AC-coupled or DC-coupled energy storage: Which architecture is better?

If you’re planning to add an energy storage system to a solar power system, the very first question isn’t about capacity, but about architecture. This decision determines the system’s efficiency, scalability, and installation cost. In this article, we’ll walk through how AC- and DC-coupled systems work, what the most important technical difference is between AC coupling and DC coupling, and what AC and DC current actually mean in practice.

We’ll examine which solution is more efficient in terms of energy conversion, when AC coupling is advantageous for expanding an existing solar array with a battery, and when it’s still worth choosing a DC-coupled energy storage system. We’ll cover inverter compatibility—whether to keep your existing inverter or replace it with a hybrid inverter—as well as power and capacity sizing for retrofit installations.

Finally, we quantify conversion losses, examine the issues of zero export and energy management in retrofit systems, compare the two architectures on a larger commercial and industrial scale, and conclude with SOLARKIT’s decision-making criteria for installers. If solar panels or batteries are added to the system later, this set of criteria will help you decide which approach is more cost-effective.

How do AC- and DC-coupled systems work?

The difference between the two architectures lies in where the solar panels and the battery circuit connect.

In an AC-connected system, the solar panel feeds into its own inverter, while the energy storage unit feeds into a separate battery-based hybrid inverter. The two devices connect on the alternating current side, typically at the distribution panel busbar. The electricity generated by the solar panel is first converted to alternating current, then, when it enters the battery, it is converted back to direct current, and during discharge, it undergoes a third conversion. The advantage of this configuration is that the two devices operate independently of each other: if one fails, the other continues to function.

In a DC-coupled system, a single hybrid inverter is used. The solar panels and the battery are connected to a common DC bus: on one side is an MPPT controller, and on the other is a battery charger DC/DC converter. The direct current from the solar panels charges the battery directly, without conversion, and is converted to alternating current only when it is fed to the loads or the grid.

Everything else—efficiency, scalability, redundancy, and installation complexity—stems from this topological difference. The same logic applies to a residential system as to an industrial facility, just on a different scale.

AC Coupling and DC Coupling: The Most Important Technical Differences

In practice, the two architectures differ in four key aspects.

Number of conversions. With AC coupling, solar energy is converted three times; with DC coupling, only once. This accounts for the difference in efficiency: when charged from solar panels, AC-coupled systems typically achieve a cycle efficiency of 90–94 percent, while DC-coupled systems achieve 95–98 percent.

Recovery of Clipped Energy. If the output of the solar array exceeds the inverter’s rated AC power, the excess is lost. A DC-coupled energy storage system stores this clipped energy before it reaches the inverter, whereas an AC-coupled system can only charge once the energy has already passed through the inverter.

Location and Redundancy. An AC-connected energy storage system can be installed freely at a greater distance from the solar array, and the two inverters operate independently. In a DC-connected configuration, the battery must remain near the inverter, and a failure in a single component shuts down the entire system.

Cost and installation. For greenfield projects, DC-coupled systems are cheaper because a single hybrid inverter is sufficient. For existing systems, however, AC-coupled systems are the faster option because they do not require modifications to the operational solar array.

Meaning of AC and DC Current

It’s worth clarifying the basics, as all the differences mentioned above are based on this.

DC (direct current) flows in one direction and has a constant voltage. Solar panels generate this type of current, and batteries also store energy in this form. In a residential string, this typically means a DC voltage between 600 and 1,000 V.

AC (alternating current) periodically changes direction, at a frequency of 50 Hz in Europe. The utility grid and most household appliances operate on this type of current.

The inverter connects the two: a DC/DC stage adjusts the voltage level, while a DC/AC stage converts the output into a sinusoidal 230-volt alternating current. Every such conversion results in a loss, so it is important to consider how many stages the current passes through.

Which solution is more efficient in terms of energy conversion?

