# Glass thicker at bottom myth

1. Sep 28, 2006

### Psi 5

A little math help please. If I wanted to build a centrifuge to spin a glass disc 12" in diameter to test the thicker at the bottom myth, how many rpm do I need to get approx. 1000G at the edge of the disc?

(I have been in an engineering career that doesn't require math for so long I forgot how to figure it out myself and I am too lazy to relearn something that would take someone here a minute to figure out)

FYI If I were going to build this (which I'm not), the glass disc would be riding inside a metal carrier (essentially a torus) and not connected to the drive shaft so that the disc was in pure compression. I just want an idea of what I would have to do to simulate a 1000 year time span on the glass (and yes I know I would have to spin it a year if all I could get was 1000G).

2. Sep 28, 2006

### Mech_Engineer

http://www.ac.wwu.edu/~vawter/PhysicsNet/Topics/RotationalKinematics/CentripetalForce.html

Go wild.

BTW, 1000g (on a 12in disc that would be about 1700rpm) would be pretty easy to get on a disc, I'd aim even higher, like 10,000g (5400 rpm on same disc) or perhaps even higher. Just make sure the thing is enclosed in case of catastrophic failure. Also: the entire disc will not be seeing the same acceleration, the center will see none while the outer edge will see the most. This will probably mess with a simple thickness measurement at the edge.

You might be interested to read this also, it's not like no-one has tried to debunk the "flowing glass" myth...

http://www.glassnotes.com/WindowPanes.html

Last edited: Sep 28, 2006
3. Sep 28, 2006

### Psi 5

Good point about the disc not seeing the same force throughout. I would probably change the design so that a small square of glass, say 1", was mounted 12" from the center of rotation. I just wanted the rpm for 1000G so I could scale it up with a good electric motor. 5400rpm should be easy.

Now that I think about it, I am planning on keeping my current car for quite a while, maybe I will make a holder for the piece of glass to fit one of my wheels and do a long term test on the cheap.

By the way your links aren't about anyone doing an actual test and I haven't heard about anyone actually doing this test though I do believe there will be no effect.

4. Sep 28, 2006

### chroot

Staff Emeritus
Frightening.

- Warren

5. Sep 28, 2006

### Mech_Engineer

Oh well, I tried.

The "test" hasn't been done because it doesn't NEED to be done, that link definitively explains why glass won't "flow." The "thicker-at-the-bottom" windows ended up that way due to the manufacturing process for the windows, not the length of time they were sitting there.

X2 on that... I hope he isn't engineering anything important.

6. Sep 28, 2006

### Mech_Engineer

Here, this is useful to note:

7. Sep 29, 2006

### zoobyshoe

I'm standing outside a cafe smoking last week and there's a group of teenagers also out there. One girl asks if anyone knows the time. I say "Look inside: there's a clock on the wall next to the counter." She looks, and turns a little red: "Someone just please tell me what time it is! I'm no good with those old fashioned clocks with the hands!"

8. Sep 29, 2006

### chroot

Staff Emeritus
I've moved Psi 5's comments about the site to the Feedback forum.

- Warren

9. Sep 30, 2006

### dlgoff

10. Dec 1, 2006

### AlephZero

Hey, just don't stand next to that thing while its running!!!!

Seriously, 10,000g is about the same conditions you get in rotating components in a typical jet engine. If you want to spin a disk up to that speed (been there done that, testing metal ones to see how long they will last before they break), first off you will need to put it in a vacuum tank to reduce the power to drive it. The tangential velocity at the edge of a 12 in disk at 5400 RPM is about Mach 0.6. That's going to generate a lot of air resistance and a lot of heat, and make a big dust storm in your workshop.

Second ,you had better do a stress analysis on the disk before you start t,o see if your glass has any chance of staying in one piece.

Third, you need a serious containment system. Typically if a disk like that bursts, if will split into two pieces, a 120 degree wedge and a 240 degree wedge (don't ask why, but that's usually what happens). Those two pieces will fly off tangentially at about half the speed of sound and a few sandbags ain't going to stop them.

And finally, you will have to do a very good balancing, job otherwise your drive shaft bearings will self destruct. Don't expect your local car tyre shop's balancing machines to be good enough, you probably need 2 orders of magnitude better than that. And getting the balance weights to stay on with a 10,000G acceleration trying to pull them off might also be an interesting design problem.

To summarize - don't try this at home...

11. Dec 4, 2006

### FredGarvin

Aleph beat me to it.