Three independent consultants have reviewed the circumstances surrounding the failure of the submarine cables. The consultants were: Novak Engineering, Inc. (Edison Sault’s primary engineering consultant), Kerite Corporation (manufacturer of the most recently installed submarine cables), and DTE Energy Technologies. Copies of their reports can be found in Appendices 5.1, 5.2 and 5.3 of this report.

Each of the consultants identified the same pattern of damage and concluded that the likely cause was thermal breakdown of the cable insulation in those portions of the submarine cables installed on land. Thermal breakdown of the insulation was caused by magnetic heating of the steel armor wires induced by current flow through the cables and due to the proximity of the cables to each other.

The consultants also identified a single contributing factor that likely accelerated the failure: excessively dry soil surrounding the portions of the cables installed on land. This condition was particularly prevalent near the lake shore due to the low lake level experienced for the last few years.

The following discussion of the failure mechanism summarizes key findings of the consultants.

Three of the submarine cables were manufactured by Okonite Corporation and were installed in 1983. Four of the cables (a three phase circuit and one spare) were manufactured by Kerite Corp. and were installed in 1988. All of the cables were specifically designed, insulated and installed for submarine service.

Starting at St. Ignace Substation, the submarine cables were installed underground for approximately 200 feet where they transition to being laid on the lake bed for approximately 15,000 feet to Mackinac Island. Approximately 300 feet from the shoreline of Mackinac Island, the cables transition to being buried in trench in order to avoid damage from watercraft and ice. From the shoreline, the cables run underground for approximately 1,500 feet to Pat Chambers Switching Substation.

Where the submarine cables are buried on land, they are laid in a three-foot-wide trench and backfilled with sand and other soil. In the water, the cables were laid on the lake bed with ten-foot separation between each cable.

In service, cable temperature is a function of the heat generated within the cable and the ability of its surroundings to dissipate that heat. In these cables, heat is generated both by the flow of current in the conductor and by the magnetic field generated by this flow interacting with the armor shield of the cable and other nearby metallic objects. The heating effect tends to increase as the load on the cables increases.

Where the cables are laid on the lake bed, the heat is dissipated through convection arising from the difference in temperature of the cable and the surrounding water. The greater separation between cables laid in the water further minimizes heat generated by the magnetic field associated with each cable.

The heat damage exhibited by the submarine cables installed between St. Ignace Substation and Pat Chambers Substation appears to be restricted to those portions installed on land where the cables were laid side by side, in a three-foot-wide trench. The proximity of the cables when laid in a trench contributes to the magnetic flux heating between the conductors and the armor shield of adjacent cables.

Normally, as demonstrated by the historic performance of these cables, moisture in the ground is adequate to dissipate the heat generated in this configuration. However, two predominate factors may have combined to the detriment of this installation.

• The region has experienced a prolonged dry spell, resulting in lower soil moisture content and lowering of the water level in the Great Lakes by more than three feet, subsequently exposing greater portions of the buried cable to the effects of heating. As the soil moisture level is reduced, the ability of the soil to dissipate heat generated by the cables is also reduced.

• In addition, heavier loading of the cables to higher electricity demands on the Island had the effect of generating more heat which subsequently acts to drive moisture away from the cables, exacerbating the problem. However, a review of company records by Novak Engineering shows that the loading on these cables had not exceeded previously revised lowered ampacity ratings that take into account the proximity of the cables installed on land. This would tend to point to reduced soil moisture as the primary cause of the heat related failure of these cables.

 

Table of Contents  |  Continue