A home standby generator is bolted to a pad beside the house, piped to your natural gas or propane, and wired to switch on by itself when the grid fails. That makes it a different noise problem from a portable. Our guide to the quietest generators for camping and RVs covers inverter generators you carry; this one covers the permanent kind, which runs for a few minutes every week whether or not the power ever goes out. If you would rather keep a portable for outages, our guide to portable generators for home backup covers the 240 V machines that feed a transfer switch or an inlet, and the CPSC rule that they run outside, 20 feet from the house.
The good news is that nearly every maker here measures at the same distance: 7 meters, or 23 feet. The bad news is that they stand the microphone in different places, and they publish two different kinds of figure. We read the product pages and spec sheets for more than thirty generators from five brands, put sixteen of them in our catalog, quoted every figure in the maker's words, and ranked each kind of figure only against its own kind.
Our picks are listed above, with full specifications for each.
Most standby spec sheets give two sound levels.
The exercise figure is the weekly self-test. The generator starts, runs for a few minutes and shuts down, usually at a lower engine speed. Generac calls this Quiet-Test low-speed exercise mode, Briggs & Stratton calls it Low Idle Mode, and Kohler calls it engine exercise. It is the noise your neighbors hear most often, on an ordinary afternoon.
The running figure is what you hear during an outage. Here the makers differ on the condition. Generac and Cummins say "at normal load". Briggs & Stratton says normal operation "based on average household usage". Kohler says "full-speed generator diagnostics and normal operation", and on three spec sheets adds "with no load". Champion gives one "Operational Volume" and no condition at all.
We never rank an exercise figure against a running figure. They describe different engine speeds, and the gap between them is large: Generac's 10 kW model lists 49 dB(A) in exercise and 61 at normal load, a 12 dB(A) jump. Our Generac generator decibels page lists every current Generac standby model the same way, exercise against exercise and running against running, with the liquid-cooled Protector series in a list of its own.
Same distance, different microphones
Every figure in this guide is at 7 meters. That is rare, and useful. But the makers do not measure from the same side of the machine.
- Generac measures from the front, and its footnote says "Sound levels taken from other sides of the generator may be higher depending on installation parameters."
- Kohler reports the "Lowest of 8 points measured around the generator" on most models. Its 10RESV sheet gives both: an 8-point average of 67 dB(A) in exercise, and 64 at the quietest point.
- Briggs & Stratton measures "Per ISO 3744", reporting the "lowest microphone at 7m".
- Cummins footnotes its figure as the "Quietest point at a normal load."
- Champion says only that sound was "measured at 23 ft. (7 m)".
A lowest-point figure is the most flattering number a maker can quote, and a front figure may or may not be. So treat a gap of a few decibels between brands as unknown. Our explainer on why you cannot compare dB ratings across brands goes into this in more detail.
All at 7 meters, in the maker's own mode:
- 49 dB(A): Generac 10 kW model 7257, Quiet-Test, front.
- 52 dB(A): Kohler 24RCLA and 38RCLC, liquid-cooled, lowest of 8 points, no load.
- 55 dB(A): Generac's newer 14, 18, 22, 24, 26 and 28 kW models, and the Guardian 14 and 18 kW.
- 56 dB(A): Kohler 26RCA, lowest of 8 points.
- 57 dB(A): Generac Guardian 10, 22, 24 and 26 kW.
- 62 dB(A): Briggs & Stratton PowerProtect+ 18kW and 22kW and both DX models, lowest microphone.
- 63 dB(A): Briggs & Stratton PowerProtect 26kW, and Kohler 14RCA.
- 64 dB(A): Briggs & Stratton PowerProtect 22kW.
- 65 dB(A): Briggs & Stratton PowerProtect 13kW.
- 66 and 67 dB(A): Kohler 20RCA, and the 10RESV as an 8-point average.
The gap from top to bottom is 18 dB(A). On the usual rule of thumb that 10 dB sounds about twice as loud, that is a large difference, and it is bigger than any plausible difference in method. Our decibel chart puts these levels beside everyday sounds.
At normal load, 7 meters: Generac's 10 kW models list 61 dB(A). Generac's 14 and 18 kW models and the Cummins RS13A, RS17A and RS20A list 65. The Generac EcoGen 15 kW lists 66. Generac's 22 to 28 kW models list 67.
Normal operation on average household usage, 7 meters, lowest microphone: Briggs & Stratton lists 65 for the PowerProtect 26kW, 66 for the PowerProtect+ range, 68 for the PowerProtect 22kW and 69 for the 13kW.
Kohler, full speed or normal operation: 60 for the 24RCLA and 38RCLC at no load, 67 for the 26RCA, 69 for the 14RCA and 70 for the 20RCA, all at the lowest point; 71 as an 8-point average for the 10RESV at no load.