On paper, the answer is clear: DC coupling. When charged by solar panels, DC-coupled systems operate at 95–98 percent efficiency, while AC-coupled systems operate at 90–94 percent efficiency, because the current passes through fewer conversion stages.

In practice, however, what matters is the path the energy takes.

If the solar panel’s output goes directly to consumers, both architectures perform the same single DC/AC conversion. The difference in efficiency applies exclusively to stored energy; in other words, the greater the system’s storage ratio, the more significant this difference becomes.

When charging from the grid, the situation is reversed: the efficiency of the DC-connected system drops to around 87 percent due to the additional conversion, while the performance of the AC-connected system remains more balanced regardless of the charging source. This factor becomes increasingly important wherever dynamic pricing or grid services are part of the business model.

AC-coupled systems may also perform better during blackouts: in such cases, the power output of the solar inverter and the battery inverter is combined, allowing for higher peak consumption.

In numbers: for a 10 kWh battery, a 3–5 percent difference in efficiency translates to an annual difference of around 200–400 kWh. This figure alone rarely justifies replacing an inverter, but it does matter in greenfield projects.

Expanding an Existing Solar System with a Battery: When Is AC Connection Advantageous?

Expanding an existing solar system with a battery is a common project in most European markets today. Most systems installed in the second half of the 2010s use standard grid-tied string inverters, which cannot accommodate a battery.

In such cases, an AC connection is clearly the better choice in four scenarios.

If the existing inverter is in good condition and still under warranty. It makes no sense to throw away a working device just to gain a few percentage points in efficiency. Moreover, the residual value of a dismantled inverter is a fraction of that of a new one.

If it’s not worth disrupting the existing billing arrangement. More favorable, previously secured rates are typically tied to the registered generation capacity and the grid connection agreement. An AC-side expansion doesn’t affect these, but replacing the inverter may trigger a new permitting process.

If the solar array and the battery are located in different places. The battery-integrated inverter can be installed anywhere near the distribution panel.

If rapid installation is the goal. Among EcoFlow energy storage systems, there are both All-in-One models and traditional configurations with separate inverters. For retrofitting, the All-in-One version is the obvious choice: the inverter and battery modules form a single tower and can be connected to the AC side, so installation can typically be completed in a single day. If the battery is added to the system later, AC connection requires the least amount of work. 

When is it advisable to choose a DC-connected energy storage system?

However, there are four situations where DC connection is the rational choice.

For greenfield projects. If the solar panels and the storage system are installed at the same time, a single hybrid inverter is cheaper and more efficient than two separate devices.

When the existing inverter needs to be replaced anyway. An 8- to 10-year-old string inverter is nearing the end of its lifespan; in this case, a hybrid inverter is no longer an additional cost but a planned replacement.

If the solar array is oversized. With a high DC/AC ratio, recovering the curtailed energy alone can pay for itself.

If space is limited. A single hybrid inverter means fewer devices, less cabling, and a single monitoring interface.

The decision is always based on an on-site assessment: the type and age of the existing inverter, along with the connection capacity, collectively determine which architecture is worthwhile.

Inverter Compatibility and System Design Considerations

Once the architecture has been selected, the success of the project is typically determined not by efficiency, but by whether the selected devices can work together.

Three factors must be coordinated: the voltage range, the communication protocol, and manufacturer approval. Most residential hybrid inverters operate with 48-volt LiFePO4 batteries via CAN or RS485 interfaces, whereas high-voltage platforms require their own battery stacks ranging from 100 to 1,000 V, often with a proprietary protocol.

A 10 kWh rating can refer to two completely different products: a 48-volt module and a high-voltage battery stack connected to a hybrid inverter. The two are not interchangeable.

Closed-loop communication is not an optional extra: without it, the inverter estimates the state of charge based solely on voltage, which is inaccurate given the flat characteristics of LiFePO4 batteries and may also be required for the manufacturer’s warranty.

Keeping the Existing Inverter or Replacing It with a Hybrid Inverter

To make this decision, four pieces of information must be gathered on-site.

