For a tank in good condition the hydro pressure is well below the elastic limit.
If the hydro pressure exceeded the elastic limit no tank would pass.
Tobin
That is not correct.
During hydro the vast majority of the cylinders experience a very small amount of plastic deformation (permanent deformation). If they were “well below the elastic limit”, there would only be elastic deformation. There would be no residual permanent expansion.
I started doing hydro’s (working at a dive shop, long before I got my engineering degrees) and in my experience very few cylinders return 100% without any plastic deformation.
The only cylinder that always return 100%, all the time is the calibration cylinder used every day a hydro is going to be performed. That cylinder is designed to truly operate within the elastic range for the hydro machines operating pressures.
I have also done the stress calculations and compared the results to the published minimum yield strength for the design material. The hydro stress seems to fall at the low end of the transition in the stress-strain curve.
The yield stress of a material is not a single point. The published data is normally for the minimum, but there is actually a transition.
The purpose of the hydro test is to determine how elastic is the material condition of the cylinder and therefore the test (by design) has to stress the cylinder to the lower acceptable elastic limit. By precisely measuring how much plastic deformation occurs at that stress level, we can approximately determine the material condition.
The hydro pass-fail criteria is a relatively conservative measure intended to determine that the stress (at the test pressure) is “mostly” within the elastic range of the material. Very little permanent (plastic) deformation is allowed.
The most common steel used under 3AA code is AISI 4130. Per CFR 49 section 1783.37:
(2) For cylinders with service pressure
of 900 psig or more the minimum
wall must be such that the wall stress
at the minimum specified test pressure
may not exceed 67 percent of the minimum
tensile strength of the steel as
determined from the physical tests required
required in paragraphs (k) and (l) of this
section and must be not over 70,000 psi.
(3) Calculation must be made by the
formula:
S = [P(1.3D2+0.4d2)]/(D2-d2)
Where:
S = wall stress in psi;
P = minimum test pressure prescribed for
water jacket test or 450 psig whichever is
the greater;
D = outside diameter in inches;
d = inside diameter in inches.
For example.
The stress level of a typical 3AA cylinder using the equation shown [ S = [P(1.3D2+0.4d2)]/(D2-d2) ] is 66,322 psi.
Data: OD= 6.83 in, minimum t=0.164 in., ID=6.502 in. test pressure P=3750 psi
If I use the classic hoop stress calculation (S = P*r / t) , the stress is actually higher at 74,119psi.
BTW, I am not at work, so I do not have access to a lot of data, but a published yield strength data I found for this alloy (AISI 4130) is 63,100 psi.
The fatigue life of steel is essentially unlimited.
Tobin
This is also inaccurate or basically incorrect.
The S-N (stress versus number of cycles) curves are available for some chrome-molybdenum steel alloys, similar to the material used on pressure vessels.
If what you are trying to say is that the number of cycles in scuba cylinder application is very small and therefore its fatigue strength life seems enormous, OK, but saying that it is unlimited (or essentially unlimited), is far from the facts.
I suppose I could get the tanks UT'd for thickness to be certain …
Normally you would not need to do UT (ultrasound Testing) on a cylinder, just to determine if they are OK. The hydro test is sufficient.
I have done UT on most of my cylinders, but it was specifically to obtain the average wall thickness on a statically significant number cylinders, so that I could calculate accurate REE (Rejection Elastic Expansion) numbers for my steel 72’s.
Now that I have about a dozen cylinders measured (6 made by PST and 6 by Norris), I have fairly good REE data for PST and Norris cylinders.
With this information my LDS can “+” stamp any cylinder of the same type by the same manufacturer.
The overfills aren't outrageous just constant, usually 2000psi, 200psi over rated.
Are you taking about cylinders that are stamped 1800 psi and you are filling them to 2000psi? That is not really an overfill.
If you are talking about filling a 2250 psi (with normal 10% fill of about 2500 psi) with 200 psi extra to 2700 psi. That is OK too. That is still relatively low.
I was told by the LDS owner that he'd been filling those tanks like that for years, which is fine with me but, since that 72 failed I've been a little leery of old well worn used tanks, which is funny because the newest tanks I have were made in 1981.
Luis, that 72 saw a LOT of service. It was the only tank I owned for probably the 1st 10 years of diving even after I got another it was still used quite a bit and always over filled a few hundred psi over. That tank was filled 3 times a day 3-4 days out of 7 at least during the summer every week for 3 months. I didn't do anything but dive and work sometime both at the same time for the 1st 3 years after getting certified. That tank has more dives on it then most posters on SB do.
Let’s assume 3 dives a day, every day of the year (365 days), for 10 years. That is only 10,950 fill cycles.
Fatigue data (in S-N curves) doesn’t start to be significant until about 10 times that. That is just where it starts.
The designed working stress level (at the design working pressure) of a cylinder is around 60% of yield.
The stresses in a pressure cylinder are always positive, they are not reversing stresses (like the reversing stresses experience in a vibrating object). The fatigue life with Non-reversing stresses is higher than with the classical S-N curves, but the data is limited.
It was a Scubapro who made it for them I don't know but it was in 1968 so I'm pretty sure you know. I actually had a second different hydro facility retest it has a favor to me for confirmation and it failed again the same way. I was there this time and watched the whole process and not one step was missed.
With the limited information I have, I am going to guess that the cylinder was galvanized and that neither hydro facility actually did the pre-test load (pre-stretch) cycle. Many hydro stations didn’t used to do it (many still don’t if you don’t insist on it).
The pre-test procedure (used for galvanized cylinders) requires pre-loading the cylinder to 90%of the test pressure and holding it there for a few minutes, just before the test. For a steel 72 (stamped 2250psi) the test pressure is 3750 psi and the pre-test is 3375 psi.
The issue is with galvanized cylinders (most commercial cylinders are not galvanized). The hot dipped galvanizing zinc coating behaves like a composite material and affects the elastic behavior of the cylinder. For testing purpose, it has been determined that pre-stretching the cylinder just before testing provides more acute results.
PST and Worthington cylinder manufacturers have provided guidance on how this procedure should be performed.
