Decision framework · sizing by proportion
The Coverage Fraction: How Much of Your House Are You Actually Backing Up — Honda EU7000iS or a Generac Guardian?
Most sizing arguments fight over absolute watts. The more useful number is a ratio: of everything in your panel, what fraction do you genuinely need alive during an outage? That single proportion — not the brand, not the kilowatt rating — sorts a portable inverter from a permanent standby. This framework builds the decision around it.
The two candidates sit at opposite ends of the fraction. A Honda EU7000iS delivers 5,500 W running / 7,000 W starting on gasoline, ~52 dBA, ~16 h per 5.1-gal tank, paralleling to ~14,000 W — it is sized to back up a slice. A Generac generator Guardian 24 kW (24 kW LP / 21 kW NG, 200 A service-rated ATS, Smart Management Modules) is sized to back up the whole panel and let load management trim only the edges. Define the fraction, and the rest follows.
f = Pbacked / Phouse
the share of your home's possible peak draw you actually intend to keep running during an outage
Dimension 1 — Where the fraction lands on the capacity curve
Mechanism: a typical all-electric home's coincident peak can run on the order of 15–25 kW once central AC, an electric range, a dryer, and a water heater can all fire together (illustrative). If your essential slice is a fridge, freezer, furnace blower, sump, lights and networking — roughly 2 kW running — then f is only about 0.1 of that peak. The Honda's 5.5 kW comfortably spans a 2 kW slice; the Guardian's 24 kW is built for f approaching 1.0.
Worked consequence: At f ≈ 0.1, buying 24 kW means paying for — and idling — roughly ten times the capacity you'll draw, since fuel burn ≈ load × bsfc keeps a large engine spinning a barely-loaded alternator. Decision driven: when your honest coverage fraction is small, the portable isn't a compromise — it's the correctly sized tool, and the standby's headroom is capacity you rent without using.
When this reverses: the moment your intended fraction climbs toward "everything I normally use" — AC included — f jumps past what 5.5 kW can hold, and only a panel-scale machine satisfies it.
Dimension 2 — The fraction hides a worse number: peak-to-average
Mechanism: f built from running watts understates the problem if the slice contains a hard-starting motor. Motor sizing is locked-rotor amps (LRA) against the source's surge ceiling, not running watts against its running rating. A single ½-hp well pump or a small AC compressor can spike to several times its run draw for a fraction of a second. The Honda buffers a 7 kW surge briefly; the Guardian's larger rotating mass and load-management board stage and absorb stacked starts.
Worked consequence: Two homes can share an identical f ≈ 0.12 by running watts, yet one contains a 4-ton AC whose inrush (illustratively ~16–17 kW) blows straight past the Honda's 7 kW surge — disqualifying the portable despite a "small" fraction. Decision driven: compute the fraction twice, once for steady watts and once for the single worst inrush; if the inrush version exceeds ~7 kW, the standby is selected regardless of how modest the average looks.
When this reverses: fit a soft starter and the same compressor's inrush can roughly halve; a borderline motor may drop back under the Honda generator's ceiling, returning the decision to the steady-state fraction.
Dimension 3 — The fraction is multiplied by event duration and presence
Mechanism: a small f is cheap to serve only if the event is short and someone is home. Runtime scales with the tank: ~16 h on the Honda, then a manual refill. The Guardian's runtime is whatever the gas pipeline sustains, and its ATS arms the slice with no human present. So the true cost of a given fraction is f × duration × (attended?) — a 2 kW slice for 6 hours with the owner home is trivial; the same slice for 4 days while traveling is not.
Worked consequence: A household that backs up f ≈ 0.1 but loses power for three days every winter while often away discovers the portable produces nothing on day two without a person to refuel it. Decision driven: a small fraction does not by itself favor the portable; multiply it by duration and presence first. Long, unattended events promote even a tiny slice to standby territory.
When this reverses: for short, attended outages — the common case in many suburbs — duration and presence both fall away, and a small fraction stays firmly in the portable's lane.
| Your situation | Coverage fraction f | Worst inrush | Duration × presence | Framework picks |
|---|
| Essentials only, home, short outages | ~0.1 | < 7 kW | Short, attended | Honda EU7000iS |
| Essentials + central AC | ~0.3 | > 7 kW | Any | Generac Guardian |
| Small slice, but multi-day & away | ~0.1 | < 7 kW | Long, unattended | Generac Guardian |
| "Run the whole house, no thinking" | ~1.0 | — | Any | Generac Guardian |
The rule, stated as a threshold on the fraction:
Compute f honestly, then check two multipliers.
• If f is below roughly 0.25 of your panel's coincident peak, the worst single inrush stays under ~7 kW, and your events are short and attended → the Honda EU7000iS is the right-sized choice; the standby's extra capacity would idle.
• If f rises above ~0.25, or any one inrush tops ~7 kW, or the slice must run multi-day and unattended → a single "yes" promotes you to the Generac Guardian, and no smallness elsewhere in the fraction pulls you back.
The decision was never the kilowatt rating on the box. It was the fraction of your house you meant to keep alive — and the two multipliers that scale it.
Topology/standards per the cited standards; all product ratings are manufacturer-stated values from the cited datasheets, current to 2026-06; derived/illustrative figures are labelled as such. This is not an independent head-to-head test. Honda is a brand affiliated with this site; competitor names are used for identification only.
Leave a Reply