Recently, a cooperative from the Midwest experienced numerous outages on its underground circuits due to cable failures. The affected circuits are approximately 14 and 16 miles (23 and 26 km) long, and were installed around 1986. The cables are unjacketed, with 4/0 aluminum conductors, extruded semiconducting screens, 175 mil wall insulation, with bare copper concentric neutral wires. The initial major fault occurred in 1995. The next three faults were within an area of 40 square yards (33 square meters) and during summer seasons. Subsequent faults spread throughout these circuits.

The cooperative dug up affected cable sections and sent cable samples to the manufacturer and to independent labs for testing. One of the reports suggested the problem is associated with metal corrosion that may be caused by the presence of sulfate reducing bacteria. The most common physically visible problem identified in all of the test reports seems to be the presence of "spots" on insulation surfaces. None of the reports stated how these spots developed but several possible mechanisms of development were suggested. One such suggested mechanism was that moisture containing dissolved calcium carbonate is migrating into the cable. When the moisture attempts to migrate back out (driven by temperature gradients or capillary action), the calcium carbonate precipitates in localized regions under the insulation shield. This buildup of precipitate could cause the insulation shield to lose contact with the insulation in these regions, leading to partial discharge and eventual failure.

The soil environment in all effected areas has a higher concentration of calcium carbonate than a "normal" environment. In one area, the soil is 12-1/2 times more conductive than seawater. Of the five failures evaluated to date, four were located in a site some 3-1/2 feet (1.1 m) lower in elevation than the surrounding terrain while the fifth was near a small lake. It is probable that these locations are water drainage sites for the respective areas. The manufacturer believes that agriculture chemicals may be concentrating in these failure locations, and these chemicals, in turn, are being absorbed into this unjacketed cable, causing the observed "spots" in the insulation/screen interface and seriously affecting the integrity of the cable dielectric.

The manufacturer also theorizes that the failure could be from the limestone deposits that leached into the cable from the surrounding soil. The company further speculates that something happened to the cables after installation resulting in the loose insulation/insulation screen adhesion area.

Electrical and physical examinations to date show evidence of impulse damage to the cables and chemical attack from something in the service environment. Cable with thicker insulation and a jacket has been proposed as a solution to this highly localized electrical phenomenon. Chemical infusion and biological attack should be minimized by using a jacketed cable.

If you would like more information or have any questions, please call Trung Hiu, Electrical Engineer, Distribution Branch, at (202) 720-1877.