Robot Vacuums and Transition Strips

The Half-Inch That Strands a Robot

Every home has them, and most people have never looked twice: the slim metal or wood transition strips where tile meets carpet, where one room's flooring butts against another's, and the raised wooden sills tucked under interior doors. To you, they are nothing — your foot clears them without a thought. To a disc-shaped robot riding on two small drive wheels, a strip even half an inch proud of the floor is a vertical wall it has to scale from a standstill.

Two failure modes show up again and again in our testing. The first is the beaching: the robot climbs partway onto the lip, the chassis bottoms out, the drive wheels lose their grip, and it sits there spinning helplessly — usually announcing the problem with a chirpy error tone — until a human lifts it off. The second is quieter and, frankly, more aggravating. The robot rolls up to the strip, decides it can't make the climb, reverses, tries a fresh angle, fails again, and eventually writes that doorway off as a dead end. The room beyond simply stops getting cleaned. You won't notice for a week, until the dust under the bed tells on it.

It gets worse on the return trip. A robot carrying a full bin and a belly of mopping water is heavier and sits lower than the empty one that set out, so a threshold it cleared on the way in can strand it on the way back. That's exactly why our protocol forces a 3cm threshold crossing on the return path of an under-furniture run — the moment a robot is lowest and least likely to make it. The strip that looks trivial in a showroom is the one that quietly defines where your robot can and can't go.

How High a Normal Robot Really Climbs

For a robot with no lifting hardware — which is still the overwhelming majority of what sells, certainly everything in the budget tier and most of the mid-range — the honest, repeatable ceiling is around 20mm, roughly four-fifths of an inch. Some clear a hair more on a good day; plenty of cheaper models start hesitating well before that. Above 20mm you're gambling, and the odds get worse every millimeter.

The thing that decides that number isn't suction, despite where the marketing spends its energy — it's geometry. Bigger drive wheels, a little wheel travel (a touch of suspension), more torque, and more ground clearance under the chassis all push the climbing height up. This is why two robots with identical headline specs behave completely differently at the same doorway: the one with larger wheels and a bit of give in its suspension walks over a strip that beaches its flatter-bottomed rival. While everyone fixates on suction ratings, the measurement that actually decides whether a room gets cleaned rarely gets a mention.

Two more real-world wrinkles. Manufacturers quote climbing height against a clean, square-edged test obstacle with a fresh battery — your beveled brass strip, hit at an angle because that's how the robot's path happened to line up, is harder than the lab figure suggests. And approach angle matters enormously: a robot that meets a strip head-on climbs far better than one sidling up to it diagonally. You don't get to pick the angle. The robot's pathing does.

The Climbing-Leg Generation, Minus the Marketing

The genuinely interesting development of the last couple of years is robots that physically lift themselves over obstacles instead of just bullying through them. Dreame's ProLeap system is the headline act: the Dreame X50 Ultra extends retractable legs to clear obstacles up to 2.36 inches (60mm), and the bigger X60 Max Ultra Complete pushes that to a frankly absurd 3.47 inches (88mm). To put that in perspective, 88mm is taller than most thresholds and most thick rugs — it is not, before anyone asks, enough for stairs.

Roborock's answer is AdaptiLift. The Roborock Qrevo CurvX clears thresholds up to about 4cm (40mm) on a tiered system, and the flagship Saros Z70 climbs to roughly 3.47 inches and adds an actual robotic arm for nudging small objects out of the way. Ecovacs runs an AdaptiLift chassis of its own on the Deebot T30S Omni — and that one earned a genuinely positive note in our hands-on testing: the AdaptiLift chassis handled every door track we threw at it, which is not something I say lightly about threshold claims.

What these systems share is that the climbing is conditional. Most of them lift only when the robot detects an obstacle it has decided is worth the effort, and the tallest figures tend to apply to specific, ideal conditions. The hardware is real and it works — but read "up to" exactly the way you'd read it on any other spec.

When the Legs Earn Their Keep — and When They're Wasted Money

Here's the part the product pages won't tell you: most homes don't have a single transition strip taller than 20mm. If your tallest threshold is a 12mm metal saddle between the kitchen and the hall, a competent mid-range robot drives over it without drama, and the climbing legs on a flagship do absolutely nothing for you all day. You'd be paying a steep premium for a capability your floors never call on.

The legs genuinely matter in a narrower set of homes than the marketing implies: older houses with tall wooden door sills, split-level or sunken rooms with chunky saddle thresholds, and floors where the robot has to climb up onto a thick rug or high-pile carpet to clean it at all. If that's your house, the lifting hardware is the difference between whole rooms getting cleaned or skipped, and it's worth every cent. There's a secondary perk too — the same mechanics tend to help a robot mount thick carpet and keep its mop pads off it — but if your home is flat and your strips are low, treat that as a nice bonus, not a reason to spend.

Fixing It Without Buying a Flagship

If you've measured your strips and one of them is over the line, you have cheaper options than a flagship robot. None of them are glamorous. All of them work.

