Offseason storage -- did I screw up?

[oxyhacker]

Not if you are talking about an aluminum cylinder - Copper actually has a higher melting point than aluminum. And not all tanks are LBB, in the US at least. Even if they are, the LBB testing is done on a cold tank so it is not certain they will behave as hoped when both heated and overpressurized.

The burst disks in the valve are made of copper and will soften and fail at a relatively low temperature (compared to the cylinder itself), regardless of internal pressure. There is no way the cylinder wall can fail due to internal pressure before the burst disk fails.

Even if there were no burst disks, scuba tanks (and generally all non-destructive pressure vessels) are designed according to leak before burst (LBB) failure theory.

 

[captndale]

I personally observed the results of a fire on several tanks. The tanks in question were all loaded aboard a boat for a dive which was to take place the next day. There were both aluminum and steel tanks. All were full and all had proper burst disks installed. During the night a fire started on the boat and the boat was destroyed. Every aluminum tank exploded. On the aluminum tanks none of the burst disks ruptured and none of the valve o-rings were extruded. On the steel tanks, all of the burst disks ruptured and none of the tanks exploded.

 

[elmer fudd]

One of the local dive shops in my area has a couple of ruptured aluminum tanks on display that were recovered from a marina fire. They definitely exploded.

 

[Spimon]

Not if you are talking about an aluminum cylinder - Copper actually has a higher melting point than aluminum.

Yes. 'Melt' probably wasn't the most appropriate word. 'Undergo heat soak' might be better. The burst disks are so thin you'd expect that the thin copper to yield before the thick tank...or at least you'd hope so!

And not all tanks are LBB, in the US at least.

That's concerning!

Even if they are, the LBB testing is done on a cold tank so it is not certain they will behave as hoped when both heated and overpressurized.

There's always uncertainty in fracture mechanics.
Are the tanks not tested at a lower temperature to induce hardness/brittleness? Is it unreasonable to expect a higher temperature tank to demonstrate an even more exaggerated LBB characteristic?

 

[Luis H]

The way I interpret it for this application; the leak before break (burst) really means that the material is not brittle and will not fragment (or break into many fragments like a fragmentation grenade).

Every ruptured steel cylinder that I have seen looked like it failed in one location and then continue to rip from there. This is typical fracture mechanics for a ductile material. The failed steel tanks that I have seen (mostly pictures, but two actual cylinders, long after the fact) stayed in one piece. But, the fracture was not just a small gap. Once a fracture (or crack) is initiated, there is enough stored energy to continue to rip the tank open (some more than others).

Since the steel tanks tend to stay in one piece, the damage is somewhat controlled. You just hope not to be in the way of that one piece.

Aluminum cylinders are totally different. The yield strength of aluminum is much lower than the ultimate strength (it yields or lets say, deforms, before it fractures), making it a ductile material. But, when it does fracture, it tends to fragment like a brittle material. This is in part due to the geometry of an aluminum tank (the heavy wall thickness).


The theory that a scuba tank should not be stored half full due to fire sounded good, but I have not seen any test that confirms that theory (either control test or anecdotal observations). I have done some simple structural engineering calculations that do not support that theory. My calculations tend to predict that in the case of a steel tank the burst disk will always fail (with any initial tank pressure) before the structural integrity of the cylinder is compromised (at least for 3AA steel tanks). I plotted curves of tank pressures in a fire (as a function of temperature) versus the structural strength at the same temperatures.

I did not do calculations on aluminum tanks, but I suspected that they would be structurally compromised at a very low temperature. Captndale observation confirms that.

BTW, the theory that the tank O-ring will extrude in a fire and relief the pressure is total non-sense. A properly seated tank valve creates metal to metal contact. There is no gap for any O-ring to extrude (even if it melted, which it doesn’t). Even if a leak is generated, it will not relief pressure fast enough to amount to anything. This confirmed by observations and analysis.


To the OP, your tanks will be fine, but IMHO, you just wasted a lot of good compressed air. Any amount of air pressure will keep moist ambient air from getting in.
I suggest that next time just save the air until the next dive and then have them VIP.

 

[scottm]

To the OP, your tanks will be fine, but IMHO, you just wasted a lot of good compressed air. Any amount of air pressure will keep moist ambient air from getting in.
I suggest that next time just save the air until the next dive and then have them VIP.

Thanks all! I'm glad an unnecessary waste of time and gas at least sparked a fascinating discussion. If it's any on consolation, none of the tanks had enough gas for a dive (I suppose the doubles could have piddled around the quarry a bit).

 

[duckbill]

Since the steel tanks tend to stay in one piece, the damage is somewhat controlled. You just hope not to be in the way of that one piece.

Like our LDS, don't remember the particulars, probably wasn't told the whole story anyway- seems like a burst disc should have been involved somehow but wasn't, but the owner wasn't paying attention while filling and the cylinder blew. The fill room sustained quite a bit of damage I hear, but the cylinder, all one of it, ended up on the other side of the highway behind the shop. Thank goodness no person or car was in its path!

 

[aquaregia]

I'm now terrified of the Luxfer 80 that's stored under my bed. On the other hand, if my bed's on fire, I'm probably not in it (or I'm beyond caring).

 

[Davidstealey3]

the only two things i have not read in this post are as follows.
1) he said he was in more of a rush this year and drained them faster. i hope this fater draining of the air did not build up moisture on the valve and tank which coul;d of allowed moisture into the tank. but since they will only be stored for a short time this springs vis should be able to fix any moisture problem that occurs.
so remember to drain them slow even if you are in a rush. wait for a day you have time to do it.

2) store your tank standing up. in the case moisture did enter the tank the bottom is thicker and will be able to withstand the moisture a little better. if you lay it on the side you risk some broadspread or line corrosion. the sidewall tollarence is not as great as the bottom of the tank and may cause for the tank to be condemed.

just two bit of info to the OP i did not see in the above post. which i thought for sure some of our resident tank experts above would of mentioned

 

[Luis H]

the only two things i have not read in this post are as follows.
1) he said he was in more of a rush this year and drained them faster. i hope this fater draining of the air did not build up moisture on the valve and tank which coul;d of allowed moisture into the tank. but since they will only be stored for a short time this springs vis should be able to fix any moisture problem that occurs.
so remember to drain them slow even if you are in a rush. wait for a day you have time to do it.

Moisture on the outside of the tank is not an issue... this is a Scuba tank.

As long as you don’t empty the tank "all the way", it doesn’t matter how fast you drain it. The moisture outside is not going to enter inside as long as there is air pressure to flow from the inside out.

As long as the pressure of dry air inside is higher than the outside pressure, moist air is not going to flow in.

The moisture outside obviously doesn’t mater… heck you can even open the valve while the tank is underwater as long as there is more pressure inside the tank than outside (you are not going to find more moisture than that). Many divers have taken breaths directly out of a tank valve underwater without a regulator... just by opening the valve underwater.


Added:
It is important to just blow some air out just before connecting the filler whip. This will clean and dry the valve connection (if it ever got wet). If the valve connection is dirty or wet, the flow of high pressure air (from the filling station) will push any droplets of water (etc.) into the tank.