[Dragon1291]

So, in the BST, there are the occasional sales where the seller will say that there are cracks in the tank (Candy coating), and follow it up by "But it doesn't leak." I'm just thinking, if a 4500 PSI tank did have a "leak", would the tank be compromised enough that explosive decompression would occur? Or is there enough structural integrity left that the air would simply leak out (at an accelerated rate none the less)

I'm just thinking that all that force on a localized weak spot will just spell "boom"

[cockerpunk]

most of the time punctured pressure vessels turn into rockets. i would suspect that any puncture would cause this.

[Snipez4664]

Not really - that assumes a directional force, like the reg coming off. Then you've got a rocket. The common catastrophic failure for pressure vessels is cracking, without fragmentation (not a grenade, but still loud and scary)

[cockerpunk]

Snipez4664, on 11 September 2012 - 10:35 AM, said:

Not really - that assumes a directional force, like the reg coming off. Then you've got a rocket. The common catastrophic failure for pressure vessels is cracking, without fragmentation (not a grenade, but still loud and scary)

i would think a crack would provide enough direction to give you a pretty decent missile.

[andrewthewookie]

We've got an ASA we use to degas all the rental tanks at my field, and when you screw it on enough, the tanks are really trying to jump out of your hand. I would think a crack or rupture that allows for the air to leave fast enough would cause the tank to take some kind of trajectory.

Also, as long as the cracks are just in the vanity coating on the outside, and not the actual resin/fiber wrap, I think it should still be fine.

[dertydood]

i imagine it would be like a balloon. if you just stab the side, it simply pops. if you blow it up and release the air from the fill hole, then it will go flying away.

[Troy]

I doubt that Lurker meant that if a tank actually cracked that it wouldn't become a projectile... it would just have a very unspecific and unstable trajectory, which would differ considerably from the reg failing. I would still worry about it hitting something, but with the impulse vectors differing wildly, it wouldn't be capable of covering the same distance.

Just as an FYI, when carbon bikes came into my shop with scratches, I usually used fingernail polish enamel to seal up the hole, to stop water penetration, and stave off any crack propagation. If you are REALLY concerned about the finish of your tank after you've dinged it, you can do like we did on 5-10k+ bikes, and use a bit of toothpaste and dental floss to polish it like new. The toothpaste has just enough grit to make a really fine polish.

This post has been edited by Troy: 11 September 2012 - 11:55 AM

[HeroForADay]

I don't know about catostrophic failure, but I have seen a tank that was basically set up to fail at a hydro shop. It had an apparent stress crack in the wall, and the side of the tank surrounding the wall blew out bending the metal all the way around the circumference of the tank. Essentially it wrapped around the tank, and sent a few small fragments around the test chamber. It didn't fully fragment if that's what you're wondering, didn't end up like a real bomb, but I imagine it's possible under specific circumstances. Much like the old diesel engine-esque oil in the tank.

[Egomaniacal]

Even if it doesn't become a projectile, one thing to worry about is the force of the air escaping. Any kind of leak - even a small one - from a 4500 psi pressure vessel is enough to literally blast away any skin or other soft tissue that is nearby.