A single-junction cell can never convert more than about 33.7% of sunlight to electricity — the Shockley-Queisser limit — and real panels fall further short of even that. Set the panel area, tilt angle, cell efficiency and inverter efficiency, then place the panel anywhere on Earth with a latitude and day of year, and watch the power track the sun's real elevation and angle of incidence across a simulated day, with a loss bar showing exactly where the rest of the sunlight goes. Feed that day-only output into a battery — capacity, load and round-trip efficiency — and watch the state of charge climb by day and ride out the night. It all runs on your device.
You are in the Physics lab.
Sunlight in, electricity out — and the losses along the way.
Charging by day, discharging by night, against a steady load.
What the sky view, the irradiance curve and the loss bar are telling you — then what the battery does with the result.
The sun arcs across the sky as the day plays, and the gauge beside the panel fills toward the rated power — the panel's output under standard test conditions. Before sunrise and after sunset the sun sits below the horizon and the gauge sits at zero, exactly where it should.
The irradiance curve plots the plane-of-array power across the day: zero in darkness, rising and falling with the sun's changing angle of incidence on the tilted panel. The loss bar breaks that same instant down into four slices of the incoming sunlight: the share no single-junction cell can ever reach (Shockley-Queisser), the further shortfall between your real cell efficiency and that ceiling, what the inverter loses, and what's left as net electrical capture — the smallest slice, and the only one you actually get.
The Storage card feeds a simulated day of sunlight — built from your latitude, day of year and tilt angle — through the same panel into a battery serving a constant load. The state-of-charge trace climbs while generation beats the load and falls through the night; if it would fall below zero that stretch counts against unmet-load time. Autonomy is simply capacity divided by load, and capacity factor compares the energy actually delivered over the day to what the panel would deliver running at its rated power the whole time.
A hard ceiling on conversion, a curve shaped by real sun geometry, and a battery to bridge the night.
Sunlight arrives as a spectrum of photon energies, but a single-junction cell has one band-gap energy: photons below it aren't absorbed at all, and photons above it give up their excess as heat the instant they free an electron. Balancing those two losses against each other, the best any single-junction silicon cell can do under standard sunlight is about 33.7% — a limit set by the physics of the junction itself, not by manufacturing.
A panel's power scales with the cosine of the angle between its face and the sun's rays, and that angle is set by latitude, tilt and the sun's declination, which shifts with the day of year. Electrical output is irradiance · area · min(cell efficiency, 33.7%) · inverter efficiency, and it falls to zero whenever the sun's elevation drops to or below the horizon.
Because a panel only generates in daylight, a battery is what actually lets a constant load be served overnight: it charges on the daytime surplus and discharges through the dark, at a round-trip efficiency below 100%. Its capacity sets how many hours of autonomy it buys against the load, and over a full day the capacity factor — actual energy over rated-power energy — tells you what fraction of the panel's nameplate rating the sun really delivered.
Sunlight frees electrons in the panel's silicon cells (the photovoltaic effect), producing direct current; an inverter turns it into AC for your loads, a battery banks the daytime surplus, and any excess flows to the grid. How much a fixed panel captures is pure geometry — it comes down to three angles.
Learn more: our data study on optimal tilt vs latitude · Solar power (Wikipedia)