[PSUBS-MAILIST] Commercial Grade O2 vs. Medical

Thu Jan 4 19:06:46 EST 2018

I probably should have mentioned that, irrespective of the other points, a vessel subject to external pressure will always experience greater forces in the vessel wall than it would with an identical pressure acting internally, because the external pressure acts on the greater area. F = P*A and all that...

Sean

-------- Original Message --------
On Jan 4, 2018, 16:41, Alan via Personal_Submersibles wrote:

> Thanks Sean,
> guilty as charged. Although I did run similar cylinder dimensions to the
> tank sighted but guessing on the thickness.
> Alan
>
> Sent from my iPad
>
>> On 5/01/2018, at 10:34 AM, Sean T. Stevenson via Personal_Submersibles  wrote:
>>
>> Alan, I'm afraid that this is a dangerous misconception. Internal and external pressure are very different load cases. The reason you are seeing similar results is that you are:
>>
>> 1) evaluating a strongly isotropic material,
>> 2) evaluating using a thin wall assumption whereby it is assumed that the net stress in the cylinder wall is evenly distributed throughout the cross-sectional area of the wall,
>> 3) evaluating a material which exhibits effectively identical compressive and tensile material yield strengths, and
>> 4) ignoring buckling failure modes, or acknowledging them to be secondary to a dominating material strength failure.
>>
>> So, the results CAN be similar for internal and external pressure, but only under specific circumstances. It is dangerous to consider this to be a rule of thumb. As a general rule, internal pressure capacity is greater than external due to buckling failure modes.
>>
>> Any material defects or deformation from applied loading only complicates the situation, as with internal pressure, any out-of-round deformation is self-limiting, whereas in the external pressure case such deformation will serve as a buckle initiation site which will propagate unchecked.
>>
>> That said, I think that collapsing an aluminum gas cylinder with external pressure is unlikely at PSub depths, although I might be inclined to select shorter, stubbier cylinders if that is a worry, to ensure that buckling will not dominate (or at least, if it is dominant it is so at much higher pressure than the intended service pressure) as could be the case with a long, slender cylinder. Regarding flat bottoms, gas cylinders are typically extruded using a die, and the internal shape of the cylinder bottom is actually a 2:1 semi-ellipse, or at minimum, generously filleted. Extra material is added in design / to the minimum head thickness to accommodate standing the cylinder upright, creating the flat bottom. Evaluating the cylinder as an unreinforced cylinder subject to external pressure, with hemispherical or semi-elliptical end caps as appropriate, seems to me to be a reasonable approach.
>>
>> Sean
>>
>> Sent from ProtonMail mobile
>>
>> -------- Original Message --------
>>
>> On Jan 4, 2018, 13:12, Alan via Personal_Submersibles < personal_submersibles at psubs.org> wrote:
>>
>> Steve, I just ran a couple of scenarios on tubes with differing aluminium grades in the free "under pressure" program. The results were exactly the same for internal pressure as external pressure. So if your medical tank is safe for 2000 psi internal it will be safe for 2000 psi external. plus there is probably at least a 100% safety margin so more likely it will crush at 4000psi or 8000ft. Perhaps Sean could confirm that with aluminium it is at least as strong under external pressure as it is under internal, then you won't need to do any calculations.
>>
>> Cheers Alan
>> _______________________________________________
>> Personal_Submersibles mailing list
>> Personal_Submersibles at psubs.org
>> http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles
>
> _______________________________________________
> Personal_Submersibles mailing list
> Personal_Submersibles at psubs.org
> http://www.psubs.org/mailman/listinfo.cgi/personal_submersibles @psubs.org>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.whoweb.com/pipermail/personal_submersibles/attachments/20180104/14263014/attachment-0001.html>