Prevalence, causes, and prevention of oxygen fires

[boulderjohn]

The alternative is that I can do everything myself, but that's a large skill set and equipment/supplies investment to saw off.

Not really. Oxygen cleaning a scuba tank is not all that difficult, especially if it is aluminum, in which case you don't have to worry about rust if you don't get it dried quickly enough after cleaning.

 

[lowviz]

A picture is worth a thousand words

The 1st (passed to me) an O2 first stage that had had a flash fire whilst attached to the cylinder

I am impressed! That is real.

The last time I saw a burn gouge like that is when I had an electric stove coil arc over and eat through a cast iron skillet. (public service notice: if you see a very bright spot on any of your electric burners -replace them now.) It is following one of two failure modes, either just stops working or arcs over and blows out a shower of hot burning metal.

 

[Colliam7]

I personally think it should be inserted in to OW & TECH courses, dive safety is not strictly beneath the waves, it starts on the surface.

OW courses are already chock full of a lot of information, a not insignificant amount of which seems to go in one ear and out the other. The students are being exposed to a lot in a relatively short time frame. Better to introduce gas handling concepts and practices at the point they will actually need to know it - the Enriched Air Diver course. And, we do A LOT of discussion on the topic of oxygen safety in that course, probably more than the students would like to have.

However, the emphasis of this thread is entirely on oxygen fires and incidents, and how to improve safety in that area, specifically in technical diving, so it is probably best to focus the discussion there.

my observations of divers in my area who are tech certified and have very little to no extended knowledge on the safety and handling of oxygen rich tanks.

A good point, and that is probably a good example of the issue being raised in this thread. We spend a considerable amount of time in tec training discussing oxygen effects, on the human body, the value of oxygen and enriched air in accelerating decompression, in gas planning and management during a dive, etc. But, there really is not a lot of instruction on the actual physical safety issues involved with handling oxygen, beyond the fact that equipment used with concentrations of O2 greater than 40% should be cleaned for oxygen service. And, quite often, that cleaning, of cylinders and regulators is done by someone other than the diver, and the tech students may well 'escape' the rigors of discussions about gas handling and equipment preparation. I just happened to do my tech training in a shop that was very active in gas blending, and there was probably more emphasis on oxygen safety and handling requirements, and blending issues than would normally be found in a tech course.

But, the reality is that it is in technical diving where cylinders with 50% O2 or greater are being routinely filled (albeit often by someone other than the diver), handled, stored, used, and that is what this discussion is about - how to improve safety where such practices are routinely being pursued. The challenge is made more difficult by the fact that many technical divers 'get away with' sloppy handling practices - they never have an issue. There are more than a few technical dives who routinely blend high percentage O2 gases, in cylinders not cleaned for oxygen service, with impunity. And, those divers understandably (but, wrongly, in my opinion) see an emphasis on cleaning for oxygen service as unnecessary.

So, although oxygen-related fires may be uncommon in diving, they can be devastating when they occur (as an example: Former WRAL photographer, wife injured in explosion :: WRAL.com). And, 2airuishuman is therefore asking 'how we can improve the safety of this area of technical diving'.

 

[rjack321]

1) It removes the risk of contamination of both ends of the DIN fitting. Instead, there's a schrader valve that is capped except at the fill station.
Schrader valves are a terrible addition to an oxygen system. They are extremely fast opening with torturous flow paths

2) No risk of rapid loss of oxygen if the valve is opened with no reg attached
You do NOT want O2 rushing into a regulator to slam into a seat which is closed or off.

3) No risk of inadvertent cross contamination by using an oxygen reg on an air cylinder, or an air whip on an oxygen cylinder, or an air "tank checker" on an oxygen cylinder, because the reg and valve are integrated and the fill fittings are not compatible.
The special fill connections are not made for oxygen service
When this regulator fails you cant swap to another reg and save your dive/trip

4) Allows the possibility of setting up the O2 fill station so that there is no DIN connector, so that it is not possible to inadvertently decant O2 into an air cylinder
You can't design out proper O2 design and handling practices. If shop monkey's are doing this they should be properly trained, properly paid, or get out of the business before they hurt themselves or others.

Doesn't fix the tortuous path problem by itself but opens the door to someone doing it. Fewer design constraints and a much relaxed maximum flow rate requirement since only one 2nd stage is served and assembly does not have to support the very high mass flow rates that can be required with back gas cylinders at depth.

Creating a "special" type of fitting is exactly what the EU's M26 DIN standard was supposed to be. Widely viewed as a worthless bureaucratic joke in search of an actual problem.

 

[boulderjohn]

There are more than a few technical dives who routinely blend high percentage O2 gases, in cylinders not cleaned for oxygen service, with impunity. And, those divers understandably (but, wrongly, in my opinion) see an emphasis on cleaning for oxygen service as unnecessary.

I am going to try to write this carefully; please read it carefully. It will be easy to misunderstand what I am trying to say.

