Almost every Earth-orbiting satellite turns sunlight into electricity with solar panels and stores some of it in batteries for when it is in Earth's shadow.
There are no power sockets in orbit. A satellite must make its own electricity, and for satellites around Earth that means sunlight.
Solar cells on large panels, or wings, turn sunlight directly into electric current. That current runs the computer, radios, heaters and payload, and charges a battery for the times the satellite passes through Earth's shadow.
Above the atmosphere, sunlight delivers about 1,361 watts on every square metre facing the Sun, according to NASA's Earth fact sheet. That is more than at the ground, where air and clouds take their share.
Space solar cells catch only part of it. NASA's small-spacecraft review gives modern multi-junction space cells a typical efficiency of about 30 percent, with the best near 34 percent. Cheaper single-junction silicon cells are around 20 percent.
Worked example: one square metre of 30 percent cells facing the Sun gives about 1,361 × 0.30 ≈ 408 W. NVS-01's 2.4 kW therefore needs at least about 6 m² of cells in full sunlight. Real panels are bigger, to allow for gaps between cells, imperfect angles and ageing.
The International Space Station is in a different league: NASA lists eight solar arrays producing 75 to 90 kilowatts.
Solar cells slowly lose output in orbit. NASA lists radiation, ageing and contamination, such as darkening of the protective cover glass, as the main causes. Designers size a panel so that it still gives enough power at the end of the mission, not just on day one.
A satellite also has to keep its panels facing the Sun. When ISRO's Mars Orbiter Mission ran out of propellant, it could no longer hold the pointing needed to keep generating power, and ISRO declared it at end of life in 2022.
Read the next-pass list: a pass only counts as visible when the station is sunlit — the same sunlight that feeds its eight solar arrays. Open the ISS tracker
The electrical power from a flat array is
P = S × A × η × cos θ
where S is the solar irradiance (about 1,361 W/m² near Earth, NSSDCA), A is the cell area, η the cell efficiency and θ the angle between the Sun and the panel's normal.
With η = 0.30 (NASA's typical multi-junction value) and the Sun square on (cos θ = 1):
P/A = 1,361 × 0.30 ≈ 408 W per m² of cells.
So NVS-01's 2.4 kW needs at least 2,400 ÷ 408 ≈ 5.9 m² of active cells, and GSAT-19's 4,500 W at least 4,500 ÷ 408 ≈ 11 m². Real arrays are larger, because cells do not cover every square centimetre, wiring loses some power, and the design must hold at end of life after radiation damage.
The cos θ term is why attitude matters. Tilt the array 30° away from the Sun and output falls to cos 30° ≈ 0.87 of the maximum; at 60° it halves.
Power must also cover eclipse. If the load is P_load and the eclipse lasts t_e in an orbit of period T, then the sunlit array must supply roughly
P_array ≈ P_load × [1 + t_e ÷ (T − t_e)] ÷ (charging efficiency)
so that the battery is refilled before the next shadow. Batteries are covered on the next page.
Sunlight in orbit is strong and steady, about 1,361 W per square metre, and solar cells turn it into electricity with no fuel to run out.
Radiation, ageing and contamination reduce their output over the years, so the array must still produce enough power at the end of the mission.
With its propellant likely exhausted, the spacecraft could not hold the pointing needed for steady power generation, and ISRO declared it non-recoverable in 2022.