Feeders We Have Fixed

The member wanted two megawatts of solar on a rural feeder about nineteen miles long. The initial screen failed on reverse power through the line regulator and the co-op denied it. That is a normal outcome and nobody did anything wrong, the screen exists to be conservative. We modelled the feeder properly, with a full year of hourly load from the substation meter rather than a single peak snapshot. Reverse flow through the regulator did occur, for about two hundred and forty hours a year, all of it in spring at light load. The regulator was on a control that could not handle it. A bidirectional control upgrade, moving one fixed capacitor bank to a switched control, and a slightly tightened voltage bandwidth cleared it. Total cost of the fixes was under forty thousand dollars against a project worth several million to the county. The study took four weeks. The lesson I take from it is that the screen is a filter, not an answer, and somebody has to be willing to do the work behind it.

A municipal utility in Wyoming was getting three or four solar enquiries a month and had one engineer, who also ran the meter shop. Every enquiry became a study and the queue was over a year long. We modelled all twenty two feeders and produced a colour map showing generation headroom by section, updated quarterly against actual load. Anything under the headroom number gets approved on a screen. Anything over it gets a study, and the map tells the developer before they apply which parts of town are worth looking at. The queue is now measured in weeks. The engineer got his evenings back. It cost less than the four individual studies the utility would otherwise have paid for that year.

The battery was installed and energised before anyone asked what it did to the protection. That is more common than it should be, because storage gets treated as a load question and it is a fault current question. We pulled the existing settings, modelled fault current with the battery in all four of its operating states, and found three lateral fuses that would now blow before the recloser had a chance to clear a temporary fault. On a rural line with trees, that turns a two second blink into a truck roll and four hours of outage for a dozen members. We reissued the settings and the fuse schedule. Two weeks of work, delivered as a sheet the technicians could act on, with the study behind it if anyone wanted to check. Nothing about this was hard. It just had to be somebody's job.

Old 4 kilovolt systems are a slow emergency. Losses are high, the conductor is small, and any new load or generation is a fight. The full conversion cost more than this utility's entire annual capital budget, which is why it had been deferred for fifteen years. We broke it into nine phases ordered by where the pain actually was, using loading, outage history and the age of the transformers rather than a map drawn by geography. Each phase stands alone and leaves the system in a working state, so the plan survives a bad year or a change of board. They are on phase four. Losses on the converted sections are down by roughly a third and they have stopped saying no to new three phase service in the north end of town.