Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Robotics Technology

Window Washing a Skyscraper Is Beyond a Robot's Reach 203

Posted by samzenpus from the does-not-compute dept.
HughPickens.com writes "Patrick McGeehan writes in the NYT that the image of a pair of window washers clinging to a scaffold dangling outside the 68th floor of 1 World Trade Center have left many wondering why robots can't rub soapy water on glass and wipe it off with a squeegee relieving humans of the risk of injury, or death, from a plunge to the sidewalk? The simple answer, several experts say, is that washing windows is something that machines still cannot do as well as people can. "Building are starting to look like huge sculptures in the sky," says Craig Caulkins. "A robot can't maneuver to get around those curves to get into the facets of the building." According to Caulkins robotic cleaning systems tend to leave dirt in the corners of the glass walls that are designed to provide panoramic views from high floors. "If you are a fastidious owner wanting clean, clean windows so you can take advantage of that very expensive view that you bought, the last thing you want to see is that gray area around the rim of the window."

Another reason for the sparse use of robots is that buildings require a lot more maintenance than just window cleaning. Equipment is needed to lower people to repair facades and broken windows, like the one that rescue workers had to cut through with diamond cutters to rescue the window washers. For many years, being a window cleaner in Manhattan was regarded as one of the most dangerous occupations in the world: by 1932, an average of one in every two hundred window cleaners in New York was killed each year. Now all new union window cleaners now take two hundred and sixteen hours of classroom instruction, three thousand hours of accredited time with an employer and their union makes sure workers follow rigorous safety protocols. In all, there are about 700 scaffolds for window washing on buildings in New York City, says union representative Gerard McEneaney. His members are willing to do the work because it pays well: as much $26.89 an hour plus benefits. Many of the window cleaners are immigrants from South America. "They're fearless guys, fearless workers."
This discussion has been archived. No new comments can be posted.

Window Washing a Skyscraper Is Beyond a Robot's Reach More

Window Washing a Skyscraper Is Beyond a Robot's Reach

Comments Filter:

  • A cost equation (Score:5, Insightful)

    by seven of five ( 578993 ) on Friday November 14, 2014 @12:06PM (#48386035)
    Human window washers must be cheaper than self-cleaning glass or robots. For now.

    • since there are no such robots, yes. The "self-cleaning" glass does need hosed down every now and then, not totally self-cleaning

      • Re: (Score:3, Informative)

        by rnws ( 554280 )
        Pilkington's "Activ" glass will self-clean in the rain. Though with credit to your comment, they recommend hosing it during prolonged dry spells.

        • freezing rain and snow might make a mess too

          • and here in chicago sometimes the moon is brown or orange because of pollution. I'd wonder how well the glass does in rain passing through fumes of oil, unburned gasoline ,coal ash, industrial solvents, etc.

            wonder if I have the lungs of a smoker from all that shit, though I've never taken one puff....

    • Re:A cost equation (Score:5, Interesting)

      by swb ( 14022 ) on Friday November 14, 2014 @12:45PM (#48386415)

      Cost seems like a good explanation. I have no idea how long it takes to wash an entire building, but at the pay rate quoted in the story they could employ two guys for a whole year for under $300,000. A really good machine would probably be a million dollars to design and implement and would still need maintenance and probably a full-time guy to operate it.

      I find it hard to believe there isn't a technological solution to this that could limit exterior manual cleaning to something done every couple of years. The idea that some machine used on the old World Trade Center sucked seems like a weak excuse -- a car from the early 70s sucks now, too, compared to current cars employing modern technology.

      Some ideas off the top of my head:

      1) A self-contained pressure washing system that recycles its own water with some combination of sheeting chemistry, forced air and microfiber to ensure the window is clean. Hard to believe you can't design a machine capable of cleaning well.

      2) Why not integrate a cleaning system into the window framing or structural system capable of washing either a bank of windows or several vertical floor sections? It might need fixing itself, but it could be combined with once-every-five years exterior maintenance and "thorough" cleaning.

      3) Are there coatings or other materials science solutions that would make dirt and pollution less likely to stick to the windows and make cleaning necessary only every 5 years or when exterior maintenance was necessary?

      • Re:A cost equation (Score:5, Informative)

        by ottothecow ( 600101 ) on Friday November 14, 2014 @01:45PM (#48386915) Homepage
        Lake Point Tower in Chicago has an automated window system that seems to work pretty well.

        https://lakepointtower.wordpress.com/2009/07/17/doing-the-windows-at-lake-point-tower/#more-461 [wordpress.com]

        Here is a terribly done youtube video [youtube.com] that purports to show the machine and process (but...well...let's just say he isn't a very good cameraman). Of course, it is a design that is fully integrated with the building. I worked in the building for a while and I always found the windows to be quite clean. Better than other buildings because they could clean the windows more often. You need tracks running down the edges of all of the winows, so you couldn't retrofit this onto an old building...but this problem does seem solvable for new constructions.