Threshold ramps

A small rubber or silicone threshold ramp — a wedge that sticks or glues against the high side of the strip and gives the robot a gentle slope instead of a wall — costs only a few dollars and solves the problem outright for most raised strips and sills. It's the single highest-leverage fix on this list. Buy one sized for the strip's height, set it on the approach side, and a 20mm-limited robot suddenly clears a 30mm sill like it isn't there.

No-go zones and a little rearranging

If you can't or won't touch the floor, you can just tell the robot to stop fighting. Drawing a no-go zone at a doorway it keeps beaching on ends the wasted retries and the random error chimes, and you clean that one room another way. Relocating the dock to the side of the house with the hardest thresholds is another quiet trick — it means the robot starts its run already on the right side of the wall and never has to cross the worst strip to get home.

Removing the strip

Plenty of transition strips, especially the screw-down metal ones, lift out in five minutes with a screwdriver, and a lot of rooms read perfectly fine without them. This is the nuclear option and it isn't for everyone — renters should leave fixtures alone, and some strips are genuinely holding a flooring edge down — but if a strip exists purely as decoration and it's stranding your robot every day, removing it is free.

What to Measure Before You Spend a Cent

Before you compare a single spec, do this: take a tape measure, walk your home, and find your tallest transition. Measure from the lower floor surface up to the top of the strip or sill — that height, not the average, is the number that matters, because your robot is only as capable as its worst doorway. Pay special attention to the sills under interior doors and the lips where hard floor meets thick carpet; those are the usual culprits.

Then ask the question that actually settles the purchase. If your tallest strip sits comfortably under 20mm, almost any robot in your budget handles it, and you can shop on cleaning, mopping, and dock features like a normal person. If it's between 20 and 40mm, you're choosing between a robot with lifting hardware and a cheap rubber ramp. Above 40mm, even most climbing robots want help, so a ramp or a no-go zone is the realistic answer. Folding this into your pre-purchase prep takes ten minutes and saves you from the most expensive mistake in this category — paying flagship money to solve a problem you didn't have, or buying cheap and watching half your house go uncleaned.

The Bottom Line

Transition strips are the limitation hiding in plain sight. They don't appear in the spec comparison, they don't show up in the unboxing video, and they strand more robots than any pascal figure ever will. The good news is that this is the easiest limitation to plan around. Measure your tallest strip first. If it's low, ignore the climbing-leg hype and buy on cleaning performance. If it's high, decide honestly whether you'd rather pay for ProLeap-class hardware or stick a rubber wedge against a doorway and move on. Both are valid. Buying blind is not.

Transition Strips: The Quick Reference

• Measure your tallest strip first: the height of your worst threshold — not the average — is the spec that decides which robots are even in the running.
• 20mm is the no-legs ceiling: a standard robot reliably clears about 20mm (0.8in); plan on a ramp or climbing legs for anything taller.
• Climbing legs only pay off above ~20mm: Dreame ProLeap, Roborock AdaptiLift and Ecovacs AdaptiLift earn their premium in homes with tall sills or split levels, and do nothing on flat floors.
• A cheap ramp often beats paying for climbing legs: a rubber threshold ramp turns a wall into a slope and fixes most raised-strip problems for a few dollars.
• Wheels and clearance, not pascals: bigger drive wheels and more ground clearance decide climbing ability — suction ratings tell you nothing about it.
• Watch the return trip: a full bin and water tank lower the robot, so a strip it cleared empty can strand it on the way back to the dock.

Frequently Asked Questions

How high a threshold can a robot vacuum climb?

A robot without special lifting hardware reliably clears a threshold of about 20mm, roughly 0.8 inches, and most start to struggle above that. Climbing-leg models go far higher — the Dreame X50 Ultra handles up to 60mm and the X60 Max Ultra up to 88mm. Measure your tallest strip and compare it to those numbers before you shop.

What's the cheapest way to get a robot over a transition strip it can't cross?

A rubber or silicone threshold ramp is the cheapest fix, usually just a few dollars. It sticks against the high side of the strip and gives the robot a slope instead of a wall, which lets a standard robot clear a sill well above its normal limit. If you'd rather not modify the floor at all, a no-go zone stops the robot from wasting time and chiming at that doorway.

Do I actually need a robot with climbing legs?

Only if your home has thresholds taller than about 20mm — tall wooden door sills, split-level rooms, or chunky saddle strips. Most homes don't, in which case climbing legs add a steep premium for a capability your floors never use. Measure your tallest strip first; if it's low, put the money toward cleaning and mopping instead.

Will a transition strip damage my robot, or just stop it?

It usually won't damage the robot, but it strands it. The robot either beaches itself on the lip with its wheels spinning, or it gives up and marks the doorway as impassable, which means the room beyond quietly stops getting cleaned. The second failure is the sneaky one, because there's no error to alert you — you just find dust building up where the robot stopped going.

How do I measure my transition strips before buying?

Use a tape measure and measure from the lower floor surface up to the top of the strip or door sill. Do this for every transition in the home and note the tallest one, since the robot is only as capable as its worst doorway. Pay particular attention to interior door sills and the edges where hard floor meets thick carpet, which are usually the highest points.