THE CHALLENGE TREES POSE TO THE RELIABILITY OF OVERHEAD DISTRIBUTION SYSTEMS is well recognized, as vegetation is a dominant cause of service interruptions at many utilities. The electric utility industry spends billions of dollars every year performing line clearance tree pruning on both a preventive and corrective maintenance basis.

Some of the ways trees caused interruptions are well understood. The damage caused to overhead energy-delivery infrastructure by the structural failure of branches and whole trees is obvious, particularly during adverse weather events. As a result, work has been done to advance our knowledge of “hazard trees” and the failure of individual branches within a tree's crown.

Trees in direct contact with energized overhead conductors can cause interruptions by providing a pathway for the flow of fault current. When fault current is detected and interrupted by protection devices, such as fuses and reclosers, an outage occurs. This electrical mode of failure and the fault pathway provided by trees has been the focus of ongoing investigations. These research efforts represent high-voltage testing of more than 2000 specimens representing 21 species of trees.

Assumptions and beliefs regarding the interaction between trees and distribution lines, largely based on anecdotal observations, have guided decisions made by utility operations and engineering staff over the years. This research effort was initiated to better understand the ways in which a tree in contact with overhead conductors may cause an interruption. The goal is to identify risk assessment criteria that can be applied to vegetation management work on distribution systems.

GENERAL FINDING

All tree contact with energized conductors can result in a fault. The branch provides a pathway for the flow of current. The contact begins as a high-impedance (resistance), very low current event. The vast majority of tree contacts remain this way. Only under the “right” combination of conditions does the fault pathway become more conductive. In these cases, the fault pathway evolves from high to low resistance, resulting in high levels of fault current and ultimately an interruption.

Several important findings have emerged from this research. It is now clear that the potential for a tree in contact with an energized conductor(s) is influenced by key characteristics of the distribution line involved and by the fault pathway provided by the tree.

Three of the most important characteristics of the fault pathway and the potential of a tree-conductor contact resulting in an interruption are:

  1. Voltage gradients

    The electrical stress impressed on the tree or branch is a major consideration. Voltage gradient is a function of the voltage differential between two points and the distance between them. Tree contacts involving higher voltage gradients are much more likely to result in an interruption. Higher operating voltages and close phase spacing create higher gradients. Equally important to reliability, there appears to be a voltage gradient threshold below, which it is unlikely that a tree in contact with energized conductors will result in an interruption.

  2. Diameter

    The diameter of the fault pathway provided by the tree is an important consideration. Large-diameter pathways are much more conductive and therefore more likely to cause a fault than small-diameter contacts.

  3. Species

    There are readily observable differences in the conductivity of individual species. These differences are significant enough to warrant consideration in planning preventative and corrective maintenance tree pruning work on distribution circuits.

CHALLENGING 10 “TREE TRIMMING” MYTHS

The industry's approach to tree-related maintenance of overhead distribution lines has come a long way from the days of “tree trimming,” where trees were thought of as structures, and the goal was simply to establish and maintain fixed clearances and corrective response (or hot spotting) was the norm. Today, we speak in terms of vegetation management and design preventative maintenance programs with an understanding of system requirements and arboricultural practices. Knowledge gained from formal investigation is providing insight into the electrical interactions between trees and conductors. This new understanding of the characteristics of tree-initiated faults supports a direct challenge to some traditionally held beliefs.

  1. Trees cause outages by growing into contact with conductors

    Not so. New, soft small-diameter vegetative growth that grows into contact with an energized distribution conductor experiences resistance heating. This leads to drying, wilt and often dieback of living tissues. This is the cause of the discolored foliage commonly referred to as a burner. These fault pathways are actually less conductive than when the initial contact occurred

  2. Reliability is all about achieving line clearance

    Reliability isn't directly addressed by creating a fixed distance of clearance between trees and conductors. The risk is that the tree will provide a low-impedance fault pathway. The greatest risk is that a large-diameter pathway encounters a high-voltage gradient where it didn't previously exist. This can occur when a tree or branch fails or is deflected or when conductors sag or swing out of alignment. The point is that the greatest risk is when something changes. Focusing solely on fixed clearance distances from conductors misses an opportunity to improve reliability without increasing cost.

  3. The threat trees pose to reliability is the same for all parts of a circuit

    This clearly isn't the case when one considers the customers affected when a fault occurs on different locations on a circuit. The research demonstrates that voltage gradients vary by an order of magnitude on a typical circuit. They are highest on multiphase lines and lowest on single-phase line segments with open spacing between an energized conductor and neutral. Trees pose a much greater risk to the reliability of multiphase lines. Preventive vegetation maintenance plans need to reflect this understanding of risk

  4. The threat trees pose to reliability is the same for all types of overhead construction framing and phase spacing

    Voltage gradients vary widely between types of overhead line construction in service today. The voltage gradient encountered by the tree is the most important variable of all. Phase-to-phase spacing is typically the most important consideration because the voltage differential between phases is so much higher than that between an energized phase and neutral. Ironically, the compact designs that are often used to avoid tree clearance problems on restricted rights of way are among the least tolerant of tree-conductor contacts.