        Significantly more complicated for old buildings. I am in a ~100 year old brick and stone facade building. The windows are inset, and not uniform in size, They have window sills and some of them are divided into multiple panes while other areas have bigger sheets of glass. I don't know how a machine would manage this...and having a custom machine might make sense for a large condo building (pretty sure the lake point tower cleaner robots are running most of the time since they have 70 stories of wall-to-wall windows to clean), but a custom machine for a shorter office building with limited windows is not going to be more effective than having a couple guys scrub the windows every few weeks.

        Here, the guys that clean the windows actually rappel down. I think a lot of older buildings have too much stuff sticking out (and the windows are more spaced out), so having a hanging scaffold doesn't make sense. Instead, they just harness in and kick themselves around the building, using a suction cup to hold themselves in place while they clean a window. Incredibly fast and even considering they get paid pretty well, incredibly cheap.

      • You could always design the building in the first place to minimize those nooks and crannys that a robot couldn't reach. That is way too practical of a solution though and would never catch on so long as the people inside need a way to show off all their money.

        • Or you could read the summary and figure out one of the real reasons:

          Another reason for the sparse use of robots is that buildings require a lot more maintenance than just window cleaning.

      • Cost seems like a good explanation. I have no idea how long it takes to wash an entire building, but at the pay rate quoted in the story they could employ two guys for a whole year for under $300,000. A really good machine would probably be a million dollars to design and implement and would still need maintenance and probably a full-time guy to operate it.

        And on top of that - it would probably be a custom solution. (Meaning expensive to maintain, and very expensive to replace when it could no longer be ma

    • Human window washers must be cheaper than self-cleaning glass or robots.

      Not necessarily. An office with a "very expensive view" is a status symbol. So is a servant. Just like having an automaton wash your windows appeals to a certain kind of personality, having an actual human being do so appeals to another.

      People who only care about the cost don't rent offices in skyscrapers and if they would, they'd just let the windows stay dirty.

      • Re: (Score:2)

        by jbengt ( 874751 )

        People who only care about the cost don't rent offices in skyscrapers . . .

        True, but people that care about having a large pool of workers to hire and close proximity to as many clients as possible tend to want offices downtown. Downtown is full of skyscrapers in order to house as many a commuters as possible during working hours. So many people rent a space with a "very expensive view" even though they don't care about it as a status symbol.

    • Human window washers must be cheaper than self-cleaning glass or robots. For now.

      high rise architecture as sculpture
      difficult to navigate, no longer a simple curtain wall.
      demands cleaning, maintenance and repair of both windows and facades -- and tenants will settle for nothing less than perfection.
      there are no robots who can do this work and that isn't going to change any time soon.
      it's all there in the summary.

      For comparison, the twin towers of the WTC had 43,600 windows --- over 600,000 square feet of glass. The World Trade Center - Facts and Figures [nysed.gov]

  • For a human, using a sponge and squeegee combo is probably the most effective way to clean a window. For a robot, I would imagine that the answer is something more like a pressure washer, with a hood which covers the work area and reclaims the wash water. The water would then be filtered and reused until the particulate count rose too high, at which point it would be flushed and replaced with fresh. A sheeting additive would be used to cause the water to run off without spotting.

    This probably wouldn't replace human window washing entirely, but it seems like it has the potential to replace at least some of the washes.

    I've often wondered if anyone has ever tried a project to make a building which washes itself, using a robot designed for the building, and a building designed for the robot. I can imagine many problems with such a project without even undertaking it, mostly related to critters taking up residence in the mechanisms and/or tracks, but if it operated continuously that might well eliminate some of those objections. A universal window washing robot has a more complicated task than such a device would.

    • For a human, using a sponge and squeegee combo is probably the most effective way to clean a window. For a robot, I would imagine that the answer is something more like a pressure washer, with a hood which covers the work area and reclaims the wash water. The water would then be filtered and reused until the particulate count rose too high, at which point it would be flushed and replaced with fresh. A sheeting additive would be used to cause the water to run off without spotting.

      This probably wouldn't replace human window washing entirely, but it seems like it has the potential to replace at least some of the washes.

