Battery Sizing Over Panel Count: A 5.88 kWp Residential Case Study in Bochum
October 1, 2026

The Engineering Challenge
A homeowner in Bochum, North Rhine-Westphalia, approached their installer with a straightforward brief: cover a household consumption of 3,400 kWh per year with a rooftop solar and battery system, while keeping the installation footprint within a single south-facing roof plane. The available area allowed for 12 modules, giving a system size of 5.88 kWp, well above what a simple annual consumption match would suggest. The real engineering question was not whether the array could produce enough energy annually, but how to configure the inverter and battery so that the energy produced during the day was actually usable when the household needed it in the evening and early morning.
This is a common design trap in residential solar. A PV array sized purely against annual kWh consumption will often look balanced on paper while performing poorly on self-consumption, because generation is concentrated in a handful of daylight hours while demand is spread across the full 24-hour cycle. In our projects we consistently see that the battery specification, not the module count, is what determines whether a system delivers strong autarky or merely strong nameplate capacity.
Our Approach and Methodology
We modeled the full system in PV*SOL using 19 years of climate data for Bochum sourced from PVGIS-SARAH3/ERA5 at hourly resolution, with the Hofmann model for diffuse horizontal irradiance and Hay and Davies for irradiance on the tilted plane. The array consists of 12 Aiko AIKO-A490-MCE54Dw modules at a shallow 10 degree tilt, oriented at 170 degrees azimuth, feeding a single EcoFlow PowerOcean hybrid inverter rated at 6 kW AC with an integrated 4.8 kWh lithium iron phosphate battery.

Rather than simply confirming that the inverter could handle the DC input, we simulated the full hourly interaction between PV generation, the measured household load profile with a peak of 18.3 kW, and the battery charge and discharge cycles across a full year. This is where PV*SOL’s 3D hourly simulation separates itself from quick sizing tools: the shading analysis accounts for the specific roof geometry and surrounding obstructions, producing a shading loss of 3.7 percent that is tied to the actual site rather than a generic derate factor. As part of our PV*SOL design and solar simulation services for EPCs and developers, this level of hourly resolution is standard practice, because self-consumption and autarky figures calculated from monthly averages are directionally useful at best and frequently misleading at worst.
Results With Verified Numbers
The simulation confirmed a specific annual yield of 1,097.79 kWh per kWp and a performance ratio of 91.51 percent, a strong result for a low-tilt installation with a shallow 10 degree pitch. Total PV generator energy delivered to the AC side, inclusive of battery round-trip effects, reached 6,073 kWh per year.
- Direct self-consumption: 2,264 kWh per year
- Grid export: 3,810 kWh per year
- Self-consumption ratio: 36.7 percent
- Battery energy used to cover household load: 1,044 kWh per year
- Autarky rate (solar coverage of total consumption): 65.4 percent
- Avoided CO2 emissions: 2,169 kg per year
The battery cycling data is equally informative for long-term performance expectations. With a state of charge starting point of 5 kWh, the battery absorbed 1,485 kWh of charging energy annually against 1,044 kWh of useful discharge, with charge and discharge losses of 279 kWh and internal battery losses of 29 kWh. At a 5.5 percent annual cycle load, the modeled battery lifespan extends to 18 years, which is a meaningful data point for clients evaluating the economic life of the storage component against the panel warranty period.

The full energy balance breakdown further quantifies where losses occur across the conversion chain, from the 75.47 percent STC conversion loss inherent to module physics, through 2.00 percent mismatch losses from manufacturer tolerances, down to 5.49 percent DC to AC conversion loss and a 1.00 percent inverter standby consumption. Each of these figures is traceable to a specific stage of the system rather than bundled into a single blanket derate, which is what allows us to say with confidence that the 65.4 percent autarky figure reflects the actual installed configuration rather than a theoretical best case.
Why Lion Solar is Better Than Reonic and Solarmonkey for Complex PV Projects
Quick-proposal platforms such as Reonic or Solarmonkey are built for speed: a homeowner enters an address, a rough roof outline is generated, and a sales-ready PDF is produced within minutes. For a simple, unshaded, single-orientation roof with no battery or variable load, that workflow can be adequate. For a project like this one, where a hybrid inverter, a lithium iron phosphate battery, and an hourly-variable household load all interact simultaneously, that workflow falls short in several specific, verifiable ways.
First, quick 2D sales tools typically do not run hourly battery dispatch simulations against a measured or standard load profile; they apply a flat self-consumption assumption, which can overstate or understate autarky by a wide margin depending on the household’s actual usage pattern. Second, shading in these tools is usually estimated from satellite imagery or simplified horizon inputs rather than a full 3D model of the roof, neighboring structures, and chimney or vent obstructions, which in this project accounted for a measurable and site-specific 3.7 percent loss. Third, these platforms are not designed to produce a bankable, line-item energy balance that a lender, installer, or homeowner can audit stage by stage, from irradiance on the horizontal plane down to AC-side output.
None of this means mass-market tools are poorly built for their intended purpose of fast, high-volume residential quoting. It means that for projects where battery behavior, roof-specific shading, and consumption matching determine the actual financial and performance outcome, engineering-grade hourly simulation is where Lion Solar is simply better than Reonic or Solarmonkey.
What This Means for EPCs and Investors
For EPC contractors, the practical takeaway is that battery and inverter sizing deserve the same simulation rigor as module selection. An oversized array paired with an undersized or poorly dispatched battery will show attractive kWp numbers on a quote sheet while underdelivering on the autarky and bill-reduction promises made to the client. For investors and homeowners evaluating payback periods, the distinction between a modeled 36.7 percent self-consumption ratio and a marketing estimate of