Here are some things we learn from either the Compressed Gas Association or Nitrox instruction:

• According to CGA standards, any cylinder that will hold a gas mixture in excess of 23.5% oxygen must be oxygen cleaned.
• According to pretty much all nitrox courses, if you use a tank for nitrox, it cannot be used for air.
• According to pretty much all nitrox courses, a cylinder that is oxygen clean cannot have air introduced into it. If that happens, it must be O2 cleaned again. (That means that if you have a newly cleaned tank and put nitrox in it using a partial pressure process, it will have to be O2 cleaned again before its next use.)
• According to manufacturers of scuba tank valves (such as Thermo), the valves can never be used for mixes with an O2 percentage above 40%. PSI/PCS tank inspection training explains that this is because the design of the valve does not conform to CGA standards for oxygen valves, so a tank inspector who certifies a scuba tank as ready for oxygen service must do so understanding that the presence of that valve makes that impossible.

In my opinion, all of those are over the top. If we really follow all the rules, we cannot dive with high levels of oxygen in our tanks. It simply can't be done safely.

So, in order to use high oxygen levels in our scuba tanks, we have ignore some rules that might be excessive. Once we decide that we are not going to follow the rules, we are suddenly left without official guidelines in some areas, and we have to make our own personal sets of rules.

I recently had a conversation about this with someone with whom I had had the same conversation a few years before. What he remembered from the previous talk was my saying that some of the rules were excessive and had to be ignored. As a result of that memory, he was ignoring pretty much all of the rules. His tanks hadn't been O2 cleaned in 3 years. After that, I became aware of the fact that several other people I know who do their own tank inspections and cleaning have been letting things go a bit. That's carrying things way too far, IMO.

Apparently human nature is such that if people think rules are ridiculous, they will ignore them, and once they are ignoring them, they will act as if there are no rules whatsoever. On the other hand, if they think a rule is reasonable, they are more likely to obey it. This was explained a few years ago by our Department of Transportation. There was a stretch of multi-lane expressway heavily used by computers who violated its 55 MPH speed limit by a huge amount. It was very dangerous at the speeds motorists were driving. The DOT explained that human nature when they paradoxically raised the speed limit to 65. As they predicted, average speeds dropped because drivers were more likely to follow a speed limit they felt was reasonable.

So, there is something of a paradox here--by setting standards that are so strict that they will ignored, the people making those rules have made things more dangerous. No, I do not follow all the O2 rules to the letter, because no one can. That does not mean, however, that I think anything goes. We all have to find a set of reasonable rules that we can follow and be safe.

 

[lowviz]

So, there is something of a paradox here--by setting standards that are so strict that they will ignored, the people making those rules have made things more dangerous. No, I do not follow all the O2 rules to the letter, because no one can. That does not mean, however, that I think anything goes. We all have to find a set of reasonable rules that we can follow and be safe.

So what it really comes down to is what are the informed oxygen users comfortable with?

How many of us have had a problem?

The data and stats speak for themselves. It isn't that hard to do it correctly. The exciting issues are always centered around cutting corners.

 

[rjack321]

Adapting your rules here...

• According to CGA standards, any cylinder that will hold a gas mixture in excess of 23.5% oxygen must be oxygen cleaned.

This hasn't proven necessary. "Air tanks" are quite happy to get 32% in them via continuous blending. Ignore and only demand >40% tanks and valves to be O2 cleaned.

• According to pretty much all nitrox courses, if you use a tank for nitrox, it cannot be used for air.

They assume that all "air" is contaminated. Making OCA is easy, just revise the dive shop standard to be OCA and this "problem" disappears.

• According to pretty much all nitrox courses, a cylinder that is oxygen clean cannot have air introduced into it. If that happens, it must be O2 cleaned again. (That means that if you have a newly cleaned tank and put nitrox in it using a partial pressure process, it will have to be O2 cleaned again before its next use.)

See above

• According to manufacturers of scuba tank valves (such as Thermo), the valves can never be used for mixes with an O2 percentage above 40%. PSI/PCS tank inspection training explains that this is because the design of the valve does not conform to CGA standards for oxygen valves, so a tank inspector who certifies a scuba tank as ready for oxygen service must do so understanding that the presence of that valve makes that impossible.

Solution: Quit partial pressure blending recreational nitrox entirely. Its a pita, wasteful, and creates a dual standard since you don't know if its a PP or CB process. Continuous blend all 32/36% nitrox using enough filtration to produce OCA (so your whips stay clean more than anything). Only PP technical (>40%) mixes and clean the valves more often since this is where the fires start. Quit charging $9 for the super special NASA approved nitrox Oring used in 32% tanks - that just breeds ill will. Don't gouge technical divers either, they know what an Oring costs.

