Beyond the Rankings: What HTW Berlin and aquu's Energy Storage Inspection 2026 Test Really Shows

<p>HTW Berlin and aquu's Energy Storage Inspection 2026 tested 12 residential PV-battery systems involving 10 manufacturers. Eight manufacturers participated publicly under their brand names, while two additional DC-coupled systems were independently purchased by the testing team and presented anonymously. Beyond comparing overall system efficiency, the report also breaks down usable capacity, conversion efficiency, battery efficiency, dynamic control, standby consumption, and other key aspects.</p><p>Looking at the final results, SAX Power ranked first in the 5 kW class, while FOX ESS set a new record for the 10 kW reference case with an SPI of 97.0%. However, the real value of the report goes beyond simply producing a ranking. It raises a more important question: <span style="color:rgb(46, 161, 33)"><strong>Why do residential storage systems with seemingly similar specifications still show meaningful differences in real operating performance? </strong></span>To answer that question, we first need to start with the ranking itself.</p><p><img src="https://energyplanet.oss-cn-shenzhen.aliyuncs.com/uploads/images/20260903160708_81f8b32b.png" alt="" loading="lazy" decoding="async" /></p><h2>5 kW and 10 kW: Who Leads in 2026?</h2><p>HTW evaluates systems using two different reference cases: <span style="color:rgb(46, 161, 33)"><strong>SPI (5 kW) and SPI (10 kW)</strong></span>. The 5 kW class represents a smaller residential PV and household-load configuration, while the 10 kW class reflects a larger PV system with additional loads such as a heat pump and an electric vehicle. Therefore, the two SPI values should not be compared directly; the meaningful comparison is within each class.</p><p>In the 5 kW class, SAX Power Home Plus ranked first with an SPI of <span style="color:rgb(46, 161, 33)"><strong>93.2%</strong></span>. SMA Sunny Boy Smart Energy 5.0 paired with Home Storage 6.5 followed closely at <span style="color:rgb(46, 161, 33)"><strong>92.8%</strong></span>, while the DC-coupled KOSTAL PLENTICORE MP G3 M 4.6 paired with BYD Battery-Box HVS+ 7.7 ranked third at 92.7%. Only 0.5 percentage points separated the top three systems, showing that the leading products in this class are already performing at a very similar overall level.</p><p><img src="https://energyplanet.oss-cn-shenzhen.aliyuncs.com/uploads/images/20260903160708_2fc4d36f.png" alt="" loading="lazy" decoding="async" /></p><p>The gap was more pronounced in the 10 kW class. FOX ESS PQ-H3-Ultra-10.0 paired with EQ3300-5 ranked first with an SPI of <span style="color:rgb(46, 161, 33)"><strong>97.0%</strong></span>, which the report identifies as a new record for this reference case. RCT Power ranked second at 96.4%, while Energy Depot, Fronius, and the DC-coupled KOSTAL + BYD system all remained above 95%. By comparison, the two anonymously purchased systems achieved only <span style="color:rgb(46, 161, 33)"><strong>91.9%</strong></span> and <span style="color:rgb(46, 161, 33)"><strong>89.3%</strong></span>, corresponding to Efficiency Classes D and G.</p><p><img src="https://energyplanet.oss-cn-shenzhen.aliyuncs.com/uploads/images/20260903160705_18953de4.png" alt="" loading="lazy" decoding="async" /></p><p>Overall, 10 of the 12 tested systems achieved Efficiency Class A or B, indicating that mainstream residential storage systems have already reached a relatively high level of overall efficiency. But this also raises another question: If most systems are already "highly efficient," why can SPI still range from 97.0% down to 89.3%? The answer lies in the different types of losses within the system.</p><h2>Where Do the Differences Between Leading Systems Come From?</h2><p>The value of SPI is that it does not focus on a single peak conversion-efficiency figure. Instead, it incorporates multiple sources of system loss. HTW divides these into <span style="color:rgb(46, 161, 33)"><strong>Sizing losses, Conversion losses, Control losses, and Standby losses</strong></span>, making it possible to see more clearly where energy is actually being lost.</p><p>In the 10 kW class, FOX ESS had total losses of just <span style="color:rgb(46, 161, 33)"><strong>3.0%</strong></span>, while the weakest anonymous system reached <span style="color:rgb(46, 161, 33)"><strong>10.7%</strong></span>. For most products, conversion losses remained the largest contributor, but they were not the only factor. How accurately and quickly the control system follows changing loads, as well as how much energy the system consumes at low power or in standby mode, also accumulates into the final SPI result.</p><p><img src="https://energyplanet.oss-cn-shenzhen.aliyuncs.com/uploads/images/20260903160712_9fb24a22.png" alt="" loading="lazy" decoding="async" /></p><p>This explains why looking only at inverter efficiency or battery capacity can easily lead to an incomplete conclusion. Even if a product has very high conversion efficiency under ideal conditions, its annual performance in a real home can still be dragged down if efficiency falls sharply at low load or if additional losses continue to accumulate during control and standby operation.</p><p>That leads to another test in the report that deserves closer attention: <span style="color:rgb(46, 161, 33)"><strong>partial-load efficiency.</strong></span></p><h2>Why Partial-Load Efficiency Matters</h2><p>Residential storage systems do not operate continuously at 5 kW or 10 kW. In reality, refrigerators, lighting, standby appliances, and basic household loads often add up to only a few hundred watts. As a result, efficiency in the low-power range may be more representative of what users actually experience day to day than peak efficiency.</p><p>The report shows that at 200 W discharge, the highest tested efficiency reached <span style="color:rgb(46, 161, 33)"><strong>92.1%</strong></span>, while the less-efficient reference inverter achieved only <span style="color:rgb(46, 161, 33)"><strong>70.7%</strong></span>. At 100 W, the gap widened further: RCT Power reached <span style="color:rgb(46, 161, 33)"><strong>86.1%</strong></span>, while the lowest reference system dropped to just <span style="color:rgb(46, 161, 33)"><strong>53.9%</strong></span>.