Luke Bell set an unofficial endurance record for a solar-powered multirotor drone last month: 5 hours, 2 minutes, and 21 seconds aloft over South Africa. New Atlas reported the flight, which Bell filmed and published on his YouTube channel. His father Mike was there. The drone came down because Luke got tired, not because the power ran out.
It is a genuinely impressive result from a garage build. It is also not what the headlines say it is.
Bell, a South African engineer and YouTuber better known for building the world's fastest battery-powered racing drones — his Peregreen 4 hit 408 mph in January 2026 — spent two iterations getting the solar endurance record right. The first version flew on pure solar, no battery, no capacitor. According to SlashGear, which covered the V1 build, 27 panels in series produced around 150 watts on the ground. It worked until a wind gust pitched it into the grass after three minutes. The panels couldn't supply enough instantaneous current to handle the power surge from fighting turbulence. Pure solar, no buffer, no margin.
The second version is what set the record. Bell added 28 panels producing over 110 watts under full sun, trimmed 70 grams from the airframe, and introduced a battery backup: a diode circuit that routes excess panel output to a small auxiliary battery when generation exceeds demand, and draws from that battery when a cloud edge or wind gust spikes consumption above what the array can deliver. The drone needs roughly 70 watts to hover. The array makes over 110. The gap is the margin, and the battery is what makes it survivable when the sky isn't cooperating.
According to New Atlas, which covered the build, Bell estimates his panels run at roughly 20-25% efficiency — typical for mass-market silicon. He told the outlet he believes indefinite flight is theoretically possible by converting the design to a winged eVTOL configuration, which would reduce hover power requirement to around 10% of current. He has not published a timeline or a detailed engineering plan for V3.
The physics Bell is fighting are real, and they don't improve linearly. A fixed-wing solar aircraft can glide when power falls short. A multirotor cannot — it must maintain rotational lift continuously, which means a power failure at any altitude is a controlled descent at best. The multirotor solar endurance problem is categorically harder than the fixed-wing version, which is why the relevant comparison points look nothing like Bell's build.
Airbus's Zephyr S, a high-altitude pseudo-satellite, holds the current solar endurance record at 67 days, 6 hours, and 52 minutes as of April 28, 2025, flying at above 60,000 feet where sunlight is unobstructed and winds are predictable. BAE Systems' PHASA-35, a competing fixed-wing solar HAPS, has logged 24 hours of actual flight time in December 2024 with a simulated 72-hour endurance capability — no multi-day endurance record exists in any public source. Both are fixed-wing, purpose-built, and flying above weather. Bell's drone operates at low altitude in South Africa's wind corridor, using hobby-grade components. The efficiency gap between 5 hours and 67 days 6 hours is not a technology roadmap — it is a physics class distinction between rotary lift and aerodynamic lift at altitude.
The battery backup is what makes the 5-hour record possible, not the solar panels alone. Strip the battery from Bell's math and V1's three-minute crash is the result. The panels are load-following with a buffer, not the direct drive that "solar drone flies indefinitely" framing implies. This is not a knock on Bell's work — it is an accurate description of how the system actually functions. The record is real. The mechanism behind it is battery-buffered photovoltaic generation, not panel-only flight.
What Bell is proposing with V3 — eliminating the battery entirely — would require either a significant increase in panel efficiency, a drastic reduction in hover power demand through aerodynamic redesign, or both. He has not specified a target panel technology or a date. NotebookCheck's coverage of the build characterizes V3 plans as Bell "looking to get rid of the battery with some clever engineering" — which is an honest description of intent, not a plan.
The multirotor solar drone space is not empty — there are active programs from defense contractors and aerospace incumbents pursuing the same problem. But the multirotor constraint creates a different engineering problem than the fixed-wing version, and Bell is working it from a different starting point: a garage in South Africa with hobby motors, YouTube documentation, and no procurement contract with a defense ministry. His 5-hour result is the current ceiling for electric multirotor solar endurance, unofficial or otherwise. That ceiling sits about 300 times below where Zephyr S already is, and the physics that make fixed-wing solar easier to scale don't help a design that needs continuous rotational lift.
The record is worth noting. The distance from record to operational system is measured in orders of magnitude, not iterations.