  5. All branches pose the same threat to reliability

    Not surprisingly, larger-diameter fault pathways provided by branches and trunks are more conductive than smaller ones. Conductivity increased dramatically with increasing diameter. Clearly, larger-diameter pathways present a much greater risk than do small branches and twigs.

  6. All trees pose the same risk to reliability regardless of species

    Substantial variability in electrical conductivity, and therefore risk, was observed among the 21 species tested to date. It is also well known that growth rates and structural integrity vary a great deal among species. It should come as no surprise that risk profiles differ among species. The good news is that, generally speaking, those species that are often targeted due to their growth rate and failure profiles are also among the most conductive.

  7. If you can't find the cause for the outage, it's probably due to a tree

    We now know that a low-impedance/high-current fault event irrevocably alters the fault pathway. The persistent fault pathway should provide observable evidence of the high-current event.

  8. Trees are a unique cause of momentary interruptions

    This is not likely to be the case. The research to date has not identified any unique characteristics of the fault pathway provided by a tree. We know that once the fault pathway is well established in the branch, it retains its low-impedance conductive characteristics. The fault pathway may be transient, for example, when a broken branch lying across two phases is ejected or burns through and falls clear of contact during the high-current event. An argument can also be made for intermittent contact with reintroduction of the fault pathway provided by the tree, though it would seem that the unique circumstances involved would be infrequently encountered. It is more likely that trees cause momentary interruptions in the same manner as other sources of faults, and it is the overcurrent protection system that manifests the event in terms of a momentary or sustained interruption.

  9. Tree-conductor contact always results in an outage

    If this were true, there would be far more interruptions on overhead distributions circuits than there are today. The majority of tree-conductor contacts remain as high-impedance, low-current events. Fault pathways provided by trees can be categorized into two discrete groups: very low conductivity or high conductivity, with few if any exhibiting stable intermediate levels of impedance. Only those fault pathways that evolve from high to low impedance are a risk to service reliability.

  10. Hot spotting “burners” is a cost-effective way to improve reliability

    Not true. As described in #1, the kinds of tree-conductor contacts that result in brown, wilted foliage are often not a serious threat to system reliability. In fact, these contacts generally become less conductive as the branch tissue and foliage dries and wilts.

UNDERSTANDING REDUCES RISK OF OUTAGE

Trees can have a significant impact on the reliability of overhead distribution systems. The risk they pose varies with a number of important attributes. Contemporary vegetation management programs based on an understanding of these factors can demonstrate increases in cost-efficiency and effectiveness. What this approach requires is a prescriptive application of maintenance resources, focusing on risk mitigations as well as productivity and workload reduction.


John Goodfellow is a vegetation management researcher with more than 25 years experience in the electric utility industry. He has held positions of responsibility for vegetation management, engineering and field services at three large investor-owned electric and gas utilities. He has also been responsible for managing T&D services for a major contracting organization. He has bachelor's degrees in forestry and natural resources management from Syracuse University and the SUNY College of Environmental Science and Forestry. jwgoodfellow@msn.com

ACKNOWLEDGEMENT

The project, coordinated by ECI, has been made possible through the support of the TREE Fund. As the principal investigator responsible for this research, the author would like to gratefully acknowledge the participation of Allegheny Power, Niagara Mohawk, Portland General Electric, Central Vermont Public Service, KeySpan Energy, Illinois Power, and Black Hills Power.

FAULT EVOLUTION MODEL

When a branch or stem of a tree comes into contact with electrical conductor(s), it provides a fault pathway between two areas of unequal electrical potential. Fault current begins to flow. The level of electrical stress and the initial characteristics of the fault pathway (branch or tree) have a major influence on what happens next.

High-stress gradients and relatively conductive pathways may result in increasing current flow as charring of the branch surface develops along the track of the fault. Areas of charring along the pathway are more conductive, leading to higher levels of current flow, which in turn makes the pathway even more conductive. And so it goes until at some point the gap between the points of unequal potential is bridged and a high-current fault occurs. However, most tree conductor contacts do not evolve into low-impedance/high-current faults. Often the electrical stress gradient and conductivity of fault pathway provided by the tree or branch are low. In these cases, the pathway experiences resistance heating. As low levels of current flow through the high-impedance branch, heat is generated, driving off internal moisture and reducing conductivity.

In actuality, the growth of a conductive carbon pathway and resistance heating and drying will occur concurrently. In effect, there is a “race.” At the same time that a carbon path may be forming, current flowing in the branch is producing resistance heating along its length. This heating has the effect of driving off moisture and thereby increasing electrical resistance. With relatively low-voltage gradients the current flow generating resistance heating is warming and drying the branch faster than any carbon pathway might form, thus no electrical flashover occurs. The branch simply warms and dries out, causing resistance to increase and current flow to tail off.

If, on the other hand, the voltage gradient is sufficient to generate localized dry band arcing, a conductive fault track develops. This charring results in the fault pathway becoming increasingly conductive. If the charred pathway reduces impedance faster than internal resistance, heating drives off moisture and increases impedance and the gap is bridged, resulting in a high-current fault and subsequent interruption.