The inverter’s age and remaining warranty period. For a 2–3-year-old unit, an AC-coupled system is almost always the better choice.

The string configuration. If you switch to a hybrid inverter, the voltage and current values of the existing strings must fall within the new unit’s MPPT window. This often requires rewiring or rearranging the strings.

The compatibility list. Always check the manufacturer’s current, approved list, including the specific model and hardware version, because firmware compatibility also matters. We’ve prepared a separate summary on FoxESS battery and inverter compatibility.

The cost of downtime. In a commercial installation, the days of lost production quickly outweigh the efficiency gains.

If any of these factors are uncertain, AC coupling is the lower-risk option: it doesn’t interfere with the solar side, and the approved settings of the existing system remain unchanged.

Power and Capacity Sizing for Retrofit Solar Panel and Battery Expansion

When sizing the system, two different figures must be considered separately: capacity (kWh) and power (kW).

Capacity is determined by the evening peak consumption, not by total daily consumption. If the evening load is 8–10 kWh per day, there is no point in installing 20 kWh: self-sufficiency improves only marginally once the ratio relative to the solar array exceeds 2:1.

Power output is limited by the C-rate and the inverter. A 20 kWh battery that can be charged at 0.5C delivers 10 kW at the battery level; the AC-side output is less than this.

Finally, calculate based on usable capacity, not nominal capacity: due to depth of discharge, efficiency, and temperature derating, the actual value may be 20–35 percent lower. This is particularly important for grid-tied expansions, because the battery inverter’s power is added to the system in parallel with the existing solar inverter.

Conversion Losses in AC- and DC-Connected Systems

The loss does not occur in the battery itself, but rather in the conversion stages and standby power consumption.

In an AC-connected system, the stored kWh passes through three inverter stages; in a DC-connected system, it passes through only one during solar charging. For a storage system handling 4,000 kWh per year, this amounts to a loss of 400 kWh at a 90 percent cycle efficiency and 160 kWh at 96 percent. The difference is around 240 kWh per year.

The second, often overlooked factor is standby power consumption. The no-load power consumption of a standalone battery inverter typically remains below 1 percent of its rated power, but even a continuous draw of 20–40 watts results in annual consumption of 175–350 kWh. With AC connection, two devices consume power in parallel; with DC connection, only one does.

Losses are not zero on the DC side either: the charge controller, the BMS, and the cabling all consume power, though on a smaller scale. In low-voltage, 48-volt systems, cable losses are more significant than in high-voltage stacks due to the higher current.

Therefore, it is not enough to simply compare the cycle efficiency listed in the catalog. Model the system based on actual production logs: how many kWh actually pass through the battery, and how many go directly to the loads. The latter costs the same in both architectures.

For zero-export and energy-management retrofit systems

Many grid permits require that feed-in be limited or completely prohibited. This regulatory requirement alone influences the choice of architecture.

In a DC-coupled system, the limitation is simple: the hybrid inverter draws back the MPPT power within a single control loop, exactly to the amount consumed by the loads and the battery.

The situation is more complicated in AC-connected retrofits because the battery inverter cannot directly curtail the existing solar inverter. This leaves two options: either absorb the excess power into the battery via variable-power charging, or regulate the solar side via communication or, possibly, frequency shifting. The latter is slower and highly brand-dependent.

This leads to a rule of thumb for design: AC-coupled zero export works reliably only if the charging power of the battery inverter reaches the rated power of the existing solar inverter.

The benefit from an energy management perspective is the reverse: any surplus that remains below the feed-in limit stays in the battery and is not lost.

In both cases, the feedback measured at the connection point—in the form of a current transformer or smart meter—closes the loop. During commissioning, check the direction of the current transformer, the phase sequence, and the sampling time, and leave a slight margin on the import side instead of setting it to zero.

AC or DC coupling for energy storage in larger commercial and industrial systems?

