Dive Cylinder Explodes - Sydney

Please register or login

Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.

Benefits of registering include

  • Ability to post and comment on topics and discussions.
  • A Free photo gallery to share your dive photos with the world.
  • You can make this box go away

Joining is quick and easy. Log in or Register now!

Yep. Otherwise, you are an expert in this. But just speaking from my amateur perspective, I'd one more point, though. Shrapnel is not the only thing that can hurt when a tank explodes, there is also the shock wave, like with any explosion. And even 1 ft of water should reduce its energy significantly.
Part of my current life is throwing bombs overboard to shock test Navy destroyers. According to the website How Stuff Works Is it worse to be near an explosion on land or in water? you'd sustain far more damage from a water blast than an air blast. Now, I'm no expert on explosions, (I'm just the ship's engineer), but the shock wave is not dissipated by the water except by distance, because the water in not compressible. Air is compressible, so some of the energy is used by compression of air and not transferred to the person. This assumes that shrapnel isn't a factor.

Back in the day when torpedoes were actually contact weapons, if you could fight the water coming in the boat, there would be little further damage inside the hull because of the compression effects of the air inside the hull. I have seen examples of this when diving on the U-2513. The U-boat was sunk during a navy exercise by hedgehogs, which opened a hole in the bow in the forward torpedo room. Hedgehogs were depth charges, and they were a "close enough" weapon. A fairly small charge, they were designed to produce a pressure wave which would transfer through the water and strike the hull. Since the outer hull is just ballast tanks, usually full of water, the pressure wave would be transferred to the pressure hull, or people tank. Assuming the people tank was full of air, damage would be limited to the hull. That was the idea, to fill the people tank full of water. The U-2513 is almost in perfect condition. All except that big hole in the torpedo room.
 
I'm no expert by any means, but I don't think he meant getting into the water tank with the cylinder as it was being refilled.
 
I'm no expert by any means, but I don't think he meant getting into the water tank with the cylinder as it was being refilled.
Point is the tank becomes as much a part of the explosion as the cylinder itself after absorbing the energy of the exploding cylinder. The nice thing about containment fill stations is that they direct the energy of the explosion away from the peoples.
 
My LDS refuses to fill any of these tanks period. It's a blanket policy.
I've been given many of these tanks over the years by people who got out of diving and have no idea that they are 6351 alloy and wouldn't even know what that means. They just think they are being nice by giving me free gear. I thank them and the first thing I do is drain them then drill a big hole in the neck and they go straight to recycle. I used to pass on them but now I beg to get them just so I can disable them and dispose of the properly. There are too many good tanks out there to mess with those turds.

Lead alloy? WTF!!
I'm no metalurgist, but why would they do that??
 
Last edited:
The problem is that in this particular cylinder's case, there were hydrocarbons, and with the oxygen was put it, it was starting to eat at the walls of the tank, fortunately not to the degree that the tank had to be condemned. I don't have the expertise to know exactly when ignition is going to occur, I just follow the industry safety guidelines.

Yes and I understand that and am glad you have taken that position. There will always be hydrocarbons n a tank. You cant have "NO" hydrocarbons. The key is the amount The ignition would probably occur at a right angle such as the valve when high volumn passes through it and because of the impact at the right angle generates heat during compression that can ignite. Much the same kind of thing when depressurizing moist air goes around a 90 angle and it freezes. with dry air the ignition would be the only scuba issue. One of the reasons for keeping O2 filling to 50 psi per minute. Again a global rule not specifically made for scuba application and more for liability coverage. It bothers me that the shop would say you have hydrocarbons in a tank and it could be a danger. Its playing on ignorance for profeit. As far as eating through the tank goes I can't comment on that in regards to hydrocarbons but it would be true for oxidation. The oxidation is what you tumble or whip for. Rust is like a fire triangle it needs moisture Oxygen and pressure. The higher the ppO2 the faster it rusts. one reason that you store you tanks for the winter with <50 psi in it. (Slows oxidation) Too many times the powers make things so simple that they take on new lives and no longer relate to the original concerns. An example. O2 less than 60% acts like air and greater than that pure O2. In its new life Nitrox 23.5% at >50psi requires O2 cleaning. I know you probably heard that it was 40% but that came from NOAA and rec nitrox line being put at 40%. Origional intent was no special cleaning would be needed for rec nitrox mixes. Look what we have now. BTW before rule makers came to the 23.5% it was 20% till they discovered atmosphere contains close to 21% and every gas station would have to have O2 clean tire air equipment and that there are places in the USA with higher than 21% atmospheric air. And you thought that dollar for 3 min of tire compressor air was expensive.
 
Last edited:
Yep. Otherwise, you are an expert in this. But just speaking from my amateur perspective, I'd one more point, though. Shrapnel is not the only thing that can hurt when a tank explodes, there is also the shock wave, like with any explosion. And even 1 ft of water should reduce its energy significantly.
Fragments of tanks or secondary projectiles are what kills people who are not right next to the tank. The shock wave dissipates a lot faster.

