The efficiency gap between AC-coupled and DC-coupled storage is real, and in the case worked below it is worth about 1.5 percent of the solar energy that passes through the battery. The figure that decides most commercial projects is a current limit, not an efficiency figure.
The quoted arithmetic and the quantity it belongs to
The standard argument is that DC coupling converts less often. It is correct, and it is attached to the wrong quantity.
A storage cycle converts twice in either architecture: once charging, once discharging, whether the converter sits on the AC bus or on a shared DC bus.
The difference sits in the surplus solar path. In a DC-coupled system the array's surplus moves from DC to DC into the battery. In an AC-coupled system the same surplus is inverted to AC by the array's inverter and rectified back to DC by the battery's converter. Two conversions, not one.
The published maximum efficiency for integrated cabinets in this class is 98.5 percent. One conversion keeps 98.5 percent of the energy; two keep 97.0 percent, because 0.985 multiplied by 0.985 is 0.9702. The gap is about 1.5 percentage points, and it applies only to the solar energy routed through the battery, not to the site's consumption as a whole.
Worked example. A 300 kWp commercial array producing 330 MWh a year, at roughly 1,100 kWh per kWp, with 40 percent of generation as surplus and a battery that absorbs 80 percent of it. The energy routed through the battery is 105.6 MWh a year. One conversion delivers 104.0 MWh of it; two deliver 102.5 MWh. The entire prize for DC coupling is 1.6 MWh a year. The published 98.5 percent is a maximum at rated power, so at part load the gap widens.
Where the two architectures put the ceiling
Efficiency is a loss question. Sizing is a ceiling question.
In AC coupling the battery owns its converter, so its ceiling is that converter's rating. The ceiling does not move when the array changes, and it does not move when battery voltage falls.
In DC coupling the array and the battery share one bus inside one published voltage window. Sharing it makes the published DC current limit an operating limit on both, and available power falls with bus voltage.
The published specification states that model BS07-265-ES-X, the integrated 265 kWh cabinet in this range, is rated 125 kW AC, with a rated battery voltage of 844.8 V, an operating window of 739.2 to 950.4 V, and a maximum charge and discharge current of 157 A.
At 950.4 V, 157 A is 149.2 kW. At 739.2 V the same 157 A is 116.1 kW. The cabinet has 28.6 percent more DC-side power at the top of its published window than at the bottom, where it sits 7.2 percent below its own 125 kW AC rating.
Units and conditions: current in A and voltage in V at the battery DC terminals; power in kW; 98.5 percent is the published maximum at rated power, not an average.
The second limit, on the inverter side
The published specification for model DWH-29.9~60K-HP3EU-S, the three-phase C&I hybrid inverter in the same range, states a battery voltage range of 135 to 850 V and dual battery inputs rated 100 A each, 200 A in parallel.
200 A equals 70 kW only at 350 V or above. At 135 V the same port moves 27 kW. A DC-coupled design that pairs this inverter with a lower-voltage battery stack does not lose a percentage point of surplus solar; it loses most of the nameplate. Neither figure says the equipment is unsuitable. They say the battery voltage window has to be checked before the inverter model is fixed.
Ruibit distributes Dawnice equipment, so every figure used here is the manufacturer's published number rather than a measurement taken by the party selling it.
The comparison only holds at one boundary
Two kinds of number are compared throughout, and they do not live in the same place. The 157 A and 200 A limits are DC-side limits. The 125 kW and 70 kW figures are AC-side ratings. Round-trip efficiency is neither; it runs from the battery terminals back to them, and system efficiency includes cooling and auxiliary load.
Energy at the battery terminals, at the converter's AC terminals and at the point of common coupling are three different quantities. A proposal that compares one with another without naming the boundary has not made a comparison. The 1.5 percentage points above are honest only because both sides sit on the same boundary.
What to put in front of the supplier
Three questions settle this before a layout is drawn.
Ask for the battery's full published voltage window and its maximum charge and discharge current, not only the kWh and the kW. A supplier who answers with capacity and power alone has withheld the two numbers the ceiling depends on.
Ask, in writing, how many times surplus solar converts before it reaches the battery. A supplier who replies that DC coupling is more efficient, without stating the stages, is repeating the argument rather than pricing it.
Ask which party owns a fault on the shared DC bus, and get the answer into the warranty documents before the order. If the reply is that the integrator will sort it out, the reply has not been given.
What stops being negotiable at signature
The array's string layout and MPPT allocation are fixed once the array is re-strung into a hybrid inverter. Returning to the original arrangement means rebuilding the DC side.
The battery's converter rating is fixed once the cabinet is ordered. No firmware change raises a current limit.
The grid connection application names converter models and quantities, and adding a second converter after approval cannot be revisited without reopening the application.
The measurement boundary in the acceptance test is fixed once the capacity test is signed off, and it cannot be reopened afterwards without a second test at the customer's cost.
The assumption worth dropping is that efficiency picks the architecture. Efficiency prices it; the published current limit and the voltage window are what choose it.
FAQs
1. Which coupling type is more efficient?
DC coupling converts less often on the surplus solar path, so it loses less energy there: about 1.5 percentage points of the solar energy routed through the battery, at a published maximum efficiency of 98.5 percent. That difference applies to battery-routed solar only, not to the site as a whole, because charge and discharge convert once each in both architectures.
2. Can a hybrid inverter work in an AC-coupled system?
Yes. The published specification for model DWH-29.9~60K-HP3EU-S states that it supports AC coupling and up to eight units in parallel, so a hybrid inverter is not automatically a DC-coupled decision. The coupling type is set by where the battery connects, not by the inverter's label.
3. What limits battery power in a DC-coupled design?
The published DC current limit together with the bus voltage. On the BS07-265-ES-X cabinet, 157 A gives 149.2 kW at the top of its published voltage window, 950.4 V, and 116.1 kW at the bottom, 739.2 V, which is 7.2 percent below its 125 kW AC rating. At low state of charge the current limit, not the kW rating, sets the ceiling.
4. What should a buyer request before the coupling is fixed?
Four things: the battery's full published voltage window, its maximum charge and discharge current, the number of conversions on the surplus solar path, and the party that owns a fault on the shared DC bus. Capacity and power ratings alone do not answer the ceiling question.
5. What cannot be changed after the contract is signed?
The array's string layout and MPPT allocation once the array is re-strung, the battery converter rating once the cabinet is ordered, the converter list in the grid connection application, and the measurement boundary in the acceptance test.
