Why not overfill Aluminum cylinders?

[Charles2]

Question: is it REALLY the case that the AL 100 is made from a different alloy such that it's rated to 3300 psi ? I find it hard to believe, given the rarity of the AL 100, that it's produced with a different alloy, but perhaps that's really the case ?

If the AL100s are manufactured to the same 3AL spec that the AL80s use, then the aluminum alloy is the same -- 6061-T6. It's the only alloy allowed by the specification.

 

[Compressor]

Thank you Captain.

 

[Bob DBF]

a US 72 is only 72 at 10% overfill

More like 71.2 cuft, but close enough for the example.


Bob

 

[lowviz]

Easier for me to remember.

How is that number determined? I've aways wondered...

 

[KevinNM]

Whats the relationship between oxygen use with aluminium vs air with aluminium and tank failures?

Aluminum burns in the presence of 100% oxygen. So an aluminum tank full of O2 has both the fuel and the oxidizer, it just needs an ignition source to cause excitement. And it will be very exciting. Like this: PHMSA Publishes RTI Report on the Root Cause Investigation of a Ruptured SCUBA Tank | RTI Group, LLC

Iirc, they never did find why the fire started, They did determine where it started in the tank, which was in the threads.

 

[Bob DBF]

Easier for me to remember.

How is that number determined? I've aways wondered...

Fill it with water, measure the volume of water and calculate, there are variances but none bring you close to 72. As for calling it a 72, well that does sound better than 71. Thus began a tradition of rounding up, sometimes considerably, the volume of SCUBA cylinders in the industry.



Bob

 

[Charles2]

I appreciate all of the comments and questions. I have tried to answer those that I could. Since asking the original question, I have spent some time going through my old text books as well as some newer information that's available on the web. If you will bear with me, I will try to summarize my analysis of the original question - "Since both 3AL and 3AA tanks are both hydro tested to 5/3rds rated pressures, what is the problem with overfilling an aluminum cylinder to the same proportionate amount as a steel cylinder?"

A few disclaimers:

• I do not recommend overfilling tanks beyond their service ratings. Never will.
• I am not an expert at anything except maybe breaking things.
• I do not have any special knowledge of any manufacturer's processes except what I may glean from the TV show "How It's Made".
• As stated in post #1, I apologize in advance for any ignorance I may have exhibited in this thread.

Resources used for this analysis:

• The chart from post #2 - thanks to @lowviz - Fatigue limit - Wikipedia
• 3AA Steel Tanks
• 3AL Aluminum Tanks

First, calculate the maximum anticipated tensile stresses in both 3AA and 3AL cylinders in accordance with the CFR 178.37 and CFR 178.46 at the service pressure. Then, a similar calculation at a pressure representing a 50% overfill of each cylinder.

3AA Steel Cylinder

• Service Pressure @ 2400 psi. -»- Tensile Stress = 28.1 ksi.
• 50% Overfill Pressure @ 3600 psi. -»- Tensile Stress = 42.2 ksi.

3AL Aluminum Cylinder

• Service Pressure @ 3000 psi. -»- Tensile Stress = 15.3 ksi.
• 50% Overfill Pressure @ 4500 psi. -»- Tensile Stress = 22.9 ksi.

Once the stress values are obtained, they can be plugged into the Stress vs. Number of Cycles Chart presented on the Wikipedia reference. I must point out that the Wikipedia S-N Chart is only representative of Aluminum and Steel Alloys in general and not a specific chart of 6061-T6 aluminum nor 4130-N steel; however, it is very illustrative for representation of this analysis. The Steel values are represented with blue lines and the Aluminum values with red. The lower of the respective lines represents the Service Pressure fill and the higher lines are the 50% Overfill stresses.

Not surprising is that when 3AA Steel cylinders are filled to their service pressure they seem to have an unlimited lifetime; however, when overfilled by 50% their lifetime is reduced to only 40,000 cycles. Now keep in mind that if that tank were overfilled daily it might last over 100 years.

When looking at the 3AL Aluminum cylinders, it appears that their lifetime when filled to service pressure is 200,000,000 cycles - I don't think that this is an unexpected value. It indicates a lifetime that is probably far beyond what most of us have left. The surprising part is that the 50% overfill lifetime is reduced to 2,000,000 cycles; or, if filled 10 times daily, it would have an expected lifetime of over 500 years.

I still don't recommend overfilling any cylinder, anytime.

 

[rjack321]

Aluminum burns in the presence of 100% oxygen. So an aluminum tank full of O2 has both the fuel and the oxidizer, it just needs an ignition source to cause excitement. And it will be very exciting. Like this: PHMSA Publishes RTI Report on the Root Cause Investigation of a Ruptured SCUBA Tank | RTI Group, LLC

Iirc, they never did find why the fire started, They did determine where it started in the tank, which was in the threads.

Based on the fire origin and where it did/didn't go, rotation friction of the valve getting turned in the threads has long been blamed for this O2 fire. The fire led to rapid cylinder gas expansion and heat that caused the valve to blow out the top and the aluminum at the crown to fail and the crack then propagated down the sidewall.

There's just no way to prove that conclusively as the only person who (presumably) had a hand on the valve at the time died.

 

[dumpsterDiver]

how do you turn a valve when the tank is filled? Wouldn't it be really tight? Like 2000 lbs of force on the valve - resisting rotation?