For example An M67 fragmentation grenade has a 6.5oz CompB bursting charge and a casualty radius of 15m (50% chance of death or serious injury), though fragments can be blown up to 230 meters. A cardboard MK3A2 concussion grenade has a bursting charge of 8oz TNT (roughly equiv) and has a casualty radius of 2 meters, though the few metal pieces can be thrown 200 meters or so. So if you take out the threat of metal fragments the danger is vastly reduced.

Unfortunately scuba tanks are made of metal.
 
Last edited:
Lead alloy? WTF!!
I'm no metalurgist, but why would they do that??

Guess what, there can be 13 - 16 different elements that make up a specific alloy depending on the properties you want. Lead like others would be used in a minute percentage given that around 96% of the alloy is aluminum.
 
As I've said before, if tanks are properly inspected there should be no fear.

The trouble is, how do you know?

Hydo doesn't prove anything except that the tank can accept a pressure and that its expansion is within limits. It can however cause a defect to grow. The Annual hydro test in Australia is just foolish.

You could use better methods which are less destructive to gauge the metallic condition of the cylinder - but I won't open that can of worms.

There are 2 contributing factors that will affect the success of the testing (Eddy current)

1. Operator training. In the real world, to gain a certification in Eddy current testing, you need 80 hrs of instruction (generally over 10 days) It covers the theroy of eddy current inspection and electrical theroy. You need to complete 3 practicals.

One. inspect a component following a procedure and find the defect or defects.
Two. You need write a procedure for another item chose the correct probes, settings and method of examination, and validate the procedure.
Three. Write up the procedure in that anyone trained can follow it without prior experience of the component.

They need to achieve 75% minimum in any one section but an overall for all the sections of 82% to pass.

So I'm guessing VIP inspectors don't get that training. They probably get a one day course from Visual plus and a pretty certificate.

2. The procedure itself. Just because it finds defects, doesn't mean it actually works. It's not the amount of defects it finds but the number it misses!! Below is an example which has good similarities with the scuba tanks problem

A real world example of this was in late 1993. I was working for the Aircraft manufacturer. We had three things that all coincided, that caused a big rethink.

It was post Gulf War II, the Harrier Aircraft operated by the RAF, RN and USMC were engaged in lots of operations, so the ration of flying hrs to landings had changed significantly. I received a quarterly report in which there had been 3 undercarriage failures, in each case they'd been recently inspected and passed. We know there to be a problem

I knew the inspectors capabilities, while not to the level of us at manufacturers (engaged in R&D and cutting edge techniques) never the less three undercarriages shearing off couldn't be down to inspector error.

The final bit of the jigsaw was an undercarriage leg that had a suspect defect (found by the procedure) arrive as us by mistake rather than the component manufacturer.

So I inspected it, using the Eddy current procedure. The best I could achieve was thinking I might have defected a defect. Not good give I believed one to be there so was using extra vigilance.

SO we stripped the paint off and chemically etched the metal (a procedure that couldn't be used in service). Using a dye penetrant, under UV light we didn't just see 1 crack but 9! One crack was marginally longer than the others which has caused the operator to reject the component. Good call!

But clearly the procedure was fallible. No one would believe it of course so we got all the concerned parties in a room. This was the first brown stuff hitting the fan moment. There was evidence clearly in view that the procedure didn't work to an acceptable degree. (Everyone had a go and failed)

So I developed a new Eddy current method, with different equipment, settings etc. and validated it (checking old legs making a call and then using the penetrant method to confirm the results. We had 100% success with no false positives and no missed defects.

Off we went to the field to inspect the world wide fleet. All the data calculations and statistics suggested we would find only 1% defective items. We found 47% of the fleet defective (about 400 aircraft) (the next fan and brown stuff moment)


So how does this relate to scuba tanks?

1. We had a known defect, but we had only a few failures - given flying hours etc is was a small probability
2. There was a Eddy Current inspection method that was deemed to be successful

The difference here is that we proved that while the procedure was finding defects, it was the 97% that it didn't find that was the issue.

It's always the defects you miss that cause the issues, not the defects you find
 
I'm no expert by any means, but I don't think he meant getting into the water tank with the cylinder as it was being refilled.
That's right. The shock wave in water is bad but only till it transforms into a splash.

I don't know what the TNT equivalent of an aluminum scuba tank explosion is (here is a link to engineers discussing the subject with formulas etc so apparently those who are interested can come up with an estimate), however I remember the case when a worker was decapitated by the sheer force of an air tank blast. This was not a scuba tank but a bigger industrial steel tank feeding pneumatic sledgehammers, so I do not know how fair such comparison is. But concussions and blown eardrums would be bad enough.
 

Back
Top Bottom