How loud is a home battery? Noise ratings from datasheets
Thirteen of the 17 home batteries we track publish a decibel figure. Only five say at what distance they measured it, and four publish nothing at all. What each datasheet actually states, and what a single number leaves out.
"Silent" is the word that appears in home battery marketing, usually next to a photo of a unit on a spotless garage wall. It is not a specification. So we opened the manufacturer datasheet behind each of the 17 models on our comparison pages and wrote down exactly what each one publishes about acoustic noise, including the basis it was measured on.
Thirteen publish a number in some form. Five of those state the distance it was measured at. Four publish nothing at all, and two of the thirteen publish a figure for a different box in the system rather than for the battery itself. That distribution is the actual answer to how loud a home battery is: mostly quiet, occasionally not, and rarely documented well enough to compare two of them.
We have not tested any of these products. Everything below is a published manufacturer figure quoted with its source.
Figures published with a stated measurement distance
These five are the only ones in our set you can meaningfully compare to each other, because they all state 1 m.
- SolarEdge Home Battery, 400V: under 25 dBA, published as "Noise (at 1m distance)". The lowest figure with a stated basis in our set. Source: SolarEdge Home Battery 400V datasheet, Jan 2026.
- Enphase IQ Battery 5P: under 30 dBA maximum, published as "Operating noise @1m". Note that Enphase labels its figure a maximum rather than a typical. Source: Enphase IQ Battery 5P datasheet, May 2025.
- Enphase IQ Battery 10C: under 33 dBA, published as "Operating noise @1m". Source: Enphase IQ Battery 10C datasheet, Feb 2026.
- SunPower SunVault: under 50 dBA, published as "Acoustic Noise Level <50dBA @1m distance, 86 degrees F (30 degrees C)". The only figure in the set that states both a distance and an ambient temperature. Discontinued, and listed here for owners. Source: SunPower SunVault ESS datasheet, RevC 2021.
- Tesla Powerwall 3: under 50 dB(A) typical and under 62 dB(A) maximum, published as "Operating Noise @1m". The highest published maximum among the models we track. Source: Tesla Powerwall 3 datasheet.
Figures published without a distance
Eight more publish a decibel figure and leave out the condition that would make it comparable.
- FranklinWH aPower 2: 30 dBA typical, 45 dBA maximum. The only unit besides the Powerwall 3 that publishes both a typical and a maximum. Source: FranklinWH aPower 2 datasheet, Dec 2025.
- Tesla Powerwall 2: under 40 dBA at 30 degrees C (86 degrees F). Tesla stated a temperature on the older unit and a distance on the newer one, which means the two Powerwall figures are not comparable to each other either. Discontinued. Source: Powerwall 2 AC datasheet.
- Anker SOLIX X1: under 35 dB at ambient below 104 degrees F, under 40 dB at ambient above 104 degrees F. No distance, but the only maker in the set to publish a noise figure that changes with temperature, which is a fair way to describe a unit that works its thermal management harder when it is hot. Source: Anker SOLIX X1 AC-coupled datasheet, US, rev 2025-03.
- LG Home 8: under 47 dB. LG also publishes the cooling method, which explains the figure: forced air cooling by fan on the Home 8, natural convection on the SE Box. Source: LG official spec page for the RA768K16A11.
- sonnen sonnenCore+: under 25 dB. sonnen publishes natural convection as the system cooling method and markets the unit for indoor installation. Source: sonnenCore+ technical specifications.
- Savant Power Storage 20: under 55 dB. The highest single published figure in the set. Savant does not state distance, mode or temperature. Source: Savant Power Storage 20 specification sheet, 009-2477-00.
- Panasonic EverVolt 2.0: under 40 db, for the SmartBox only. This is the case worth reading twice. The 40 db figure sits in the SmartBox mechanical table, next to "Cooling: Fan (user replaceable)". The battery cabinet and inverter tables publish no noise figure at all, and the inverter is listed as natural convection cooled. So Panasonic's published number describes the control box, not the battery. Source: Panasonic EverVolt Home Battery System datasheet, marked preliminary, rev 2023.
