Expect 300-600 m2/h from a good skylight cleaning robot in ideal conditions, dropping to 250-350 m2/h with obstacles. A typical 2-3 kWh battery gives 3-5 hours of cleaning, and a 25-40 litre tank covers roughly 800-1,200 m2 before a refill.
Vendor throughput numbers are measured on clean, open, rectangular glass with a fresh brush. Your roof is not that. Here is how to translate the brochure into a realistic shift plan.
Why does real throughput fall short of the spec?
Every obstacle costs time. A drainage channel, a mullion line or a rooftop vent forces a slow-down or a repositioning pause. On a cluttered industrial sawtooth roof, real throughput can drop 40% below spec. On a clean mall atrium, you may come within 10%.
Two squeegee passes instead of one halves throughput again. If streak-free finish matters, price that in from the start.
Should you swap batteries or recharge?
Recharging mid-shift is a productivity killer because you wait two hours for a 45-minute top-up. For anything beyond three hours of cleaning, carry a second battery. Swap time on a well-designed unit is under five minutes, and the clean continues. A spare battery costs $400-$900 and pays back quickly on large roofs.
- Under 1,500 m2: single battery is usually enough.
- 1,500-3,000 m2: buy a spare battery.
- Over 3,000 m2: two spares and a charging plan.
How far does a tank of water go?
Water use depends on surface dust. Light dust might need 25-35 ml/m2; a post-construction roof with cement film can burn double that. A 30 L tank covers about 850-1,200 m2 on light dirt, or 400-600 m2 on heavy dirt. On big roofs, plan a refill point and a hose run, or you lose time hauling.
Waterless and low-water modes exist for dry climates, but they trade speed for finish. Expect more passes and more brush wear.
| Roof condition | Real throughput | Tank coverage (30 L) |
|---|---|---|
| Open atrium, light dust | 450-600 m2/h | 1,000+ m2 |
| Mall roof, some obstacles | 300-400 m2/h | 800-1,000 m2 |
| Sawtooth industrial | 200-350 m2/h | 400-600 m2 |
| Post-construction film | 200-300 m2/h | 400-500 m2 |
Who overestimates this the most?
Buyers who plan one robot to clean a whole portfolio in a day. Travel, setup and refills eat hours that the spec sheet ignores. If your buildings are far apart, budget one full day per roof, not a shared morning.
We can model your shift plan around the Lingfeng S1 or Lingyun Y3 via the contact page.
How do you build a shift plan from the numbers?
Take your roof area, divide by the realistic throughput, then add setup, refills and battery swaps. A 2,500 m2 mall roof at 350 m2/h needs about seven hours of cleaning. Add two refills and a battery swap and you are looking at a full night. That is the plan you take to the client, not the 600 m2/h headline.
| Roof area | Real hours | Shifts needed |
|---|---|---|
| 1,000 m2 | 3-4 h | 1 night |
| 2,500 m2 | 7-8 h | 1 long night |
| 5,000 m2 | 14-16 h | 2 nights |
| 10,000 m2 | 28-32 h | 4-5 nights |
Why do two robots beat one big one?
On large roofs, two units working separate bays finish faster than one unit doing passes end to end, because there is no travel time between far-apart sections. Two mid-size units also give redundancy: if one fails, the clean still finishes. The cost of a second unit often recovers within one avoided re-visit.
How does dust load change water use?
Water consumption tracks dirt. Light dust needs a thin film to lift particles; heavy film needs wetting, dwell time and a second pass. That can double water use and halve throughput on the same roof. Post-construction cleans are the extreme case, which is why many contractors run a dry rough pass first, then a wet finish.
What is the biggest efficiency mistake?
Running out of water or charge in the middle of a bay. Plan refills at bay boundaries, not mid-panel, or you leave a visible line. Carry a spare battery and a full tank from the start, and the shift finishes clean and on time.
How does battery age change the plan?
A new battery delivers its rated runtime; by year three, capacity may fall 20-30%. That means a shift that once just fit no longer does. Track cycles and plan a battery replacement around year three, before the runtime gap starts causing mid-shift top-ups that eat productivity.
Cold weather cuts usable capacity too. A battery that lasts four hours in summer may manage two and a half in near-freezing conditions, which changes the number of shifts for a winter clean.
What slows the robot most?
- Obstacle density and repositioning pauses.
- Two-pass finishes.
- Heavy dirt requiring dwell time.
- Battery swaps and refills.
- Tether re-routing over structures.
How do you compare two robots fairly?
Measure both on your own roof, same day, same dirt. Compare effective m2/h after refills and swaps, not the spec sheet. The faster machine on paper is often the slower one in practice once water and battery logistics are counted.
How do you set a water refill plan?
Map refill points to bay boundaries before the first clean. Running dry mid-panel leaves a visible mark and forces a re-pass. On a roof without a nearby tap, a small bowser or a hose run from the plant room saves more time than any throughput gain. Plan the water like you plan the battery, at section breaks and never in the middle of a run.
Key Takeaways
- Budget 250-350 m2/h on real roofs, not the 600 m2/h headline.
- A spare battery (under $900) beats mid-shift charging.
- 30 L of water covers 400-1,200 m2 depending on dirt.
- Two-pass finishes halve throughput; plan for them.
- Travel and setup dominate multi-building plans.