Glass viewport at vacuum chamber

Uncle Al gave very good advice.

Please consider a much smaller window, or no window at all. A dome will be stronger than a flat window. (Domed to put the glass in compression, NOT IN TENSION.)

You ask a question with serious consequences, as failure of this window could cause serious injury, or even death.

Do not rely blindly on the information you get here. Your design should be reviewed by someone at your institution (a licensed engineer, or a safety review committee, preferably both) before you try it. Also try contacting vendors. You're talking about an expensive window, so they should be willing to provide assistance.

I remember years ago that a manufacturer of optical windows (can't remember name, sorry) had a product brochure with a table showing required window thickness versus diameter and pressure difference. Later versions of this brochure omitted the table -- I suspect for liability reasons.

[Along similar lines, all I'm trying to do here is point you in the right direction. I'm not designing this window for you, and I assume no liability. Your final design should be reviewed and blessed by someone qualified to do so.]

Normally, the way you would design something like this is that you would get a formula for the stress in the window from a source like Roark's Formulas for Stress and Strain, and then compare the calculated stress to the allowable stress in the material you're using.

But, for glass, manufacturers seem reluctant to tell you what value to use for a safe working stress.

This may be partially for liability reasons, but also because the strength of glass is a complicated subject:

1. It's brittle, so failure is sudden and catastrophic. Be very generous with factors of safety.
2. Failure is a statistical process, as pre-existing cracks grow with time under the applied load. The stress required to produce failure is a function of how long the stress is applied, and the initial surface finish.
3. Glass is subject to "static fatigue". Many people believe that glass will eventually fail under any load, no matter how small, if you just wait long enough.

Reading material:

According to : "The mechanical properties of glass differ from those of metals. The lack of ductility of glass prevents the equalization of stresses at local irregularities or flaws and the breaking strength varies considerably about a mean value. This latter is commonly found to occur at a tensile strength of about 70 kg/cm2 (1000 psi)."

According to

"The practical tensile strength of glass is about 27 MPa to 62 MPa." (This converts to 3900 to 9000 psia)

Doyle and Kahan

state: "Brittle materials such as glass do not possess a single characteristic strength. The strength of the material is dependent on the distribution of cracks or surface flaws. These factors, coupled with the inherent brittleness (cause of catastrophic or rapid failure) mean that extremely conservative design approaches are typically used for optical elements made of glass. Determining a design allowable for glass elements is critical for optical systems using relatively brittle glass types or for optical elements subject to relatively high stress levels. Rule-of-thumb tensile design strengths are typically at 1000 - 1500 psi for nominal glass materials."

The CRC Handbook of Mechanical Engineering (pg. 12-97) states: "For most types of glass a nominal strength of 70 MPa and a design strength of 7 MPa are typical." (70 MPa = 10,000 psi and 7 MPa = 1,000 psi.)

Additional considerations:

1. If your window glass is scratched, replace it. Do not use it. A surface scratch is the beginning of a crack.
2. Compression is much better than tension.
3. Pay attention to how you support your window at the edges. Once we built a pressure vessel with a window. To withstand the pressure we had bought a window made of sapphire, which has an incredibly high strength. But, we broke the window as we were tightening the flange to hold it in place. Apparently we had some direct sapphire-metal contact, and, due either to a small burr or pinching as we tightened the flange bolts, we broke the window. You'd like to clamp this window uniformly around the edge, with some sort of compliant gasket or o-ring between the "glass" and the metal.

MOST IMPORTANT: I apologize if I'm wrong, but your question causes me to think that you've never had any training or experience in stress analysis. Particularly in a situation where failure could cause serious injury or death, I can't think of a worse situation than having someone who doesn't know what they're doing relying on advice from a newsgroup. Get help from a qualified professional. Then have another qualified professional review the design.

Olin Perry Norton