[ajb]

The alternative is counterweights. Not exactly new, but there looks like there is enough space for them in this case. The ones in the house I grew up in were a pair of concrete cylinders on either side of the door, maybe 10-15 cm in radius and 1m in length, mounted vertically. Not sure why they're not more common - seemed a completely reliable system.

[derekp7]

The door needs less counter weight as it goes up. In order for weights to work you either need several weights, with different weights bottoming out as the door rises. Or have the cable on a spiral pulley may work.

[ajb]

I don't think that's true of the mechanism I saw, which I believe was quite common, but it's 20 years since I looked at the mechanism.

Ok I've found a video with one in: https://m.youtube.com/shorts/zLuJYlHBSkU

The counter weight is always taking the full weight of the door, there's just a linkage which rotates the door on the way up. So, not the same as the spring ones, which follow a curved channel.

This makes me less keen on them as in theory it's going to fall when it wears out, albeit with only 50% of the weight. Unless there's another failsafe.

[ajb]

Thinking about it, the failsafe should be that if one side fails first, it gets stuck against the track. Although I don't think the one in the vid would, because the track points inward

[scotty79]

Some gas piston would be cool.