- Qcells Q.HOME CORE: under 30 dBA for the Q.VOLT inverter and under 50 dBA for the Q.HOME HUB. Same pattern as Panasonic. The Q.SAVE battery block itself has no noise line and is listed as natural convection cooled, while the hub that does carry a noise figure is fan cooled. Source: Qcells Q.HOME CORE datasheet, H3S and H7S, May 2025.
Four datasheets publish nothing
- Generac PWRcell. We read the full PWRcell 2 spec guide, covering the system, the Smart Disconnect Switch, the battery cabinet, the DCB battery module and the inverter. There is no acoustic line in any of the five tables. Source: Generac PWRcell 2 spec guide, Jul 2025.
- EG4 PowerPro WallMount. No noise figure. Source: EG4 14.3 kWh PowerPro WallMount all-weather battery specification sheet, ver. 1.0.1.
- Fortress Power eVault Max 18.5. No noise figure. Source: Fortress Power eVault Max 18.5 technical datasheet.
- BYD Battery-Box Premium HVS. No noise figure. Source: BYD Battery-Box Premium HVS and HVM datasheet V1.7 EN.
Three of those four are DC batteries with no inverter and no power electronics of their own, which is a plausible reason to publish nothing: on those products the audible hardware is the inverter you buy separately, and it comes with its own sheet. Generac is the exception, a full AC system with an inverter, a switch and a cabinet, and no acoustic figure anywhere.
What a single number leaves out
Two makers publish both a typical and a maximum, and both gaps are large. Tesla prints under 50 dB(A) typical and under 62 dB(A) maximum for the Powerwall 3, a 12 dB spread. FranklinWH prints 30 dBA typical and 45 dBA maximum for the aPower 2, a 15 dB spread. Decibels are logarithmic, so those are not small differences in perceived loudness.
Every other product in the set publishes one number, and does not say which kind it is. A unit quoting 40 dB might be quoting a typical figure with a 55 dB worst case, or a maximum with a much quieter typical. There is no way to tell from the sheet, and a battery is loudest exactly when it is working hardest, which is the hot afternoon it is charging at full rate or the outage evening it is carrying the house.
Distance compounds it. As a rough physical rule, doubling the distance from a source drops the sound pressure level about 6 dB in open air, and a real garage with hard walls does not behave like open air. So a figure measured at 1 m and an unlabelled figure are not interchangeable, and the eight unlabelled figures above could each sit several decibels either side of where they look.
On reference points, we will not import a chart of everyday sound levels and pretend it is precise. Published charts disagree with each other by roughly 10 dB on the same appliance, some putting a refrigerator at 40 dB and others at 50. What is safe to say is that the published figures in this set span the range from something you would need a quiet room to hear at all, in the low 20s, up through the range where a household appliance is plainly audible from across a room, in the high 40s and above. The Powerwall 3's published maximum of under 62 dB(A) at 1 m sits above that, in the range where a conversation next to it would be affected.
What to do with this before you sign
Ask three questions, and ask them of the model in the quote rather than the brand.
- Is the number a typical or a maximum, and at what distance? If the sheet does not say, ask the installer to get it in writing from the manufacturer. That is a fair question about a spec the maker chose to publish incompletely.
- What is the cooling method? Where a maker publishes it, natural convection tracks with the low figures in this set and forced air with the high ones. It is on the sheet for LG, Panasonic, Qcells, sonnen and Enphase's controller.
- Which box is the number about? On Panasonic and Qcells the published noise figure belongs to a control box, not the battery. If the control box is going on the other side of a bedroom wall, that is the figure that matters, and it is the fan-cooled one.
Where a unit's placement is the real question rather than its decibels, where to install a home battery covers the enclosure ratings and clearances, and home battery dimensions and weight covers what physically fits. Every figure above is a manufacturer datasheet claim recorded with its published basis, or the absence of one, and none of it is a measurement we made.