      I've often wondered if anyone has ever tried a project to make a building which washes itself, using a robot designed for the building, and a building designed for the robot. I can imagine many problems with such a project without even undertaking it, mostly related to critters taking up residence in the mechanisms and/or tracks, but if it operated continuously that might well eliminate some of those objections. A universal window washing robot has a more complicated task than such a device would.

      Did you even read the article? You'll find it discusses how the old World Trade Center Towers had built in devices that were made specifically for the building that would automatically go up and down cleaning it. The only problem was they missed the corners and creases of each pane and the rich people at the top of the building didn't want the grimy borders to their new expensive view of NYC.

      It sounds like you have a lot of ideas for building a nice big heavy expensive machine that moves up and down

      • Did you even read the article?

        YMBNH

        You'll find it discusses how the old World Trade Center Towers had built in devices that were made specifically for the building that would automatically go up and down cleaning it. The only problem was they missed the corners and creases of each pane and the rich people at the top of the building didn't want the grimy borders to their new expensive view of NYC.

        That seems like a solvable problem, in a variety of ways. The easiest would be to add a window border on the inside of the glass that covers up the area that doesn't get cleaned, but actually solving the problem also seems doable.

        • It seems that designing the building exterior and windows to allow for robotic or other automatic means of washing would be the right approach. After the fact, its a lot harder.

        • That seems like a solvable problem, in a variety of ways. The easiest would be to add a window border on the inside of the glass that covers up the area that doesn't get cleaned, but actually solving the problem also seems doable.

          Hmm... I would say it is a work around, not a solution to the problem. Why? Because the dirt is still there. your suggestion would simply hide the dirtiness from seeing rather is a way to clean it (solution)...

          • Hmm... I would say it is a work around, not a solution to the problem. Why? Because the dirt is still there. your suggestion would simply hide the dirtiness from seeing rather is a way to clean it (solution)...

            Dirt on painted surfaces is a problem. It traps more dirt, it traps grease, it traps water and holds it against the surface. Dirt on glass is only a problem insofar as it impedes visibility. If nobody sees it, it isn't a problem. This technique is already used in the case of your car's windshield, and it works quite well. We're talking about only a small percentage of the total area of the window.

          • Re: (Score:2)

            by itzly ( 3699663 )
            The dirt on the brick facade is still there too.

          • The problem isn't dirt on the windows, the dirt doesn't affect the performance of the building in any meaningful way. The non-window areas don't need to be cleaned spotless, because the occupants can't see them. The GP's post provides the most economical solution - masking to avoid the appearance of the problem.

            Consider this is a solution which has been done for ages in the area of movie projections and photography: no lens can make a perfectly rectangular, evenly lit, properly focused area on a flat surfac

      • Re: (Score:2)

        by jandrese ( 485 )
        The article read like "The window washing robots we tried sucked, the problem must be impossible!"

        • Re: (Score:2)

          by swb ( 14022 )

          Worse, it seems to also add "...and everybody has a financial stake in the current system, too"

          It read to me like "I rode in a car once 40 years ago, and it sucked. Cars don't work well."

          • On another hand, anyone here knows an automated car wash that can actually do a great job at washing a car? So far, I haven't find a single one that do the job like a human is doing it. My guess is the business case for an automated car wash capable to wash clean a car is significantly better than any skyscaper window washing robot and we are still waiting for a decent one.

            • This isn't a mechanical problem, it is a vision problem. Humans miss spots while cleaning all the time, but we have one advantage, we can easily tell what is and isn't dirty, then we go back and correct. The issue I see is we aren't solving the vision and object recognition problem for some time yet so humans will still have that advantage.

      • yet google can develop self driving cars.

        I think we could implement AI/pattern recognition that could do it. It's more an economics question for the reasons you just listed. 26ish bucks an hour vs multi-million dollar machine, and the R&D needed to develop it..

    • It is cheaper to hire people to do it, even if occasionally you have to pay a higher spike in insurance rates when a person dies.

      Until human life is A LOT more expensive, or robot solutions are a lot cheaper, it will be cost effective to skinsource menial but complex and dangerous jobs.

      We like to pay lipservice to all kinds of factors, but it all comes down to money.

    • If you designed that sort of building (and went through that cost), I'd imagine you could do something cool like flip the windows 180 degrees so someone inside could wash them.

      Or you'd do just windows with no seam or edges for robots to miss.

      • Or you'd do just windows with no seam or edges for robots to miss.

