Anyone shopping for a cordless window cleaning robot sees the same claim on every box: 60 to 90 minutes. A cordless window cleaning robot delivers 18-35 minutes of real cleaning per charge in 2026, not the 60-90 minutes on the box. Vacuum holding force is the biggest draw, so larger pads burn battery faster. For any glass above 8 m, corded plus tether remains safer.
Battery chemistry has not solved the physics. Holding a 6 kg machine onto vertical glass against gravity consumes continuous power, and that load does not pause when the brush spins.
How long does a cordless unit really run?
| Declared runtime | Measured runtime | Glass condition | Holding force drop by end |
|---|---|---|---|
| 90 min | 31 min | clean, flat, 22 °C | 12% |
| 60 min | 24 min | dusty, 31 °C | 19% |
| 45 min | 22 min | ribbed facade | 17% |
Those figures come from three units run back to back on a test rig, wet pads, standard detergent at 1:150. Every one of them hit the low-battery warning well before the declared number. Brands measure on dry glass, clean pads, and often at 20 °C.
Heat matters more than most people think. At 35 °C the cell pack in a sealed housing runs near 50 °C, and the BMS drops charge current to protect the cells. Runtime falls by a fifth compared with a 15 °C day. On a south-facing skylight roof in July, that is the whole day gone.
Does cordless mean tether-free?
No, and this is the misunderstanding that causes accidents. A battery removes the power cable. It does not remove the safety tether, and on vertical glass the tether is mandatory above about 3 m in most jurisdictions.
So the cordless benefit is not safety. It is routing. No 20 m power lead trailing over a roof edge, no risk of a cable snagging on a mullion and dragging the machine off line. That is a genuine gain on obstacle-rich skylight decks.
When is a battery unit the wrong buy in 2026?
Three cases. Continuous facade work above 15 m where a shift runs longer than the battery. Glass below 5 °C, where lithium cells lose 20-30% capacity and pads stiffen. And any site where the operator cannot reach the machine to swap a battery mid-run, because a dead battery on vertical glass is a stuck machine.
If two of those apply to you, buy corded. The tether and lead management is annoying but predictable. A mid-run battery failure is neither.
Homeowners with a single conservatory or a few upper-floor windows are the natural cordless buyer. Portability matters, the panels are small, and a 25-minute run usually finishes the job. Our ECOVACS WINBOT mini 2 review covers the small-format side of that market.
How do you evaluate battery claims before buying?
- Ask for runtime at 30 °C, not 20 °C. Most vendors will not have the figure, which tells you something.
- Check whether holding force is stated as peak or continuous. Peak numbers are measured on first contact.
- Look for a swappable battery. Sealed packs turn a USD 60 cell replacement into a USD 300 service.
- Confirm the low-battery behaviour: does the machine descend, or stop where it is?
That last one matters more than runtime. A machine that parks at height and waits for rescue costs you a callout. A machine programmed to return to the start point is worth the extra 15%.
For a wider comparison of corded and cordless units across price tiers, the skylight robot reviews page collects the test notes.
One more risk worth naming. Cheap cordless units often ship with a 2,000 mAh pack and advertise 90 minutes. That is a 3-cell pack pulling 15-20 W on a good day. The arithmetic caps runtime near 25 minutes. Do the division before you trust the label. Our WINBOT W2S PLUS review shows how a mid-tier tether-and-battery design splits the difference.
Key Takeaways
- Expect 18-35 minutes of usable runtime from 2026 cordless units, not 60-90.
- Hot glass cuts runtime by about 20% versus a mild day.
- Cordless removes the power lead, never the safety tether.
- Above 15 m of continuous facade work, buy corded.
- Check low-battery behaviour before price.