"Generator or inverter?" is a question I get asked at least once a week, and it's one of those questions people expect a single answer to. There isn't one. After four years of reviewing spec sheets and field failures, I've learned that the right answer usually depends on which of three situations you're actually in:
- You're grid-tied, already have or are adding solar, and want short-term backup during outages.
- You're fully off-grid — remote site, farm, cabin — and self-reliance is the goal.
- You need critical backup (medical equipment, cold chain, telecom) where power quality matters more than upfront cost.
Quick context on where this comes from: I'm a quality and brand compliance manager at a solar equipment distributor. I review every spec sheet and warranty document before it reaches our installers — roughly 200+ documents a year — and in 2023 I rejected about 12% of first submissions at the labelling stage, mostly for missing or incorrect grid-code marks. When I first started in this role back in 2020, I assumed the inverter datasheet was the whole story. Two field failures later (both blamed on the inverter, neither caused by it), I realized a datasheet only tells you what the equipment can do, not what the installation context allows.
So here's how I'd break it down, scenario by scenario.
Scenario 1: Grid-Tied Solar With Backup Needs
If you have a grid connection and solar panels, an SMA Sunny Boy solar inverter (or another qualified string inverter) is usually the default choice, and for solid reasons. A grid-tied inverter converts the array's DC output into grid-compliant AC, which means it has to meet the interconnection rules where you live — IEEE 1547-2018 in the US (updated with smart-inverter functions), the G98/G99 or VDE-AR-N 4105 equivalents in the UK and Germany.
The flip side of grid compliance is anti-islanding: when the grid goes down, a standard grid-tied inverter shuts off, by design. That's a safety requirement, not a defect. If you want the lights to stay on during an outage, you're not choosing an inverter — you're specifying a system, and that system needs a battery or a generator to keep working while the grid is dead.
On the hardware side, the first things I check are range and certifications. SMA's string range covers roughly 5 kW to 150 kW+, from residential through small utility, which matters when you're standardizing across a mixed project pipeline. Published efficiencies for the Sunny Boy family typically land around 97–98% (European weighted), and the units carry IEC 62109-1/-2 for safety. For a sense of installed base — which affects spare-parts availability and long-term serviceability — SMA's 2023 annual report cited roughly 20 GW of inverter output shipped. That number comes up whenever a client asks, quite reasonably, "will I still be able to get a replacement in 10 years?"
"The vendors I trust most are the ones who tell me where their product stops and something else should start. I try to hold myself to the same rule on this side of the table."
Here's where I have to be candid about boundaries. SMA builds inverters. It does not build generators. If your actual requirement is "run the well pump for three days during a hurricane," the inverter is one component, and the part doing the heavy lifting may be a battery bank or a generator — neither of which is an SMA product. I'd rather say that plainly than frame a grid-tied inverter as something it isn't. A specialist who knows their limits is easier to trust than a generalist who claims to cover everything.
Scenario 2: Fully Off-Grid Sites
Off-grid is where most generic "inverter vs. generator" advice falls apart, because the cost math flips depending on how remote you are and how much power you actually need.
For a remote cabin running a fridge, lighting, and a well pump, a generator is almost always the cheapest way to get started. The Champion inverter generator 3100 (roughly 3,400 starting watts / 3,100 running watts — ratings vary by model year, so check the current spec sheet) is a decent example of the modern portable category: inverter-based output, fairly clean power, and a total harmonic distortion low enough that most sensitive electronics tolerate it. The "inverter" in the name is the same basic idea as what SMA does at utility scale — convert and condition — just optimized for a fuel tank instead of a grid.
Here's the counterintuitive part. For a cabin used more than a few weeks a year, the generator alone often isn't the cheapest option over five years. Fuel, oil changes, and — the piece people underestimate — fuel-system maintenance eat into the savings. A Chrysler 300 fuel pump is a known example of how modern fuel systems dislike short trips and ethanol-blended gasoline sitting in the tank; my neighbour replaced his twice in six years, both times after long stretches of infrequent use. Generators are worse about this, not better: they sit idle for months on stale fuel, and the carburetor and fuel pump are usually the first things to suffer.
That's the case where a small battery bank (say 10–15 kWh) plus a modest solar array and an off-grid inverter more or less wins on total cost of ownership versus "run the generator for everything." But it's genuinely load- and latitude-dependent, and I've seen sites where the generator-plus-battery hybrid comes out ahead. Off-grid isn't a product category — it's a sizing exercise.
Scenario 3: Critical Loads, Where Power Quality Decides
People ask me what is the difference between a generator and an inverter as if the two were alternatives. They aren't, really. A generator turns fuel into electricity. An inverter converts and conditions electricity. Most serious backup systems use both, and the actual question is which piece goes where.
Medical equipment, telecom sites, food cold chain — in these cases the decision usually stops being "generator or inverter" and becomes "generator plus inverter plus battery." What you're buying is transfer time and power quality, not peak wattage. The few seconds between the grid dropping and the backup carrying the load, the inrush current behind a compressor start — those details decide whether the system works on a Tuesday afternoon in July, not the headline kilowatt number.
This is where hybrid inverter platforms earn their keep, because they can coordinate PV, battery, grid, and generator inputs. On the compliance side, I review equipment safety against IEC 62109, and grid-interactive functions against IEEE 1547-2018 and UL 1741 SB in North America. On sizing, the rule I apply is to count inductive loads at roughly 1.25× running watts minimum, more if multiple motors can start simultaneously.
And this is where gut instinct has earned its place next to the spreadsheet. On a cold-storage project, the sizing spreadsheet said the generator was sufficient — the continuous-load math checked out. Something felt off about the start-up sequence. I asked for a live start-up test anyway. All three compressors starting together tripped the system in under two seconds. The spreadsheet was right about kilowatts and wrong about inrush current. We added soft starters; the redo cost about $18,000 and pushed commissioning back three weeks. Now every critical-load spec we issue requires a documented start-up test.
How to Tell Which Scenario You're In
Four questions usually settle it.
- How often does the grid fail, and for how long? A couple of flickers a year is a power-quality conversation. Multi-day hurricane-season outages are a resilience conversation. They lead to different budgets.
- What are the non-negotiable loads? Write the list before you look at products — fridge and lights is one category, a ventilator or a server rack is another. (The second category rarely ends at "just an inverter.")
- Who actually maintains it? Honest answer, not the aspirational one. A generator nobody fuels, oils, and exercises is a liability with a pull cord. If there's no maintenance owner, a sealed battery system is usually the more realistic choice.
- What's the five-year cost? Sticker price is the least useful number in the conversation. Add fuel, maintenance, replacement intervals, and the downtime risk. Compare cost per usable kWh delivered, not cost per kW on the label.
My rule of thumb, after too many site visits: the grid connection determines your starting point, and the critical loads determine whether you stop at an inverter or add a generator. Everything between those two poles is a budget-and-redundancy conversation, and it's usually worth having it with a specialist who will tell you which parts of your problem aren't theirs to solve.
This is accurate as of Q1 2025. Inverter efficiency figures, portable generator specs, and grid-code editions change fairly often — verify current datasheets and local utility requirements before specifying anything. Pricing for commercial three-phase inverters in the 30 kW class, for reference, was quoting in the low-to-mid four figures per unit in our Q1 2025 vendor rounds; treat that as a planning range, not a quote.