        Making sure that you retain the glass, even in a seismic event (at least one insufficient to shatter it) is an important job. You can glue a window onto a car, but you don't want to glue it onto a skyscraper. It's pretty hard to do that job with frameless edges. However, you could solve that problem the same way they do in cars. Look at your windshield sometime, it's got a frame painted onto the glass. That covers up from the inside the ugly metal bits which are covered up by trim on the outside, as well as

      • At the top of a taller building the winds are 30 to 50MPH stronger than at ground level. When you popped that window out of the frame the average office would explode in a fury of paperwork.

    • Re: (Score:2)

      by njnnja ( 2833511 )

      Still not thinking big enough. Who needs windows? Giant LCD screens on the inside, tiny pinhole cameras on the outside FTW!

      I know, it uses a lot of energy to power those screens, so it's not very environmentally friendly. But they could place big solar cells on the sides of the building to power them. Unfortunately, eventually those solar cells would get dirty and lose efficiency. In theory you could lower someone down on a scaffolding who could clean the solar cells when they got dirty, but as a practic

    • For a human, using a sponge and squeegee combo is probably the most effective way to clean a window.

      The high tech alternative to the sponge and squeegee isn't a robot, it's a chemical coating on the glass.

      Titanium dioxide is a photocatalyst: it's a material that makes chemical reactions happen when the right kind of light shines on it. The right kind of light for titanium dioxide is ultraviolet (UV), the super-blue, high-energy part of sunlight that our eyes can't see, but that nevertheless can give us sunburn even on a cloudy day. When ultraviolet light hits the titanium dioxide coating of a self-cleaning window, electrons are generated. These turn water molecules from the air into hydroxyl radicals that make chemical oxidation and reduction reactions take place on the coating. In effect, the hydroxyl radicals attack organic (carbon-based) dirt and chop it up into smaller pieces that are much easier for rain to wash away. Since the reactions happen on the titanium coating, on the very surface of the glass, they attack the lowest layers of the dirt, loosening encrusted muck from the glass very effectively by chipping it away from the inside out (the opposite of normal window cleaning, where you effectively scrub the dirt from the outside in.)

      Self-cleaning windows [explainthatstuff.com]

      There are two problems:

      Rain will probably not reach every corner of the window.

      The literal curtain wall of the early box like glass towers has gone out of fashion.

      The coating can add maybe 20% to the cost of a window.

      The Twin Towers had 43,600 windows --- 600,000 square feet of glass. 'Mind you, the architect of the WTC was notoriously afraid of heights and windows were kept a

  • Using simple magnets the windows could self clean in the same way someone can clean the inside of a fishtank

    • Having several fishtanks myself. I can attest to the fact that those magnet cleaners suck.

      • I wonder if you could reasonably make one that had a rotating element, or was one. I'm thinking either a rotating drum on the inside with magnets in and brushes on which is attracted to your outside device which is the same, but the drum has rubber on it instead of brushes — or a motorized jobber on the outside, with a part which rotates around an axis perpendicular to the glass.

        • Both sides of the glass need to be cleaned at some point. Why not brushes on both sides?

          • Both sides of the glass need to be cleaned at some point. Why not brushes on both sides?

            Because cleaning accessible glass is a job easily done with typical glass-cleaning methods which don't involve fiddling with magnetic pucks. You can get it done with soap, water, and an ordinary squeegee/sponge combo.

            There is a window roomba already, it leaves streaks.

      • When I had a salt water tank, I found these magnets to be awesome. The only real issue was that you had to have one for each pane. It would be kind of clever if they made one magnet slightly larger than the other so you could get the inside one to scrub slightly behind the outside trim piece.

        • When I had a salt water tank, I found these magnets to be awesome. The only real issue was that you had to have one for each pane. It would be kind of clever if they made one magnet slightly larger than the other so you could get the inside one to scrub slightly behind the outside trim piece.

          If you're clever you can get it to hop from pane to pane. I Could do it ever time. It works for very basic cleaning. But algae will definitely build up despite it and you have to go in with a real scrubber.

  • I'm going to say there will be a window roomba for skyscrapers within a decade. It's too lucrative a market not to pursue.

    The improvement in suction cups have been here for a while. Short of some innovative cleaning system that require little/no water, resupply and dirt offloading can be handled by some ancillary robot that runs back and forth to some main hub.

    All that will really be needed is some safety system to keep it falling from pedestrians. If it's a cable, then the ancilliary robot might be done

    • And, they need the humans for more than just washing the windows.

      From TFS:

      Another reason for the sparse use of robots is that buildings require a lot more maintenance than just window cleaning. Equipment is needed to lower people to repair facades and broken windows

      So, your Skyscraper Edition Roomba still doesn't cover all the needs of buildings like this, and you'll still need to be able to get people there for other tasks. Someone still needs to be able to physically get there anyway.

