AC couplingDC couplinground trip efficiencydatasheetssystem design

AC-coupled vs DC-coupled home batteries: what conversion costs

AC-coupled vs DC-coupled home batteries: one maker publishes 89 percent AC coupled and 94 percent DC coupled for the same battery, with the measurement basis.

The Home Battery ReportAugust 19, 20267 min read✓ Verified

Every conversion between DC and AC loses a little energy. A solar panel makes DC. A battery stores DC. A house consumes AC. So the design question is how many times you cross that boundary between the roof and the outlet, and the two answers to it have names.

The argument gets oversold in both directions. It is a real engineering difference with a published number attached, and that number is smaller than most sales conversations imply. Every figure below is a manufacturer specification quoted with the basis the datasheet states. We have not tested any of these products.

The two paths a solar electron takes

DC coupled. The array and the battery share one inverter, usually called a hybrid inverter. Solar comes down from the roof as DC, a charge stage puts it into the battery as DC, and it becomes AC once on its way out to the house. Solar into the battery crosses the boundary zero times. Battery out to the house crosses it once.

AC coupled. The array has its own solar inverter and the battery has its own inverter. Solar comes down as DC, the solar inverter turns it into AC, the battery's inverter turns that AC back into DC to store it, and turns it into AC again to discharge. Solar into the battery crosses the boundary twice. Battery out to the house crosses it once more.

That is the whole mechanism. Two extra conversions on the solar charging leg, and only on that leg. Charging from the grid is identical on both architectures, because grid power arrives as AC either way and has to be converted once to be stored. This is why the coupling question disappears entirely for a battery installed without solar, which is the case covered in our no-solar guide.

The one datasheet that prices the difference

Most of the efficiency claims you will read compare one maker's DC-coupled product against a different maker's AC-coupled product, which mixes the architecture difference with every other difference between two pieces of hardware. Panasonic removes that problem by publishing both numbers for the same system.

The Panasonic EverVolt 2.0 datasheet lists a single row, Battery Roundtrip Efficiency, with two values: AC coupled 89 percent and DC coupled 94 percent. The two footnotes on that row are what make the figures usable. Footnote 1, attached to the 89 percent: "At the beginning of battery life, AC-to-AC at 50% power rating." Footnote 2, attached to the 94 percent: "At the beginning of battery life, DC-to-AC at 50% power rating." Source: Panasonic EverVolt Home Battery System datasheet, marked preliminary, rev 2023.

Five percentage points, same manufacturer, same cabinet, same test power, both bases printed. Note also that the two figures are not measuring identical journeys: one is AC in and AC out, the other is DC in and AC out, which is exactly the difference the architectures create.

What five points is worth depends on how much of your battery's throughput comes from solar rather than the grid. On a system that charges mostly from the array, it is five percent of a large number over many years. On a system that charges mostly off-peak from the grid to arbitrage a time-of-use rate, it is close to nothing, because grid charging takes the same path on both architectures. The general mechanics of what round trip losses cost are in our round-trip efficiency glossary entry.

Why Tesla's 89 percent is not the same 89 percent

The Tesla Powerwall 3 datasheet publishes two efficiency lines and neither is an AC-to-AC round trip:

Solar to Battery to Home/Grid Efficiency: 89 percent, footnoted as a typical solar shifting use case.
Solar to Home/Grid Efficiency: 97.5 percent, footnoted as tested using CEC weighted efficiency methodology.

Source: Tesla Powerwall 3 datasheet.

The first figure measures a solar round trip on an integrated solar-and-battery product. It is not the AC-to-battery-to-AC figure that the Tesla Powerwall 2 publishes at 90 percent and footnotes explicitly as "AC to battery to AC, at beginning of life", and it is not the number FranklinWH publishes at 90 percent labelled "Round Trip Efficiency (Grid to Battery to Load)". Lining Tesla's 89 up against a rival's AC-to-AC 90 and declaring a winner is a category error, and it is one that appears on comparison sites that look authoritative. The full model-by-model breakdown of who measures what is in our round trip efficiency comparison.

The coincidence is worth naming, because it is confusing: Panasonic's AC-coupled figure and Tesla's solar round trip figure are both 89 percent, and they are measurements of different things.

What each tracked model publishes about its own coupling

We record the term only where the manufacturer prints it. Where a datasheet publishes no coupling statement, we say not published rather than inferring one from a wiring diagram or a retailer page.

| Model | What the datasheet publishes |

| --- | --- |

| FranklinWH aPower 2 | "Coupling: AC-Coupled" as a spec row; headline "AC-coupled battery" |

| Enphase IQ Battery 5P | "all-in-one AC-coupled system" |

| Enphase IQ Battery 10C | "all-in-one AC-coupled system" |

| Generac PWRcell | "the AC-coupled PWRcell 2 Home Energy Storage System"; SDS supports "AC-coupled PV" |

| sonnen sonnenCore+ | "Grid integration: AC coupled" |

| Tesla Powerwall 2 | "a fully-integrated AC battery system" |

| SolarEdge Home Battery 400V | "DC coupled battery"; "Compatible Inverters: SolarEdge Home Hub Inverters" |

| Qcells Q.HOME CORE | Datasheet heading "H3S / H7S : DC or AC-coupled" |

| Fortress Power eVault Max 18.5 | "The eVault Max is AC/DC coupled to solar arrays" |

| Savant Power Storage 20 | "AC Coupled Solar Support: Yes", alongside a DC solar section with 17 kW maximum solar input and 2 MPPTs |

| Tesla Powerwall 3 | No coupling term published; the datasheet publishes PV DC inputs, six MPPTs and 20 kW maximum solar STC input |

| Anker SOLIX X1 | No coupling row in the spec tables; the US document is Anker's AC-coupled datasheet |

| EG4 PowerPro WallMount | Not published; a DC module offering "selectable closed-loop communications with EG4, Schneider, Solark, Victron, Growatt, Megarevo, Luxpower, and Deye inverters" |

| BYD Battery-Box Premium HVS | Not published; a DC battery for a compatible external hybrid inverter |

| LG Home 8 | Not published |

| Panasonic EverVolt 2.0 | Both, and priced: separate AC coupling and DC coupling system diagrams and separate efficiency figures |

| SunPower SunVault | Not published. Discontinued |

Three things fall out of that table. Most makers who name a coupling name AC, because AC coupling sells into the larger market of houses that already have solar. Several products are configurable rather than fixed, which means the question belongs to the system design and not to the battery. And the DC-only battery modules leave the answer entirely to whichever inverter an installer pairs them with, which is why their published power figures assume an inverter you have not bought yet.

