If Thomas Edison's first electric system under the cobblestones of lower Manhattan, were still the industry standard, our industry would be a curiosity shop. These direct-current mains consisted of two 20-ft (6-m) sections of half-moon-shaped solid copper bars, separated by insulation, inserted into iron tubes filled with hot liquid asphalt. Prior to installation in 1881/1882, Edison himself considered making even more changes to these problematic mains, which had been tested and modified many times in Menlo Park, New Jersey, U.S. Obviously, Edison's first-generation distribution systems are curiosities, properly housed in museums. This is because of many, many technological advances and changes. But change is never easy; it is never routine. To change is to shake hands with risk, your constant and unforgiving companion in any new venture.

American Electric Power (AEP; Columbus, Ohio, U.S.) recently confronted a situation involving an upgrade to its transmission system in Texas that cried out for a new approach that pushed the envelope. The potential rewards were great, and AEP carefully balanced them against the perceived risks.

AEP's 15.7-mile (25-km) 138-kV transmission line from its Pleasanton Substation to CPS Energy's portion of the line south of San Antonio, Texas, U.S., had become a power-delivery constraint. CPS Energy, the municipal electric system serving San Antonio, had upgraded its line from the Bexar County line into Leon Creek Substation with new structures and Drake 795 kcmil aluminum-conductor-steel-reinforced (ACSR) conductor. AEP's portion of the line was a 50-year-old wood H-frame line with Partridge 266.8 kcmil ACSR conductor.

The conventional choice for this project would be to replace the conductor with a much larger one that would require replacing more than 75% of the 139 existing structures. The tantalizing alternative would be to continue using the existing structures and replace the conductor with technology wire that promised to have limited conductor sag at elevated operating temperatures.

If AEP went with the conventional choice, structures would have to be replaced because of the increased mechanical load and sag of the much larger conductor. However, the new technology conductor would not increase mechanical loads, and would meet ground-clearance requirements with existing attachment elevations. Using existing structures would shorten construction and outage time. Furthermore, optimizing an existing right-of-way is a key element of AEP's transmission strategy. Employing new technology also supports the company's vision of maintaining leadership in technical innovation.

The reported attributes of the new aluminum-conductor-composite-core/trapezoidal-wire (ACCC/TW) conductor are very positive and allow continued use of existing structures while providing the required incremental circuit capacity. ACCC/TW has a lightweight, high-strength composite core wrapped with fully annealed trapezoidal-shaped aluminum strands. ACCC core is a 30-to-70 resin-to-fiber ratio composite produced in a pultrusion die. The structural core consists of carbon fibers wrapped with a “shell” of continuous glass fibers, a hybrid polymer matrix. This ACCC core has more than 1.3 times the tensile strength of the same size ACSR core.

The composite core's coefficient of thermal expansion (CTE) is four times lower than the steel core of ACSR. The result is an expected sixfold reduction in high-temperature sag. The composite core's low weight allows more conductor for the same weight. Plus, soft-tempered aluminum improves conductivity. Trapezoidal-shaped wires eliminate the interstitial spaces, characteristic of round strands, and increase the conductor aluminum cross-sectional area.

The bottom line for all these attractive mechanical and electrical attributes — in addition to no tower replacements — was US$600,000 in savings. This potential cost savings led AEP to undertake an extensive review of ACCC/TW.

The review process became one of extensive due diligence. AEP had to get comfortable with the new technology and manage the risks of applying the technology to its transmission system. Due diligence confirmed the advantages of ACCC/TW manufactured by General Cable Corp. (Highland Heights, Kentucky, U.S.) and Composite Technology Corp. (CTC; Irvine, California, U.S.). Another plus was a three-year full material and labor replacement warranty and an optional additional seven-year warranty covering material and labor to remove and replace wire. Longevity and durability testing to determine fatigue, brittleness, aging and environmental resistance helped bolster confidence.

EPRI's testing of ACCC/TW conductor hardware at high temperatures gave AEP even more reassurance. Ice-loading capability was a concern and could limit ACCC/TW's thermal advantages. Because conductor icing is limited in south Texas, AEP's Leon Creek project was able to take full advantage of ACCC/TW's reduced sag at elevated temperatures.

To quote Sherlock Holmes's esteemed associate Dr. Watson in The Hound of the Baskervilles: “Has anything escaped me? I trust there is nothing of consequence which I have overlooked.”

AEP decided to go with the new conductor and reap the savings. This would be the longest ACCC/TW transmission line in the United States — in fact, the world.

During the conductor pulling operation, installation crews experienced three breaks of the ACCC/TW conductor. The first break occurred at the outset of wire pulling when the conductor jumped off the tensioner bull wheel and wrapped around the tensioner axle. This event was plainly an operator error. During this specific operation, the operator should have applied the reel brake. Instead, the operator lessened brake tension, and the loosened wire left its track on the drum and broke when it was bent too severely around the tensioner axle. In this case, wire already pulled for this particular phase was simply eliminated, and pulling resumed with fresh wire.

The second and third wire breaks were more troubling and required in-depth forensics. Break two occurred as wire was pulled into a span and being tensioned for sagging. Break three, involving several breaks in the core, occurred less than two weeks later in the middle phase, after all three phases had been pulled in, tied off and the crew had left for the day.

CTC submitted samples of broken wire to the University of Southern California's (USC's) Composites Center for testing. USC conducted thermal and mechanical tests on the damaged ACCC/TW wire. The post-mortem on the breaks included:

• Dynamic mechanical analysis (DMA) to determine the glass transition temperature (Tg) of the failed cores versus a control group.

• Differential scanning calorimetry (DSC) on the two damaged resin compositions to determine differences in their reactivity as possible causes.

• Three-point short beam shear tests on the control group and damaged composite parts to measure shear strength.

From these tests, the USC Composites Center determined that there were no significant differences in the thermal and mechanical (tensile and flexural) properties of the control group and the failed composite samples.

Based on these tests, CTC attributed the failures to excessive bending of the core during installation. “What events transpired that caused the excessive bending may never be known,” CTC concluded in its own review.