Rotating skyscraper idea

[wasunka]

Has anyone else done any thinking on this? 

If you're not sure what this is, Google "rotating skyscraper"

I think I am closing in on a possible way to build one in N, say 18 - 24 floors.

Joe

[CaptOblivious]

<a href="http://www.youtube.com/v/A-cfB6xFSKM" target="_blank">http://www.youtube.com/v/A-cfB6xFSKM</a>

That's pretty cool! I'd hate to be in the pool or a rolling office chair when my floor switched directions though!

Doing this in N would be very tricky indeed; it would require quite a lot of mechanical cleverness that is totally beyond my ken...any watchmakers here? But it would be way awesome.

[grumbeast]

This is seriously cool,

 but I don't think its a mechanically sophisticated as it looks (I'm not claiming to be able to do it mind!!! :) :) ).  All that is happening is that each floor is rotating simply about x, is when and how they rotate in relation to each other that causes the cool effects (and because they aren't round floors (they simply look square))  A computer program to track and specify the direction and rotation of each floor would be pretty simple, actually what what would really cool would be to have the building reshape or change based on environmental factors (imagine something like the direction an d speed changing based on the weather forecast, you could just look at the skyline to see if it was going to rain tomorrow, add coloured lights to that and it could also indicate the status of the stockmarket!   I have to stop, I'm getting waaaaaay too carried away!   

Graham

[CaptOblivious]

It's even simpler than that, Graham: I bet you could do it with a single rotating shaft running up through the building, with a series of cams, one for each floor, each slightly offset from the previous, that deflect each floor in turn; a spring then returns the floor to it's resting position. However, you'd have to rig it so that the spring didn't just snap it back; use something to generate friction so it returns at the same rate it was deflected.

There's probably a way to do it with two camshafts and no springs, but I'm beating my head for a clear picture of what it would look like.

[Mudkip Orange]

Except that no one in Dubai can get financing anymore, they're starting to default on all their loans, so this has about as much chance of getting built as... well...

[grumbeast]

You're undoubtedly right!  I'm rather mechanically challenged and always think in terms of software! :)

Graham

It's even simpler than that, Graham: I bet you could do it with a single rotating shaft running up through the building, with a series of cams, one for each floor, each slightly offset from the previous, that deflect each floor in turn; a spring then returns the floor to it's resting position. However, you'd have to rig it so that the spring didn't just snap it back; use something to generate friction so it returns at the same rate it was deflected.

There's probably a way to do it with two camshafts and no springs, but I'm beating my head for a clear picture of what it would look like.

[CaptOblivious]

You're undoubtedly right!  I'm rather mechanically challenged and always think in terms of software! :)

Graham

As a friend of mine was once said-and he was right!-there is always a software solution (first said in response to the catsup pump at Wendy's not working).

[wasunka]

Well, I've played with the idea a bit more in the back of my mind (dangerous place - I don't like to go there!) and have a possible solution as to how.

First, a rigid central column, possibly PVC, just to maintain alignment.  Second, each floor is separated from the next by a 3" lazy susan - cheap and low friction.  Third, even in N scale, each floor should be thick enough to mount a small RC servo.  It would have a linkage to the central column.  Fourth, run control cables (3 wires) down the central column.  And fifth, there are servo controller boards available that will handle up to 32 servos from one small controller.  That should be plenty to get the visual effects.

After that, it is all software.  While 360 degree rotation wouldn't be possible, somewhere between 180 and 270 should be achieveable.  Just think of the sequences, and the effect at shows.

Financially worth it?  Probably not.  But, who knows, someone, just for the challenge, . . .

Joe

[Martijn Meerts]

Small enough servo's might get a bit expensive if you need 32 of them ;)

Of course, it'd be possible to just test something with say 4 or 5 servos, hook them up and write some controller code for that. That way you can try some things without shelling out a lot of cash.

[wasunka]

Small enough servo's might get a bit expensive if you need 32 of them ;)

Of course, it'd be possible to just test something with say 4 or 5 servos, hook them up and write some controller code for that. That way you can try some things without shelling out a lot of cash.

Yep, after that, it's "just a bit more software." <BG>

[cteno4]

Joe,

hey what about just staring with the rotating roof restaurant?! lots of those around! would be fun to see people in the windows spinning by.

cheers

jeff