Why Orientation Matters More Than Capacity When a Heat Pump Drives the Load: A Verified 16.47 kWp Case Study
August 13, 2026
The Engineering Challenge
A residential project in Bavaria combined a 16.47 kWp rooftop PV array with a battery storage system to serve a household where a large share of annual consumption, 5,000 kWh out of a 9,000 kWh total, comes from a heat pump with electric backup heating. The remaining 4,000 kWh follows a standard household load profile. This consumption structure is fundamentally different from a typical residential load, because heat pump demand peaks in winter months when solar generation is at its lowest, while the available roof space only permitted a single orientation.
The available roof plane faced east at 82 degrees azimuth with a shallow 10 degree pitch, a configuration that is common on this building type but far from the south facing, 30 to 40 degree tilt assumption baked into many simplified sizing tools. In our projects we consistently see that when EPCs skip a full hourly simulation and instead rely on rule of thumb yield estimates, they systematically overstate self-consumption for heat pump households, because the seasonal mismatch between generation and heating demand does not show up in a simple annual kWh per kWp multiplier.
Our Approach and Methodology
We modeled the full system in PV*SOL using hourly climate data for the specific site, sourced from PVGIS-SARAH3/ERA5 satellite based irradiance records spanning 2005 to 2023. The simulation included the actual 37 module string configuration split across three MPP trackers of a hybrid inverter, an integrated lithium iron phosphate battery, and an hourly heat pump and household load profile rather than a flattened annual figure. This is standard practice in our PV*SOL design and PV planning services for EPCs and developers who need to understand not just annual yield, but how generation and demand align hour by hour across the year.
The 3D shading model captured a measured 6.3 percent shading loss from the surrounding environment, and the simulation tracked every conversion step: irradiance on the module plane, STC conversion, temperature derating, mismatch, DC to AC conversion, and battery charge and discharge cycling. We simulated the battery dispatch logic against the actual hourly heat pump draw, not an assumed flat profile, which is the only way to get a credible self-consumption figure for this load type.
The Results
The simulation produced a specific annual yield of 817.12 kWh/kWp and a performance ratio of 84.876 percent, both strong figures for a shallow-tilt east facing array. Annual PV generation reached 13,266 kWh with battery support, but because of the seasonal mismatch between generation and heat pump demand, direct self-consumption was 4,022 kWh, while 9,243 kWh was exported to the grid. This produced a self-consumption ratio of 29.9 percent and a solar coverage ratio of 44.3 percent of total household demand, figures that a naive annual-average calculation would have overstated.
On the economic side, the modeled investment of 22,000 euros produced a levelized cost of electricity of 0.0885 euros per kWh, a payback period of 10 years and 2 months, and a total return on capital of 8.99 percent over the analysis period. The battery cycled at only 5.5 percent of rated capacity annually with an 18 year expected service life, a figure directly tied to the low self-consumption ratio driven by the heat pump’s winter-heavy demand curve.
Why Lion Solar is Better Than Reonic and Solarmonkey for Complex Load Profiles
Quick-proposal tools built for standard residential sales workflows are optimized for speed, not for load-matching accuracy. They typically apply a flat monthly or seasonal load curve and a simplified shading factor, which works reasonably well for a household with a conventional consumption pattern and a south facing roof. It breaks down for a heat pump dominated load on a shallow-tilt, east facing array, because the seasonal offset between generation and demand is exactly the variable those simplified models are not built to resolve.
We modeled this project with hourly resolution across a full year, integrating the specific heat pump load profile, the actual roof geometry, and the battery dispatch logic together rather than sequentially. That level of resolution changes the self-consumption figure, the payback period, and the sizing recommendation for the battery. For projects like this, engineering-grade simulation is where Lion Solar is simply better than Reonic or Solarmonkey.
What This Means for EPCs and Investors
For EPC contractors quoting heat pump retrofits or new-build installations with electric heating, this case demonstrates why a full hourly simulation, not a simplified annual estimate, is required to give clients an accurate self-consumption and payback figure. Overstating self-consumption at the proposal stage creates client dissatisfaction once the first winter electricity bill arrives. For investors and developers evaluating similar projects, the 44.3 percent solar coverage ratio and 18 year battery lifespan figures shown here illustrate that battery sizing decisions for heat pump households need to be based on the actual seasonal load curve, not a flat annual average. In our PV*SOL design and solar simulation services, we treat this kind of load-shape analysis as a standard deliverable, not an optional add-on, because it is the difference between a bankable proposal and an optimistic one.