No condition stated: Champion lists 62 dB(A) for its 14 kW aXis, 67 for the 22 kW aXis and 68 for its 26 kW fleX.
We do not rank across these groups. A Kohler figure at no load is not a figure at normal load, and Champion's 62 could describe either. Champion's smaller standby units publish figures too: our Champion generator decibels page lists the 8.5 kW Select Circuit 100177 at 59.5 dB(A) at 23 ft (7 m), the lowest Champion standby figure, again with no load stated.
When the page and the spec sheet disagree
Kohler's product pages and spec sheets give different numbers for four of the six Kohler models we checked. The 24RCLA page says 54 dB(A) in exercise and 61 at full speed; the sheet says 52 and 60, and the same page's feature text also says 52. The 38RCLC page says 54 and 62; the sheet says 52 and 60. The 14RCA page gives 67 at full speed, the sheet 69. The 20RCA page gives 64 and 69, while the sheet gives 66 and 70 and calls them both an average and the lowest of 8 points. Only the 26RCA matched, which is one reason it is our Kohler air-cooled pick.
Where they differ, we quote the spec sheet, because it states the distance and the method. Kohler's pages also say of the 26RCA that "No air-cooled generator is quieter". On exercise figures, Generac's air-cooled units list 49 and 55 against Kohler's 56. The methods differ, so neither claim settles it, but the Kohler slogan is not something the published figures support.
What the neighbors hear, and the rules that apply
There is no federal noise limit for a home standby generator. The rule that matters is your local noise ordinance, and it is measured at the property line, not at 7 meters.
Montgomery County, Maryland, is a useful example because its Department of Environmental Protection writes about standby generators directly. It says standby generators, which can operate on a 24-hour basis, "are required to meet Montgomery County's more restrictive nighttime standard of 55 dBA" at the nearest receiving residential property line. Its advice is to "Purchase the quietest generator available", to choose the location to minimize noise for neighbors, and to "Schedule the recommended periodic testing of generators to occur at times when building occupants and nearby neighbors are least likely to be disturbed." It adds that compliance is the property owner's responsibility. The county also gives a rule of thumb, that each doubling of distance from a point source brings about a 6 dBA drop. That is the county's estimate, not a maker figure, and we do not use it to convert any of the figures above.
Your county or city will have its own numbers and hours, and its own treatment of generators running during an outage. Read the code itself, or ask your permitting office, before you choose a spot for the pad. Our yard noise overview covers the other outdoor machines that count toward what neighbors hear.
Placement also has a safety side. Generac's spec sheets say its units can go "as close as 18 in (457 mm)" to a structure, but "Must be located away from doors, windows, and fresh air intakes and in accordance with local codes."
Exercise settings that lower the weekly noise
Since the exercise is the noise neighbors hear most, its settings matter as much as the figure. Generac's newer controller offers exercise weekly, bi-weekly or monthly, for 5 to 20 minutes, with a low-speed option. Its Smart Exercise skips a scheduled test if the generator ran for more than 5 consecutive minutes in the prior 72 hours. Kohler offers a 90-second eco-exercise on the 26RCA. Cummins describes a crank-only exercise that runs diagnostics without starting the engine. Ask the installer to set the quietest mode and a time of day your neighbors will not mind.
Liquid-cooled or air-cooled
Most homes get an air-cooled generator, which runs at 3,600 rpm. Kohler's 24RCLA and 38RCLC are liquid-cooled, run at 1,800 rpm, and list 52 dB(A) in exercise and 60 running at no load, the lowest Kohler figures. They also weigh well over 1,000 lb shipped and start at $17,900 and $22,949 MSRP. For most buyers, a smaller air-cooled unit sized to the house is the better route to a lower running figure: across Generac's range, the running figure climbs from 61 at 10 kW to 67 from 22 kW up.
Brands that publish less
Champion gives an Operational Volume at 23 feet with no load or mode. Cummins gives a normal-load figure but no exercise figure, on a spec sheet dated 2019, and cummins.com refused our requests, so we could not check a current page or a price. Generac's EcoGen 15 kW lists 66 dB(A) at normal load and no Quiet-Test figure, and its sheet says it is designed for use with alternative energy and storage. We did not find a PowerPact model in Generac's current standby listing. Briggs & Stratton's product pages list "Normal Operating Sound" with no distance; the distance and the ISO 3744 method are in its spec sheets.
How we chose
We read each maker's product page and spec sheet for more than thirty home standby generators sold in the US, and recorded every sound figure verbatim with its distance, mode, load and measuring position. We ranked exercise figures only against exercise figures, and running figures only against running figures with the same stated condition. We then chose picks on those figures, output, weight, warranty and the maker's own MSRP, which excludes installation. Where a product page and a spec sheet disagreed, we used the spec sheet.
We did not measure any of these generators. Every figure is the maker's own, quoted as published, and we did not convert any figure to another distance.