Short term results:
No regular dirty air at all
OCA keeps everything (whips, DIN fittings) cleaner even if not actually getting 100% introduced to the cylinder
More resources and dedication to frequent cleaning of the valves actually getting 100% through them
Shop level honesty about what is genuinely required for recreational vs technical nitrox mixes

Long term goal:
Redesign scuba valve flow paths and seats to be more aligned with CGA standards

 

[Colliam7]

Here are some things we learn from either the Compressed Gas Association or Nitrox instruction:

• According to CGA standards, any cylinder that will hold a gas mixture in excess of 23.5% oxygen must be oxygen cleaned.
• According to pretty much all nitrox courses, if you use a tank for nitrox, it cannot be used for air.
• According to pretty much all nitrox courses, a cylinder that is oxygen clean cannot have air introduced into it. If that happens, it must be O2 cleaned again. (That means that if you have a newly cleaned tank and put nitrox in it using a partial pressure process, it will have to be O2 cleaned again before its next use.)
• According to manufacturers of scuba tank valves (such as Thermo), the valves can never be used for mixes with an O2 percentage above 40%. PSI/PCS tank inspection training explains that this is because the design of the valve does not conform to CGA standards for oxygen valves, so a tank inspector who certifies a scuba tank as ready for oxygen service must do so understanding that the presence of that valve makes that impossible.

In my opinion, all of those are over the top. If we really follow all the rules, we cannot dive with high levels of oxygen in our tanks. It simply can't be done safely.

John, I had to chuckle. I was going to bring the CGA '23.5%' rule into the discussion, but was afraid of opening a Pandora's Box of comment. But, it is very relevant to the point you make, which is a good one. So, I am glad you brought it up. If we were required to abide by that rule, use of enriched air in diving would essentially disappear. And, diving would be arguably less safe. Bullet 4, I think I understand - it is a CYA position. I routinely ignore bullets 2 and 3. In fact, I tell nitrox students that bullet 2 is NOT true, and that they should ask their fill provider whether the air they are providing in fills meets OCA requirements (quite a number of shops have fill stations that do).

I completely agree with your conclusion:

If we really follow all the rules, we cannot dive with high levels of oxygen in our tanks. It simply can't be done safely.

My point, quoted in your post, was not actually about nitrox, but about O2 filling practices at a high volume operation in a state south of me, that some consider to be among their very best friends. In fact, I have routinely used that facility for fills when diving in that state. I even use that operation as an example in my nitrox class and my Gas Blender class. A LOT of cylinders are filled with O2 there, that have never been cleaned for oxygen service. And, I am not aware of a fire / explosion at that facility, ever. Now, my cylinders always are, as are my deco tec regs. That is my choice. I think risks are low, but the 'insurance' (real or perceived) is worth it to me.

 

[rjack321]

My point, quoted above, was not actually about nitrox, but about O2 filling practices at a high volume operation in a state south of me, that some consider to be among their very best friends. In fact, I routinely go there for fills when I dive in that state. I even use that operation as an example in my nitrox class and my Gas Blender class. A LOT of cylinders are filled with O2 there, that have never been cleaned for oxygen service. And, I am not aware of a fire / explosion at that facility, ever. Now, my cylinders always are, as are my deco tec regs. That is my choice. I think risks are low, but the 'insurance' (real or perceived) is worth it to me.

Even oxyhacker recognizes that if you go slow enough and keep the heat side of the fire triangle low enough, you can basically put 100% into a gasoline container. Not really a great idea but you only have to take away one side of the fire triangle. We agonize over the fuel side and give lip service to the heat side. I have yet to see anyone in this 4 page thread even mention heat (except in the context of fast opening valves).

 

[2airishuman]

I am going to try to write this carefully; please read it carefully. It will be easy to misunderstand what I am trying to say.
[...]
In my opinion, all of those are over the top. If we really follow all the rules, we cannot dive with high levels of oxygen in our tanks. It simply can't be done safely.

So, in order to use high oxygen levels in our scuba tanks, we have ignore some rules that might be excessive. Once we decide that we are not going to follow the rules, we are suddenly left without official guidelines in some areas, and we have to make our own personal sets of rules.

My take on this is that it is incumbent upon the dive industry to come up with products that are actually compliant with the CGA rules so that these sort of dives can be undertaken in a structured manner. I think there would be fewer accidents and injuries.

There's no reason we can't use CGA540 or some other oxygen-specific valve and regulator attachment interface for oxygen cylinders. We don't have to use DIN. There's no reason why there can't be regulators for SCUBA use that actually comply with all the of the best practices and CGA rules and whatnot for use with oxygen. I realize that the wonky oversize DIN connection for Nitrox that was promulgated in Europe has poisoned the conversation somewhat but that's because it was mandated for 32%, which everyone knows is ridiculous and counterproductive. Tecreational diving has been a thing for how many decades, and this hasn't been solved?! Come on.

As for OCA (oxygen compatible air), well, the problem is that the industry pretends that it doesn't exist, because anyone who has a product that purports to produce OCA opens themselves up to product liability. Which is all a cop out, because there are industrial gas suppliers that meet the technical standards that would be appropriate for OCA.