One Solar Plant Outage, Three Different Root Causes.

A 150 MW solar plant loses a 35 kV feeder. SCADA lights up red, the main transformer‘s differential relay operates, and a 110 kV motorized breaker clears the fault in under two cycles. To the plant operator, this is an outage. To whoever has to explain it afterward — an insurer, a subrogation team, an asset owner facing a warranty dispute — it’s a question with more than one plausible answer, and picking the wrong one is expensive.

Utility-scale solar compresses an entire grid’s worth of electrical engineering into a few hundred acres: inverter-based generation with its own fault behaviour, layered protection and control logic, step-up transformers, and heavy MV/HV switching apparatus. A single trip can originate in any one of these systems, and the evidence it leaves behind can look, at first glance, like it points somewhere else entirely. Investigating a failure like this properly means being fluent across all of it — not handing the file to whichever specialist happens to be available.

Four things make that harder than it sounds:

• The generation side breaks the usual fault assumptions. Inverter-based resources shape and limit their own fault contribution, so reading fault current the way you would for synchronous generation gives a distorted picture of what actually happened.
• Protection is stacked in layers that don’t explain themselves. Knowing which relay operated first tells you what the protection scheme saw — not necessarily what happened at the equipment.
• The physical failure modes mimic each other. A breaker mechanism problem can present, downstream, as a transformer fault. Transformer symptoms can point convincingly at the wrong cause entirely.
• Reviewed alone, each domain produces a defensible answer. That’s the real trap: a power-systems read, a protection read, and an apparatus read of the same trip can each sound complete, and each be wrong, because each is only seeing a slice of the event.

What follows works through each of these three domains in turn, and where the blind spots sit when they’re reviewed in isolation.

Inverter-Based Generation Changes What a Fault Looks Like

In conventional thermal or hydro generation, fault current is large, predictable, and governed by synchronous machine physics. A PV plant behaves differently. Inverter-based resources limit and shape their fault contribution according to ride-through control logic, responding to a grid disturbance in a way that has more in common with a control system than a spinning generator. A power-systems review of the disturbance — fault contribution, network response, whether upstream protection saw a genuine fault or an artifact introduced by inverter behavior — is a necessary part of any investigation at these sites. It’s also, on its own, incomplete. It explains how the network responded. It doesn’t explain why a specific piece of equipment failed.

Protection Logic Tells You What Tripped, Not Why

A modern PV site is protected in layers — inverter and string-level protection, transformer differential and restricted earth fault schemes, feeder overcurrent, bus protection, main transformer protection — and when something trips, the sequence-of-events record shows which relay operated first. It’s tempting to stop there. Relay operation reflects the electrical signature the protection scheme was built to detect, which isn’t always the same thing as the physical event that produced it. Instrument transformer behaviour and relay coordination settings can all shift what the record shows without changing what actually happened at the equipment. Reading a fault recorder correctly means understanding where a protection scheme can be misled — not just where it responded.

Transformers Fail in Ways That Mimic Other Things

Step-up transformers sit at the center of most PV failure investigations, and they’re also the easiest equipment to misdiagnose. Winding configuration changes how a unit responds to a given fault type. On-load tap changer problems can produce internal gas and pressure signatures that read, from the outside, like a winding fault. Re-energization inrush, under the wrong conditions, can resemble a low-impedance fault to both a protection relay and a first-pass reviewer. None of this is exotic — it’s the normal range of possibilities a transformer specialist has to rule in or out before a unit gets called failed. The risk is a report that arrives at “transformer failure” as its finding when the transformer was the last piece of equipment in the sequence, not the first.

The Breaker Is a Machine, Not a Switch

MV and HV circuit breakers at a PV site — motorized spring or motor-charged mechanisms, vacuum or SF6 interruption — are electromechanical systems with their own duty cycle, wear pattern, and failure modes, entirely separate from anything happening electrically. A mechanism that doesn’t fully reset between operations, a contact system worn past its expected service life, or an interrupting medium that’s degraded below rated performance can each turn the breaker itself into the source of a fault, rather than the device that clears one. When that happens, the resulting damage often shows up somewhere else — most often on the equipment the breaker was supposed to protect — and gets attributed there instead.

Why the Same Event Reads Differently to Different Specialists

Picture a main HV breaker tripping on overcurrent. Reviewed on fault current magnitude alone, it looks like an external grid event. Reviewed on relay timing alone, it looks like a coordination gap with the utility. Reviewed because gas showed up in the Buchholz relay, it looks like a winding problem. Each of these conclusions is defensible from inside its own discipline, and each can be wrong for the same reason: it’s an accurate read of a slice of the event, not the whole of it. In our experience, this pattern shows up often enough that a single-discipline finding on multi-discipline equipment gets treated as a flag, not a conclusion, until it’s been checked against the other two views.

What This Means for Claims, Subrogation, and Asset Owners

Root cause drives everything downstream of a loss — repair scope, business interruption timelines, and whether there’s a viable subrogation target in an OEM, an installer, or a maintenance contractor. Assigning root cause to the wrong system means the wrong party pays, the wrong equipment gets replaced, and the conclusion doesn’t hold up to closer scrutiny. Getting it right the first time takes an investigator who isn’t confined to one corner of the electrical system, and who knows exactly where each discipline’s blind spots are.

The Case for One Investigator Who Covers All of It

Splitting a PV failure investigation across separate power-systems, protection, and apparatus specialists isn’t just slower — it produces three defensible-sounding partial answers instead of one complete one. Mission Critical Forensics investigates electrical equipment failures — transformers, MV/HV switchgear, protection and control systems, and the complex, multi-cause losses that don’t fit neatly into one discipline — for insurers, subrogation counsel, and asset owners.