reynolds ([email protected]) has 18 years of experience in the electric utility industry. He has performed a variety of transmission line projects in both engineering and project management roles, including a variety of avian mitigation projects. He presently is a standards engineer supporting overhead line design at both distribution and transmission voltages and underground distribution. Reynolds holds a BSCE degree from the University of Massachusetts and is a registered professional engineer in the Commonwealth of Massachusetts.

Richard Dupre ([email protected]) has more than 40 years of experience in the electric utility industry. During his successful tenure at National Grid, he performed all aspects of transmission line engineering, supported many key projects including the HVDC Phase II project, was a mentor to many engineers and achieved the utility’s highest engineering rank. Presently, Dupre is employed at Energy Initiatives Group, supporting National Grid on its key NEEWS project. Dupre holds a BSCE degree from Northeastern University and is a registered professional engineer in the Commonwealth of Massachusetts.

Duncan Eccleston ([email protected]) has been involved in numerous avian protection and wildlife outage management projects since joining EDM. Additional areas of expertise include environmental compliance and permitting, storm water and post-fire erosion. He is a graduate of Middlebury College and holds a master’s degree in watershed science from Colorado State University.

Mentioned in this article:

APLIC | www.aplic.org

EDM International | www.edmlink.com

Kaddas Enterprises | www.kaddas.com

Mission Environmental Products | www.missionenviro.co.za

Voice your opinion!

To join the conversation, and become an exclusive member of T&D World, create an account today!