A cordless window cleaning robot in 2026 realistically runs 20 to 40 minutes per charge on exterior glass, not the 90 minutes marketing pages quote. The gap comes from vacuum duty cycle, and it decides how large a skylight you can finish in one pass.
Cordless is the feature buyers ask for first and regret second. It solves a real problem, the cable snagging on roof penetrations, and it creates a different one: every watt spent holding the machine to the glass is a watt not spent cleaning.
Why do cordless run times fall short in 2026?
Because the vacuum pump runs continuously. On a corded unit that load is external. On a battery unit it competes with the drive motors and the brush, and the pump draws the biggest share. A 5,200 mAh pack quoted at 90 minutes is usually measured in eco mode, on vertical glass, at 25°C, with no brush running.
Add exterior conditions and the number collapses. Sun-warmed glass at 55°C heats the pack; the pump works harder to hold a chamber whose seal is softening; the brush motor pulls extra current against dried-on film. That is how you get 22 minutes from a 90-minute claim.
How many square metres can one charge clean?
Depends on the path. A simple zigzag on a clean flat pane covers maybe 30 to 45 m² in half an hour. A skylight with frames, mullions and edge-detection retries drops to 15 to 25 m². Our own test runs on pitched frames land near 18 m² per 30-minute charge, which is the honest planning number for a 1.2 m × 1.5 m skylight bay.
Practical translation for a facility: if your atrium is 200 m² of glass, budget eight to eleven battery cycles, which means two packs if you want to avoid idle time. That single line item changes the quote more than choosing between brands.
Scenario Quoted run time Realistic run time Area per charge Indoor vertical glass, eco 90 min 70–85 min 60–80 m² Indoor vertical, standard 90 min 45–60 min 35–50 m² Exterior pitched, 25°C 90 min 30–40 min 20–28 m² Exterior pitched, 45°C+ 90 min 18–25 min 12–18 m²
Is a corded robot better for skylights?
Usually, yes, and this is where most buying advice gets it backwards. On a fixed skylight, the cable path is knowable in advance. You can route it along a frame, tie it off, and it never moves. The cordless advantage mostly applies to glass with no clean cable route, such as a large open atrium where the unit crosses far from any edge.
Corded also means lighter packs or no packs at all, which cuts weight. On a 30° slope, 600 g matters more than an hour of runtime, because it is the mass the seal has to hold against gravity at the worst moment.
One more consideration in 2026: swappable packs are now standard in the mid tier, so the cordless argument has shifted. It is no longer "cordless or corded" but "how many packs and how fast do they swap." Ask for the swap time. A 3-second hot swap is fine. A 90-second swap with a screw means you lose the advantage entirely.
What breaks first on a cordless unit?
The connector, then the pack. Pogo pins and spring contacts corrode on roofs, and a corroded contact reads as a dead battery. Keep a dielectric grease routine and inspect the contacts monthly. Pack degradation shows as shorter runs at similar temperatures; once capacity falls below 70 percent, the pump starts sagging on slopes and holding force drops before the warning beep.
If you want comparative runtime data rather than spec-sheet claims, our runtime reality check breaks it down pass by pass, and the reviews archive includes the cordless models we tested on the same pitched frame, among them the WINBOT mini 2 review.
Who should buy cordless in 2026?
Owners of one or two skylights with no cable route and a technician who can be on the roof for a two-minute pack swap. Anyone cleaning more than 100 m² in a session, or working above people where a mid-pass stop is a risk, should take corded and route it properly. The freedom is real. It is just smaller than the brochure suggests.
Cabinets and service corridors change the answer again. A machine stored at roof level starts warm, so the first ten minutes of a summer run are the most wasteful. Store packs in a conditioned room and carry them up, or accept that the first pass of the day costs you a third of a charge.
Charging discipline is the other half. Fast chargers get a pack to 80 percent quickly and then trickle; running a partial charge up to full every time shortens life. If your pattern is one pass in the morning and one in the afternoon, a pair of packs and a standard charger will outlast a single pack on a fast charger by a wide margin.
Compare this with the older rail-fed facade units, which never had a runtime problem because power arrived through the cable. That is worth remembering when a vendor’s headline feature is the runtime figure. In most buildings the constraint is not energy, it is whether someone can reach the glass at all.
Key Takeaways
- Expect 20–40 minutes of exterior runtime from a 90-minute cordless claim in 2026.
- Plan on 18–25 m² per charge for framed pitched glass, not 60 m².
- Corded often wins on skylights because the cable route is fixed and knowable.
- Weight matters more than runtime on slopes above 25°.
- Check battery contact corrosion monthly; it reads as a failed pack.