2025 Update: Ferroresonance Explained
Ferroresonance is a complicated issue, one that industry workers must be educated about. That’s because as the number of URD system installations grows and systems age, instances of ferroresonance increase – as do threats to worker and customer safety, equipment and service reliability.
I first became acquainted with ferroresonance in the 1980s while troubleshooting a pad-mounted, three-phase transformer at night. The pad fed a chemical plant, closed for the evening, in the middle of nowhere. The 480-volt, 2000-amp main was single-phasing, so plant electricians had dropped all the 480-volt sub-feeds except one: a single-phase, 240-volt sub-panel that fed the plant’s fire control systems plus some lighting that was barely functional (keep this information in mind – it will be important a little later).
The underground radial feed was long, and the A-phase pothead fuse had blown; my first thought was a bad cable. We pulled the radial-fed elbows at the transformer, opened the 2000-amp main and re-fused the pothead. It held – so why had the other fuse blown? Next, we checked the meter’s kilowatt demand, which was barely more than half the peak-load capacity of the 1500-kVA transformer. Then we smoke-tested the transformer, closing it in from the potheads, opening the potheads with a load-break tool, and finally closing in the elbows on the de-energized pad.
The 23-kV potheads were dead-ended on a 10-foot heavy-duty double arm. The fuse for A-phase was on the far side of the arm, on the other side of the neutral from where the bucket truck had been set up. I wasn’t going to side-sling the fuse barrel feeding a 1500-kVA pad, so I boomed over the neutral positioning for the A-phase fuse and closed it. As I was booming back over the neutral to close in B-phase and C-phase, I heard what sounded like a car crash coming from the vicinity of the pad, followed by a flash and something going to ground. The A-phase pothead fuse erupted behind me and the feeder relayed. Back at the pad, A-phase and B-phase elbows were blown off the bushings. The A-phase polymer arrester that was plugged into the feed-through bushings had split down the middle and was still smoking. Additionally, the current transformer cabinet wiring and polyphase Class 10 meter were on fire.
What Happened?
Given that the CEO of the chemical plant was on the utility company’s advisory board, some plant personnel were alerted when we blew up their transformer. Three crew members stayed at the plant to pull the bad transformer and ready a spare three-phase. An apprentice and I went to retrieve a new transformer. On our way to the yard, an engineer on-site at the plant radioed us, requesting that we bring back a 2000-kVA transformer. I asked if he thought the damaged transformer had been overloaded; he said no and told us he would explain more later.
His eventual explanation? You guessed it: ferroresonance.
As it turned out, during our troubleshooting, we had created perfect conditions for the loud noise and fire. No one had known that the capacitive reactance of the cable on A-phase and B-phase was nearly equal to – and in series with – the inductive reactance of the 1500-kVA transformer windings. In the evenings during the off-season, the plant reduced operation and electrical loads. With the matching series reactance of the cable and transformer impedance, the lightly loaded transformer would begin to react, creating low-level ferroresonance that overheated and prematurely aged the transformer until the pothead fuse blew. When we began troubleshooting and opened the main, conditions became ideal for runaway core excitation or ferroresonance. The fact that it had taken me a couple minutes to get from the A-phase fuse to the B- and C-phase fuses – combined with the open main and no secondary load – triggered everything needed to blow up that $60,000 installation.
You may still be wondering why the engineer requested that 2000-kVA transformer. The answer: in observing damage from the incident and speaking with the crew, he recognized the problem and opted for the new transformer to raise impedance. That way, the cable capacitance and transformer inductance would no longer be almost equal.
Need-to-Know Info
Ferroresonance is a rare condition most likely to occur with three-phase, pad-mounted, delta-connected transformers. Not nearly as often, ferroresonance has been documented in wye-wye transformers as well as in aerial three-pot banks served by long-dedicated aerial circuits.
When ferroresonance occurs in a transformer, high voltages three to five times the rated primary can appear on the primary and secondary and in the core. Oil heats to temperature extremes in minutes, blowing out of vents and bubbling paint on top of the transformer. Surge arresters – not designed to clamp sustained overvoltages – can be cooked to destruction and potentially fragment during failure. The rise in primary also increases the secondary voltage, sometimes blowing up meters like bombs. Other times, purely coincidental yet ideal conditions create low-level ferroresonance that can boil the life out of a transformer with barely a whimper. I know of one case in which a 1000-kVA transformer was replaced three times in five years. Finally, the utility realized a low-level resonant circuit – yes, without an open phase – was killing the transformer every night when the commercial building load dropped to about 6% of the transformer’s rating.
URD cables are capacitors; transformer coils are magnetic inductors. To create resonant circuits, there must be capacitive reactance and inductive reactance of almost equal value in series with each other, and the inductor must have very little to no load. The most likely situation is a three-phase transformer fed by a long underground circuit. In some of the most dramatic events, a pothead fuse was opened or blew, allowing a still-energized primary cable (capacitance) to be more or less in series with a coil (inductance). If a series-connected phase-to-coil connection is allowed to remain energized, and there is low loading on the transformer’s secondary or the customer’s mains are open, no impedance exists in the primary circuit. Current is free to flow, and runaway voltage rises in the laminated core – hence the “ferro-” component of ferroresonance.
In testing, loading the secondary above 20% has proven sufficient to prevent resonance. The first indication of the condition is typically a loud rattling noise – often described as shaking a coffee can full of marbles – emanating from a transformer due to magnetostriction in the laminated core (note that normally, magnetostriction causes the 60-Hz hum in transformers). The noise is wicked enough that almost no one who hears it stands around to see what is going to happen next.
The two other ferroresonance cases I have worked on both involved amorphous-core, three-phase, pad-mounted, wye-delta transformers. One was 1,700 feet of 1/0 to a 1000-kVA; the other was 2,000 feet of 1/0 primary to a 1500-kVA pad-mount. These are not formulas for determining the potential for ferroresonance but examples of the conditions present when it has occurred.
Preventive Efforts
So, what’s the best way to avoid suspected resonant circuits? Never open three-phase transformers one phase at a time from potheads or lateral taps. Some utilities are using an air-break or AB switch to isolate the coil before switching potheads.
Other methods include shortening the primary run to change capacitance or replacing transformers to ensure different inductive impedance values. Rural electric association specs add a fourth dropout to temporarily ground the high-side floating neutral when energizing or de-energizing three-pot banks. The goal is to split the series path between the primary feed and the transformer coils, removing the series reactance, which is a prime condition necessary to create the resonant circuit.
The last preventive effort is leaving some or all of the secondary load connected. This, of course, contradicts what we have always been told (i.e., “Don’t single-phase the customer”), but not to worry. You won’t kill the three-phase customer equipment in the time it takes to close three pothead switches – and you’ll almost certainly avoid blowing up a costly transformer.
About the Author: After 25 years as a transmission-distribution lineman and foreman, Jim Vaughn, CUSP, has devoted the last 28 years to safety and training. A noted author, trainer and lecturer, he is a senior consultant for the Institute for Safety in Powerline Construction. He can be reached at jim@ispconline.com.
Editor’s Note: This is an update to Jim Vaughn’s article “Ferroresonance Explained,” first published by Incident Prevention magazine in 2012.