I get the call around 8 p.m. on a Thursday. A farm operation has lost grid power. The solar system, which was supposed to keep the milking parlor and a freezer room running, is showing a red status light. The installer says: “The SMA inverter won't start. It must be bad.”
Actually, it's almost never the inverter.
I've been on the emergency side of the solar industry for 7 years. In that time, I've coordinated more than 300 urgent service calls for off-grid and hybrid systems. I've flown replacement units to remote sites, watched technicians line up multimeters at midnight, and seen the same pattern repeat: the equipment gets blamed, but the failure started long before the outage.
That's the surface problem. Let's go deeper.
The Problem You Think You Have: “The Inverter Died”
When something stops working, it's natural to look at the biggest, most expensive component. People search for “sma-inverter,” assume the inverter is dead, and ask for a replacement. Sometimes they do need a new one — I've authorized those replacements. But in my experience, maybe one in five emergency callouts is an actual inverter failure. The other four are design mismatches, ignored auxiliary equipment, and decisions made during the first install.
Most grid-tied inverters are required to shut down when the grid is lost. In the U.S., that's IEEE 1547. So if your “backup” inverter is utility-interactive, disconnecting isn't a failure — it's compliance. Reference: IEEE 1547.
The Problem Behind the Problem
1. You chose the right inverter for the wrong mode
A standard grid-tied string inverter has one job: send DC power from solar panels into the grid. It's efficient, reliable, and quite boring. But it is not a standalone power source.
If the system has to keep running after the grid disappears, you need an inverter that's designed for islanding. That means an SMA off grid inverter or a hybrid inverter that can create its own voltage reference. The SMA Sunny Island series does exactly that — it forms a microgrid, controls a generator, and manages batteries. A standard Sunny Boy won't do that on its own.
I see this mismatch all the time. Someone buys a large string inverter because it's cost-effective, then tries to use it as a backup power source. It doesn't work. The inverter shows an error, and the client says it failed. It didn't fail — it was never the right tool for the operating mode.
2. The auxiliary equipment is the real weak spot
Off-grid systems are full of small, unglamorous devices. A controller, a communication module, a shading motor, a gate opener. And too often, all of them rely on a 12V battery that isn't being maintained.
This is where a trickle battery charger matters more than people think. It sounds like a weekend workshop toy. But in a commercial off-grid system, a battery that sits at 12.1V instead of 12.6V can make a PLC or router restart in a loop. The “smart” inverter then sees no communication, throws a fault, and stops charging. When we eventually trace the fault, it's a dead 12V battery, not an expensive inverter.
I only believed this after ignoring it once. In 2023, a client's 40 kW off-grid system went into fail-safe mode three times in one month. We replaced communication cards, checked firmware, even swapped a main board. The real problem was a 12V instrumentation battery fed by a cheap trickle charger. A $60 charger had caused a $6,000 service bill.
3. Motor loads are a different animal
If you're powering pumps, compressors, conveyors, or any equipment with a three-phase motor, the start-up current is the problem. Motors can pull five to eight times their rated current during inrush. An off-grid inverter can trip on overload even if the motor's running current is well within limits.
What actually works is a VFD for 3 phase motor applications. A variable frequency drive soft-starts the motor, manages the inrush, and often has a DC bus that can connect directly to the battery bank. That changes how the system behaves. Instead of the inverter seeing a sudden demand spike, it sees a gentle ramp. But many installers don't include a VFD because it wasn't in the original single-line diagram.
One of my biggest regrets: not catching this before a dairy installation went live. We had a 35 kW SMA off grid system, a 15 kW feed pump, and a VFD that was “only needed if the pump struggled.” It struggled. The whole system shut down twice in the first week. We paid for another site visit and a rushed VFD shipment. We saved $800 on the initial quote; the correction cost $4,200. In the emergency service business, that ratio is depressingly common.
4. You're asking the wrong question about microinverters
The “microinverter vs string inverter” debate is one of the most popular conversations in solar. It's also one of the least useful for an off-grid project.
Microinverters are excellent for rooftop installs with shading and monitoring complexity. They do not magically make a system more flexible in an emergency. If you need backup power with a battery, a microinverter setup is harder to AC-couple to an off-grid inverter, requires more communication, and gives you more points of failure during an outage. String inverters — especially those designed for hybrid operation — are easier to integrate into a battery-backed island grid.
I'm not saying microinverters are bad. I'm saying the question is wrong. Don't ask “microinverter vs string inverter” as if one answer applies to every roof. Ask: “What happens to this system when the grid goes down?” That answer is what determines the architecture.
Before we sign off on a system, my team checks the inverter against IEC 62109 and IEC 62116. Those standards don't tell you whether to choose a microinverter or a string inverter, but they tell you what has to happen when the grid fails. That's more useful than a marketing sheet. Reference: IEC 62109-1, IEC 62116.
What Ignoring This Actually Costs
I can give you the textbook answer: downtime, spoiled product, missed deadlines, penalty clauses. But the real cost is less visible.
It's the client who calls your competitor at 2 a.m. because you didn't answer.
It's the project manager explaining to a utility that the “guaranteed uptime” system was down for 14 hours. It's the warranty claim processor who has to tell a customer their “failed inverter” was actually a design flaw — and the awkward silence on the other end of the phone.
In March 2024, I coordinated a same-day replacement for an SMA off grid inverter that had actually failed. We shipped a unit from a regional warehouse, paid $400 in rush freight, and had the system back online by evening. The client's alternative was a $12,000 penalty for failing to provide power to a critical communications site. That ended well. But the much harder calls are the ones where the inverter is fine, the auxiliary system is the culprit, and there's no single product to swap.
The cost, in those cases, is not just money. It's trust.
A Short Fix (Kept Brief on Purpose)
I've spent most of this article on the problem, because the problem is rarely where people look first. But if you want the short version:
- If the system needs to run during an outage, choose an SMA off grid inverter or a hybrid inverter with true islanding — not just a standard grid-tied string inverter.
- Don't forget the 12V side. Add a reliable trickle battery charger to every control panel and critical load battery.
- For any three-phase motor load, plan for a VFD for 3 phase motor applications from the beginning. It's cheaper than a service call.
- Ignore the microinverter vs string inverter hype unless you're optimizing rooftop yield. For emergency backup, a string inverter with battery is the simpler and more resilient choice for most backup designs.
- If you want room to grow, look at an SMA flex inverter architecture that lets you expand storage or hybrid capability without replacing the core unit.
None of this is exotic. It's just design discipline. The SMA inverter itself is rarely the whole story, and the sooner you understand what's around it, the less likely you are to need someone like me at 8 p.m.
At least, that's been my experience. I'd rather you never find out what your system does in an emergency — and if you do, I'd rather the only emergency be the one you planned for.