      Me, no way I'd do th

  • all robots, 3d printers, burning man all the time? Perhaps Bennett could weigh in on my infrequent submitter observation.

  • Question:

    Why are we not designing these buildings with the "robots" built-in?

    Surely it can't be that hard to include "self-cleaning window system" in the multi-million dollar installation costs?

    • I worked at the Sears tower many years ago, and we had machines that cleaned the outside of the building, so it definitely can be done.

  • Shocked... (Score:3, Interesting)

    by camazotz ( 1242344 ) on Friday November 14, 2014 @12:20PM (#48386181) Homepage
    I'm shocked that those window cleaners make "up to $26.89" WTF seriously? They just bragged like that was a good number for that sort of work....? I guess, relatively speaking, it might be good for the alternative choices those workers have, but I sure as hell wouldn't do that for $26.89. Why is it that all the high mortality rate jobs have such shitty wages?

    • Why is it that all the high mortality rate jobs have such shitty wages?

      Because it's a crime to be broke.

      • Re: (Score:2)

        by mark-t ( 151149 )
        Citation?

    • I'm shocked that those window cleaners make "up to $26.89" WTF seriously? They just bragged like that was a good number for that sort of work....?

      That is $27/hr + benefits.

      The median household income in the US is $52,000, something the geek tends to forget.

    • Re: (Score:2)

      by radtea ( 464814 )

      Why is it that all the high mortality rate jobs have such shitty wages?

      Because they are all done by men, whose lives are not valued by our society.

  • Patrick McGeehan has apparently never seen a windshield wiper.

    Apparently he's never heard of youtube either:
    https://www.youtube.com/watch?... [youtube.com]

    Seriously people, don't make declarative statements in your professional life without at least doing a google search first.

    • , windshield wipes cannot and do not clean the entire window; you are silly

      • , windshield wipes cannot and do not clean the entire window; you are silly

        Two windshield wipers at opposing corners can cover a window. And if you designed the wipers cleverly, you might well get away with having one wiper ride a cam and hop over the frame to wipe two windows. Problem is, wipers don't wash anything, even if you spray water on the window. They just wipe it. But perhaps you could have a roller brush or sponge on one side which would be motorized and roll counter to the movement of the arm, and a squeegee on the other. It'd scrub across the window, then flip over at

  • Surely a remote control device of some sort should suffice, be smaller, faster and good enough if not perfect.

    I’m amazed the windows aren't pre-designed for some kind of semi-automatic, rc-controled cleaning device.

    There may then be some difficult areas that occasionally need a human crew on the outside, but if you could get this down to 5% or even 20% it would be a big safety win.

    I imagine a range of remote units. Some very dumb and cleaning the bulk of the windows, then bigger more expensive units t

  • Umm, the original twin towers... (Score:3, Interesting)

    by ZorinLynx ( 31751 ) on Friday November 14, 2014 @12:42PM (#48386377) Homepage

    The original twin towers DID use robotic window washers. They were built in 1971.

    This problem has already been resolved, but they didn't design the new building in a way that robotic window washers could be implemented easily.

    Oops.

  • That would work out to less than 34000$ a year of full time work. Even with a 10 hr/week overtime, double time for over time it will be around 50K a year. With words like "could" and "as much as" thrown in, it is probably the maximum pay. This is not high pay by NYC standards. It just shows lots of poor people are willing to risk their life and limb for a relatively low salary.

    Any pay increase to these low end workers will almost immediately be spent creating economic activity in this country, boosting GD

  • It wasn't that long ago that all LCDs were produced with a manual process of wiping and buffing the liquid crystal onto the glass substrate. No machine could be made to perform the task. This limited the size of panels that could be made and reduced yield with flaws from mistakes and contaminants. They finally automated the process to achieve the panels we have today. The same will be true of window washing robots.

  • is hampering the free market. Why, without the union and all that pesky classroom instruction and mentored apprenticing, I bet we could hire 100 window washers at $5 an hour! No need to worry about why robots can't do what my $5 South American Mexican can do. May the power of Axioms absolve you!

  • The Tower which I will always know as the Sears Tower had purpose-built automated washing systems. When you are up in said tower, you can see dozens of other buildings that have rails on the roof for their automated washing systems. It sure seems like a lot of skyscrapers have these systems. Kind of flies in the face of what the article says.

  • Assume for a moment that robotic window washers could work:

    Who will wash the washers?

    • Assume for a moment that robotic window washers could work:

      Who will wash the washers?

      The Window Washer Washers.

Slashdot Top Deals

He has not acquired a fortune; the fortune has acquired him. -- Bion

Close