Where DC coupling loses, and it is not on efficiency

The efficiency argument favours DC coupling. Almost everything else favours AC coupling, and the reasons are practical.

A DC-coupled battery is not inverter agnostic. SolarEdge publishes "Compatible Inverters: SolarEdge Home Hub Inverters" for its Home Battery 400V. That is a clean sentence and a hard constraint: on any array not already running a SolarEdge Home Hub inverter, adding that battery means replacing the inverter.

The DC path does not exist on a microinverter array at all. If every panel has its own inverter on the roof, the DC has already become AC before it reaches the house, and there is no shared DC bus for a battery to join.

Multi-box DC systems put the power rating on a component you buy separately. The BYD, Fortress and EG4 modules publish battery-side currents and voltages, and the AC power your house actually sees is set by the paired inverter. That is not a flaw, but it does mean a DC battery's spec sheet answers fewer of your questions than an all-in-one's does.

Against that, DC coupling on a purpose-built new system removes a box from the wall, removes a set of conversions, and removes one warranty relationship. Where a maker offers both, as Qcells and Panasonic do, and the design is new rather than a retrofit, the DC configuration is usually the better engineering answer.

What actually decides it

For most buyers, the coupling choice is made by the situation rather than chosen on merit.

Existing solar, string inverter with years of warranty left: AC coupling, almost always. The retrofit mechanics are in adding a battery to existing solar.

Existing solar, microinverters: AC coupling, by construction.

Existing solar, ageing inverter: the choice reopens, because replacing the inverter is on the table anyway and a battery with a solar inverter inside it can take the array over.

New solar and storage together: DC coupling is worth asking for, and worth roughly five points of solar round trip efficiency on the one datasheet that publishes both.

No solar: the question does not apply.

Every figure on this page is a manufacturer datasheet claim recorded with its published basis, verified against the documents cited on our model pages. Manufacturers revise datasheets and the same model can carry different figures across revisions, so confirm the current document for the configuration you are quoted.

Frequently asked questions

What is the difference between an AC-coupled and a DC-coupled battery?

It is where the battery joins the system electrically. A DC-coupled battery shares one inverter with the solar array, so solar arrives as DC and is stored as DC without ever becoming AC. An AC-coupled battery has its own inverter, so solar is converted to AC by the solar inverter and then converted back to DC by the battery's inverter before it is stored. The DC path has fewer conversions on the solar-to-battery leg, which is where the efficiency argument comes from.

How much efficiency does AC coupling actually cost?

Panasonic is the one maker in our set that publishes both figures for the same product. Its EverVolt Home Battery System datasheet lists Battery Roundtrip Efficiency as AC coupled 89 percent and DC coupled 94 percent, and footnotes the first as AC-to-AC at 50 percent power rating at the beginning of battery life and the second as DC-to-AC at 50 percent power rating. Five points, from the same manufacturer, on the same hardware, with both bases printed. Treat that as the honest size of the gap rather than the larger numbers that circulate.

Is Tesla's 89 percent the same kind of number?

No, and this is the most common misreading in the category. The Powerwall 3 datasheet publishes 89 percent as Solar to Battery to Home/Grid Efficiency, which is a solar round trip, and separately publishes 97.5 percent as Solar to Home/Grid Efficiency tested using CEC weighted efficiency methodology. Neither is the AC-to-AC round trip that most rivals quote. Setting Tesla's 89 against another maker's AC-to-AC 90 is comparing two different measurements, and Panasonic's 89 percent AC coupled figure is coincidentally the same number measured a different way.

Which home batteries are AC-coupled and which are DC-coupled?

Where the datasheet says so: FranklinWH prints Coupling: AC-Coupled, Enphase describes both the 5P and the 10C as an all-in-one AC-coupled system, Generac calls the PWRcell 2 AC-coupled, sonnen publishes Grid integration: AC coupled, and the Tesla Powerwall 2 is described as a fully-integrated AC battery system. SolarEdge describes its Home Battery 400V as a DC coupled battery. Qcells heads its datasheet DC or AC-coupled, Fortress says the eVault Max is AC/DC coupled to solar arrays, and Savant publishes both a DC solar input and an AC Coupled Solar Support row. Several others publish no coupling term at all, and we record that rather than inferring one.

Does DC coupling matter if I have no solar?

No. Both arguments are about solar. A battery charging from the grid takes AC in and gives AC back regardless of how it would have been wired to an array, so the coupling question collapses and what is left is the plain AC-to-AC round trip figure. Our guide on batteries without solar covers what that does to the economics.

Should I choose a battery based on coupling?

Rarely. On a new build with a maker who offers both, DC coupling is the better engineering answer and the five point efficiency difference is real. On a retrofit, the constraint usually decides for you, because DC coupling means the battery shares an inverter and that inverter has to be one the battery maker supports. Warranty terms, continuous power and the size of the backed-up panel move the total outcome more than five points of round trip efficiency does.