</p><p><img src="https://energyplanet.oss-cn-shenzhen.aliyuncs.com/uploads/images/20260903160714_5939d440.png" alt="" loading="lazy" decoding="async" /></p><p>This result matters because it shows that while leading systems may perform very similarly at rated or higher power levels, the gap can widen again quickly once they move into the low-load range that is more common in residential use. For home-storage users, the more useful question in the future may no longer be simply "What is the maximum efficiency?" but rather "How efficient is the system at the power levels my home actually uses most of the time?"</p><p>However, partial-load efficiency only explains how much energy is lost. It does not yet answer another key question: <span style="color:rgb(46, 161, 33)"><strong>Can the system respond quickly enough when household demand keeps changing?</strong></span></p><h2>Response Speed and Standby Performance Also Start to Matter</h2><p>Residential loads are not static. Heat pumps switch on, EV charging changes, and even ordinary household appliances continuously alter the power demand. A storage system therefore needs not only to be efficient, but also to adjust its output quickly enough.</p><p>In the dynamic control tests, Energy Depot and RCT Power both recorded a settling time of just 0.2 seconds, the fastest result in the report. The two anonymous systems, by contrast, required 13.7 seconds and 10.9 seconds respectively.</p><p><img src="https://energyplanet.oss-cn-shenzhen.aliyuncs.com/uploads/images/20260903160716_0d823d37.png" alt="" loading="lazy" decoding="async" /></p><p>This difference means that when household demand changes suddenly, a faster system can adjust battery power more quickly, while a slower system may temporarily import additional electricity from the grid or feed more power into it. In other words, system performance depends not only on how much energy is lost during each conversion, but also on whether the system can respond correctly at the right time.</p><p>The same logic applies to standby consumption. The report shows that some high-performing systems can keep standby power consumption to only a few watts, while certain systems consume noticeably more when the battery is discharged. These losses may not appear significant in a single moment, but for a residential storage system that spends many hours each year at low load or in standby mode, they eventually accumulate into real operating costs.</p><p>By this point, the report has moved from "Who ranks first?" to "Why do different systems behave differently in real operation?" But for users, there is still one more practical question: <span style="color:rgb(46, 161, 33)"><strong>What are these efficiency differences actually worth?</strong></span></p><h2>Dynamic Tariffs Make Efficiency More Important</h2><p>The second half of the report goes on to examine grid charging under dynamic electricity tariffs. The basic logic is straightforward: charge the battery from the grid when electricity prices are low, then use the stored energy to supply the home when prices are higher, reducing electricity costs through the price spread.</p><p>But this model only works under one condition: the price spread must be large enough to cover the system losses generated during charging, storage, and discharging.</p><p>HTW provides a clear example. In a more efficient system, grid electricity can achieve an overall efficiency of 78% after going through the complete charge-storage-discharge process, while a less-efficient system reaches only 63%. If the system is also operating at low power, the decline in partial-load efficiency can further reduce the economic benefit.</p><p><img src="https://energyplanet.oss-cn-shenzhen.aliyuncs.com/uploads/images/20260903160719_2a736d6c.png" alt="" loading="lazy" decoding="async" /></p><p>As dynamic tariffs, smart meters, and intelligent EMS become more common across Europe, efficiency is no longer just a laboratory performance metric. It is increasingly likely to determine how much real value a storage system can capture from electricity-market price fluctuations.</p><p>This also suggests that the competitive logic of residential storage may change. Simply increasing battery capacity will not guarantee better economics. The system must also be able to charge and discharge with lower losses, while responding quickly to household demand and electricity-price signals.</p><h2>What Do the 2026 Test Results Tell Us?</h2><p>On the surface, Energy Storage Inspection 2026 is an efficiency ranking. But the logic connecting the report is actually quite clear: <span style="color:rgb(46, 161, 33)"><strong>the ranking is the outcome, system losses explain the outcome, partial-load and dynamic-control tests reveal the differences in real-world operation, and dynamic electricity pricing ultimately turns those technical differences into economic differences.</strong></span></p><p>SAX Power and FOX ESS took first place in the 5 kW and 10 kW classes respectively. But as peak efficiency among mainstream products continues to converge, the factors that may increasingly differentiate products are partial-load efficiency, control response, standby consumption, and how well the entire system performs together under real residential conditions.</p><p>For consumers, installers, and distributors, this means that comparing residential storage systems may increasingly require moving beyond a single nominal capacity or maximum-efficiency figure toward a broader question: <span style="color:rgb(46, 161, 33)"><strong>How much of the energy handled by this system can actually be turned into real value under real household conditions and real electricity prices?</strong></span></p><p><a href="https://solar.htw-berlin.de/wp-content/uploads/HTW-aquu-PPT-Energy-Storage-Inspection-2026.pdf" target="_blank" rel="noopener noreferrer"><span style="font-size:12px"><em>Source: HTW Berlin &amp; aquu, Energy Storage Inspection 2026, Version 1.0, March 2026.</em></span></a></p>