Hacker News Reader

Comment by ItsHarper | Hacker News Reader

I believe it's to allow room for the handrail belt to wear down, which brings its speed closer to the stairs until it starts de-syncing in the opposite direction. If it started perfectly, you'd have to replace the belt more frequently to maintain the same level of tolerance.

View on HN

If I’m understanding what you’re saying, couldn’t that be solved with some kind of belt tensioner?

According to https://www.youtube.com/watch?v=tZ8ehplVFp4&t=636 the handrail is driven by a friction wheel that wears out over time, so its diameter gradually decreases and the handrail speed slows down (until it gets too out of sync, and the friction wheel is replaced).

I wonder why they can't just use a toothed belt.

Because you'd have teeth rubbing out all through the course, which would eat them through very quickly, or you'd have to add toothed rollers throughout the course which would increase complexity tremendously.

If I were designing it and wanted to extend the maintenance interval, I think I’d use something like a CVT for a tensioner. The drive pulley would wear as normal, and the tensioner would be of a harder material and cone-shaped. As the drive pulley wears, the tensioner slides toward the large end so that the diameter of the drive pulley and the diameter of the section of the tensioner in contact with the belt remains a consistent value.

You could probably use a roller on the back side of the drive pulley to automatically adjust it, too. No electromechanical stuff needed.

Oh, and add a “squealer” - a piece of metal that makes an awful racket once the drive pulley wears down to a critical point, prompting those responsible that it should be replaced before it begins to induce unacceptable wear on the belt itself.

So when it’s new it feels broken, the friction wheel wears until it works properly for a period of time, and then it feels broken again, so they replace the wheel with on that goes back to making it still feel broken and start the cycle again. How frustrating.

That’s the case for many/most mechanical assemblies with wear parts.

I installed a new steering box in my Jeep a few years ago - or at least, I think that’s the name of the component, but it seems wrong to my ear. The part that converts rotational motion of the steering wheel into linear motion in the steering components for the front wheels.

Anyhow, it was very tight. Not only was there no play, but I could tell it was taxing my power steering system more than normal. It took about a year for that to level out, and another 2-3 years for it to feel “normal” again. Another ~5 years and it would have exhibited perceptible “play” in the steering.

Reply on news.ycombinator.com