On Wed, 7 Apr 2010 12:25:50 -0700 (PDT), harry wrote:

If you are pressure testing bottles, on NO account use air or gas, the
bottle will explode violently at some point.

This reference backs up the observation that the smaller bottles rupture at
higher pressures ...
http://answers.yahoo.com/question/in...9125950AAdiFbU

5 atmospheres is about 73.5 psi. I know that 16 ounce plastic coke bottles
are rated up to 175psi(11.9 atm). 2L bottles hold somewhat less. A coke can
holds about 100 psi(6.8atm). I don't know the rating of champagne bottles
but that the thick glass can withstand a marginally greater pressure than
the thin plastic. However the plastic will begin to stretch (audiblly) as
it nears failure and the glass will just shatter and send shards
scattering. I prefer the plastic.
Source(s):
The Coca Cola people told me... and I've also test popped a few first hand
as a demonstration as to ability of expanding gas to do work using liquid
nitrogen source in the capped bottles.

The coca cola contact is :
Gina M. L'Heureux
The Coca-Cola Company
Industry and Consumer Affairs

This answer, way too conservative, at least shows a mathematical process:
http://answers.yahoo.com/question/in...4051635AAoQDT9

Best Answer - Chosen by Voters
You would have to do a bit of research: You need to know the dimensions of
the bottle: Diameter and wall thickness.

You need to know the plastic it is made from and the corresponding tensile
strength (yield) of the material. Then you can update these calculations:

Assuming that the diameter of the bottle D=5 inches, wall thickness t =
0.025 inches, and the plastic has a yield strength of 5000 psi:

The hoop stress in the wall of the bottle = PD/2t
The longitudinal stress in the wall = PD/4t

For this pressure vessel situation, those 2 stresses are orthogonal and and
the principal stresses s1 and s2, the von mises failure theory suggests
that the stress levels are acceptable if: sqrt(s1^2 - s1*s2 +s2^2) yield,
so:

sqrt((PD/4t)^2 - (PD/4t)*(PD/2t) + (PD/2t)^2) 5000

expanding and collecting the LHS =

0.433 P D / t 5000

Filling in the example numbers, P 57.7 psi

And ofcourse with anything safety related, a safety factor should be
applied in proportion to the risk severity. In this case an exploding
bottle probably would not cause death, but could cause serious injury - A
safety factor of 5 is likely appropriate... thus, assuming the example
numbers are about right, you should not pressurize to more than 57.7/5 =
11.5 psi (this is delta compared to 1 ATM)

But, of course, these guys are the most reliable I can find:
http://home.people.net.au/~aircommand/procedures.htm