On a larger scale, the priorities shift: it’s not efficiency that matters most, but rather the revenue strategy and grid connection.

A single connection point is an advantage of a DC-coupled configuration. This means simpler grid analysis and fewer transformers; moreover, in the case of a heavily oversized solar array, recovering curtailed energy generates a large volume of power. Therefore, a DC-coupled energy storage system designed for a newly built solar power plant is now the standard.

Independence is a key advantage of an AC-connected configuration. An AC-connected energy storage system can provide grid services independently of solar generation, allows for modular expansion, and the failure of a single inverter does not take down the entire plant. This is the viable approach for existing industrial rooftop systems.

One argument against AC-connected systems is that they respond more slowly to rapid grid commands—such as frequency response—because multiple inverters must be coordinated simultaneously.

Be aware of two pitfalls. Retrofitting a DC-coupled system is difficult because the voltage range of new batteries typically does not match that of the old ones. For peak load shaving, size the PCS power output to match the peak to be shaved and the capacity to match the duration of the peak.

For larger projects, you’ll also find scalable, cabinet- and container-based solutions in the FoxEss energy storage product line; however, always base your sizing on the site’s quarter-hourly load profile.

SOLARKIT’s Recommendation: Decision-Making Criteria for Solar Panel Installers

In our experience, most flawed architectural decisions stem from a lack of assessment, not from a lack of technical knowledge. Therefore, we recommend noting down five pieces of information before submitting any proposal.

The exact model and year of manufacture of the inverter. This determines whether it can be retained.

The voltage and current values of the strings. This is a prerequisite for switching to a hybrid inverter.

The connection capacity and the feed-in limit. If modifying these requires a permit, the balance tilts toward an AC-connected system.

The need for a backup power supply. During a blackout, in an AC-connected system, the battery-powered inverter must operate in grid-forming mode, and the existing solar inverter must be able to work with it. If not, the solar side simply shuts down during a power outage.

The evening load profile. This determines the required capacity, not the total daily consumption.

Our conclusion is simple: for a new system, use DC connection; for an existing, well-functioning inverter, use AC connection. The exception is if the inverter needs to be replaced anyway, in which case the hybrid solution resolves the issue in one step.

As an installer, it’s worth keeping inventory ready for both architectures. Those who offer only hybrid inverter solutions will lose a large portion of the retrofit market, while those who offer only AC-connected storage will miss out on new construction projects.

FAQ: AC- and DC-Connected Energy Storage

Can a DC-connected energy storage system be installed with a microinverter system?

No. Microinverters already convert the generated power to AC behind the module, so there is no DC busbar to which the battery could be connected. With such a system, only AC-connected storage is an option; however, this is a natural solution: the storage system is simply connected to the AC-connected inverter.

Does an AC-connected expansion void the warranty on the existing inverter?

Typically not, because the AC-connected storage system is connected from the distribution panel, after the solar inverter, and does not interfere with its circuitry. However, when replacing the inverter with a hybrid model, the remaining warranty on the old unit is voided. Always request written confirmation from the manufacturer or distributor before installation.

Can AC and DC connections be combined in a single system?

Yes, in a well-designed system, the two can coexist. A typical scenario is when the existing solar inverter remains on the AC side, and a new hybrid inverter is added alongside it, complete with its own solar array and battery. In this case, an energy management system must coordinate the control.

How much more solar power is needed to charge an AC-connected battery?

Roughly 5–10 percent. Due to two conversion losses of around 95 percent each, charging a 20 kWh battery requires approximately 2 kWh more solar generation. This extra amount is negligible when midday peak generation would not be utilized anyway.

Does an AC-connected battery charge from solar panels during a power outage?

Only if the battery inverter operates in grid-forming mode to create a stable local grid, and the solar inverter is capable of working in conjunction with it. Some hybrid inverters raise the frequency to 52–53 Hz when the battery is full in order to disconnect the solar side. This behavior must be verified at the model level prior to installation.