Why not overfill Aluminum cylinders?

[Eric Sedletzky]

A question for the experts:
If an Aluminum cylinder sits unused for say a year filled to 3000 psi, is it slightly compromised at the end of that year from when it was filled? Or does the very slight compromise happen only when the tank is being filled?
Would it be prudent to store aluminum cylinders with only 200 or 300 psi in them and then fill only when you plan to use the tank?
Would that possibly extend the life of the tank?
I've heard the term "sustained load cracking".
Does this happen when the tank just sits full?

I don't own any aluminum tanks anymore, I've since discovered old steel 72's and have a whole gang of them, but I'm just curious.

Besides aluminum tanks, I know of an marine construction outfit here in tbe SF Bay area that used to have their small skiff tenders welded up out of aluminum, but after several mishaps with sudden crack failures and a few boats sinking fast enough that the workers barely got off without going swimming lead them to ordering skiffs built out of steel. The skiffs do go through torture though getting squeezed in between larger vessels/barges and used almost like fenders sometimes.

 

[scubadada]

Anyone,

Do routine overfills of AL cylinders to 10-20% over working pressure (3300-3600 psi) result in ultimate harm compared to 3000 psi fills?

Thanks, Craig

 

[lowviz]

A question for the experts:

This thread is in desperate need of one.
My familiarity with aluminum comes from an old materials science course, the instructor was all about aluminum. My second-hand "experience" comes from the tank commander in my old LDS. Many, many years of filling all sorts of tanks for a very diverse group of divers.

I've heard the term "sustained load cracking".
Does this happen when the tank just sits full?

Yes, that is the problem with the "bad" alloy. 6061 claims to not have that problem.

If an Aluminum cylinder sits unused for say a year filled to 3000 psi, is it slightly compromised at the end of that year from when it was filled? Or does the very slight compromise happen only when the tank is being filled?

Other than sustained load cracking which is a definite "Yes", I would expect a material like aluminum to continue to deform under continuous stress. "How much" is the issue and I have no idea.

Would it be prudent to store aluminum cylinders with only 200 or 300 psi in them and then fill only when you plan to use the tank?
Would that possibly extend the life of the tank?

That is what I was told to do, exactly that. Drain them down to around 500 so some other fill station doesn't have a bird when you bring them in.

Anyone,

Do routine overfills of AL cylinders to 10-20% over working pressure (3300-3600 psi) result in ultimate harm compared to 3000 psi fills? ...

My old fill-guy would flat out refuse to fill even a brand new tank to 3600. You could get 3200 out of him if you had good reason, but he didn't really like filling above 3000. I could get 3200 in a steel 72 anytime I wanted.

My guess, 3600 in your tank will do little more than shorten its very long life. But I would still advise against it.

 

[Charles2]

So I posted earlier that I commonly get AL cylinders filled 10-20% over working pressure. Are there any implications here or is this below worrisome pressures? These are cool pressures, not immediately after filling.

It is my belief that the filling of any cylinder (whether steel or aluminum) beyond its service pressure will shorten its life expectancy. I also suspect (meaning: I don't really know but I do have an educated guess) that design life expectancy is well beyond 100,000,000 cycles. As @Doppler mentioned in his post, "overfilling an aluminum cylinder reduces that number 'exponentially'". The worrisome issue becomes: How much does overfilling a tank reduce its life expectancy? Does a 10% or 20% overfill reduce it to 10,000,000 cycles or 10,000 cycles, or 10 cycles? For 3AL (aluminum) cylinders, we do know that their specification requires that three samples must be subjected to 100,000 pressure reversal cycles between zero and service pressure or 10,000 pressure reversal cycles between zero and test pressure, at a rate not in excess of 10 cycles per minute without failure. Three samples must also be pressurized to destruction and failure may not occur at less than 2.5 times the marked cylinder service pressure. These tests alone should give us all great confidence in the safety of our tanks when filled in accordance with the manufacture's instructions. However, each person will have to make their own decision about their comfort level of safety when overfilling high pressure cylinders.

 

[Charles2]

My primary computer is AI, so I have all my tank pressures downloaded. For fun, I checked the starting pressure for my last 50 dives using AL80s. The average fill pressure was 3291 psi with a range of 3032-3552 psi. I found these results very interesting, average overfill of 9.7%, peak overfill of 18.4%. Not exactly cave fills, what do folks think?

My primary computer also is AI. Thanks to your post, I went back to check my starting pressures when using AL80s and found that my highest was 3292 psi with many, many above 3200 psi. I do recall seeing 3400 to 3500 psi on my SPG (before AI) after leaving tanks out in the hot, Texas sun. I calmed down a little after deciding that I was not the only one this had ever happened to nor would it probably be the last time it ever happened to me.

 

[Charles2]

I mentioned in the blog that I even Eddy Current new cylinders...

Impressive. Have you ever had to reject any new, 6061 cylinders?

 

[Charles2]

If so, a new question arises. Is the service rating only determined by the manufactured tank’s ultimate yield strength? Or does the elastic limit of a tank's material also have something to do with the tank's service rating?

For 3AA tanks, the specification states: "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 in paragraphs (k) and (l) of this section and must be not over 70,000 psi."

3AL tanks have a similar specification: "The minimum wall thickness must be such that the wall stress at the minimum specified test pressure will not exceed 80% of the minimum yield strength nor exceed 67% of the minimum ultimate tensile strength as verified by physical tests in paragraph (i) of this section."

It is also interesting that earlier in the 3AL spec the 6061-T6 properties are specified with an ultimate tensile strength of 38,000 psi and a yield strength of 35,000 psi.

As far as design parameters go, I think that the desired dimensions, volume and service pressure are determined first and then the strength of the various materials are used to determine the final physical characteristics of the finished product.

 

[Charles2]

What is the same numbers for AL please?

The generally accepted numbers for 6061-T6 aluminum are: Yield Strength of 40,000 psi and an Ultimate Strength of 45,000 psi.

 

[kelemvor]

The generally accepted numbers for 6061-T6 aluminum are: Yield Strength of 40,000 psi and an Ultimate Strength of 45,000 psi.

Those numbers must not mean what it sounds like they mean to those of us without a metallurgical background...

 

[Ghetto Diver]

As far as metals go in general: Yield is when it starts to become elastic and stretch. Strength is when it actually breaks.
I don't know the accuracy of the numbers quoted, but these tests are conducted on a 'coupon', or sample of the subject matter. Literally, the piece is put on rack and stretched, and some guy in a lab coat observes and records.

As far as overfilling AL, I get mine +10% pretty regular. To point that if I don't, I get miffed and have a WTH expression upon viewing my gauge.
Also another nice feature of getting fills like this is you can top off your bailout and still have a full main tank.
YMMV