Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Design of Glass as a Walking Surface

[Subject Prev][Subject Next][Thread Prev][Thread Next]


As Jordan notes, glass design has to deal with the fact that glass is a very brittle material.  I don't know much about crack propogation, but the little I do know about glass design is to make it tolerant of cracking.  Practically, what this means is to favor uniform support on all sides, so that one crack does not constitute failure but rather serves as a warning to replace the pane that cracked.

Actually, that's all I know about glass design, but if you can provide such support your design should be safe.  Of course, you still have to deal with issues such as scuffing (if you choose to use plastics) and finding an appropriate degree of friction so people don't slip.


Mike Hemstad

-----Original Message-----

From: Joseph Harouni [mailto:jharouni(--nospam--at)]

Sent: Monday, September 26, 2005 11:28 AM

To: seaint(--nospam--at)

Subject: Design of Glass as a Walking Surface


I need some ideas about designing glass as a beam to support a

walking surface. Any ideas and or references?


Joseph Harouni


25 Message:0025 25


From: "Jordan Truesdell, PE" <seaint1(--nospam--at)>

To: seaint(--nospam--at)

Subject: Re: Design of Glass as a Walking Surface

Its been a while (over a decade) since I dealt with glass design, so I

would highly recommend checking any of the few "facts" I remember. The

modulus of elasticity is near that of aluminum - we simulated glass

(fused silica, to be exact, though BK-7, aka pyrex, is similar) contact

with oring seats using aluminum to verify deflection and clearances.

Working strength is 3-4x lower than Al...I'm guessing 4-6ksi, but I'm

not certain. The critical mode of failure is flaw growth. The analysis

program used years ago was called FLAGRO and is probably available free

from NASA, at least in source form. The most difficult part is assuming

a flaw/crack size. The program then simulated repeated loadings (as a

uniformly distributed pressure, iirc) and gives an idea of how many

cycles to failure. Hmmmm...this is for annealed glasses...might not be

applicable for tempered, which is, I'm sure what you'll need. If you

can hang out for a day, email me off list (glassinfo at truesdell

engineering dot com) and I'll look at my glass notebook in the office