Prevalence, causes, and prevention of oxygen fires

[2airishuman]

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).

The problem with this line of reasoning is that, given enough repetitions, even careful people make a mistake. Maybe only once in 1,000 fills. Maybe less often than that. But it's going to happen.

Safety is all about putting controls in place to mitigate the effect of mistakes.

 

[uwxplorer]

That specific all-in-one reg showed earlier is a patented pain in the rear to service (check the drawings and count the O-rings). I have given up on it for my drysuit bottle. If someone wants it, it's up for grab.

Regarding O2 fills (and all that can happen to your cylinder once you are out of the shop), there is unfortunately not much more than trust. My regular shop was out of O2 before a weekend trip, so I went to another, which requested a visual inspection and O2 cleaning. Long story short, after my trip, I went back to my regular shop for a top off. While getting ready to pay my bill, we hear a hiss and then a pop... the valve main O-ring was bulging underneath... Turns out the wrong O-ring had been installed. Needless to say, I quickly disclosed the name of the shop which had botched the job, got my cylinder (re)cleaned and thanked Neptune for having spared me the responsibility of burning down a shop or worse.

Scary stuff. O2 hacker is in my bedside reading stack.

 

[lowviz]

The problem with this line of reasoning is that, given enough repetitions, even careful people make a mistake. Maybe only once in 1,000 fills. Maybe less often than that. But it's going to happen.

Safety is all about putting controls in place to mitigate the effect of mistakes.

For the unwashed masses, yes. Yes, indeed.

For the informed and diligent, no. Not at all. Safety is a state of mind supported by self-discipline training/education/understanding and a desire to learn more about the risks. Safety is not a burden, it is an easy natural thing that only gets easier as one gets better at handling the risks. I had the benefit of working at a very large company with a very intelligent approach to safety.

 

[RainPilot]

Uh, bring a spare bottle?

A spare deco tank for every one you actually need, for everyone on the team? Not going to catch on I’m afraid. A single 1st and second and SPG in your bag that can replace any one of the 5 or 6 regs on your kit or your buddies is much simpler.

 

[JackD342]

Oxygen Clean - Is a Myth • ADVANCED DIVER MAGAZINE • By Bart Bjorkman

 

[rjack321]

The problem with this line of reasoning is that, given enough repetitions, even careful people make a mistake. Maybe only once in 1,000 fills. Maybe less often than that. But it's going to happen.

Safety is all about putting controls in place to mitigate the effect of mistakes.

Yea well if you don't understand the fire triangle and that to a certain extent handling high pressure oxidants is rolling the dice you have no business in fill business.

So here's the synopsis from the advanced diver article:
We must understand that oxygen clean is important, but it plays only a partial role in the big picture. The fraction of oxygen, maximum pressure, gas velocity, temperature, material compatibility, equipment design, contamination and ignition sources are all inter linked. If the fraction of oxygen is high, insure that proper equipment design is utilized, the oxygen service is current, keep pressures and temperatures low and open valves slowly.

Fraction of O2? How do you propose to reduce that? I proposed to cease PPing recreational nitrox to cut the numbers of cylinders exposed to >40% way down in the first place. CB all recreational nitrox (membrane or stick doesn't matter)

Maximum pressure? I don't know anyone filling O2 over 3,000 psi even if their booster can do it

Gas velocity? Long term we need a new scuba valve seat design

Temperature? Reduce the number of >40% fills which are the only ones PPed in the first place and you don't have to rush so much with the remaining cylinders. Temp is directly related to fill rate, but the allowable fill rate (in L/min) is cylinder size dependent. This requires a knowledgable human

Material compatibility? Stop agonizing over Orings and cleanings in a 32% cylinder, they aren't at fire risk. Cease using non O2 lubricants in dive shops altogether. If only O2 compatible lubricants are sold and used in the shop there's less risk of mistakes in lube choice. Regs do fine with cristolube/triolube.

contamination sources? Making OCA is easy. Stop the idea of having 2 air standards and that "some" hydrocarbons are ok in "air only" cylinders. If everything is OCA there's less contamination to share around cylinders and whips. Charge people a fair price to O2 clean their valves frequently (annual would be good). Tanks don't need annual cleanings or overpriced "nitrox visuals" usually and tank cleanings could be biannual (still an annual crack/corrosion vip just knock off the stupid idiotic upcharges)

ignition sources? Comes back to equipment redesigns and operator knowledge per above

But really if you go into any gas supplier 99% of the guys handling ten of thousands of Ls of O2 per day have no clue about this stuff. The gas industry is even more moronic than dive shops in their "rules" and pricing. (see john's example of greasing the helium supply bottle nipple) So there's little or no chance of any practical changes happening unless they are at the individual level.

 

[Colliam7]

So here's the synopsis from the advanced diver article:
We must understand that oxygen clean is important, but it plays only a partial role in the big picture. The fraction of oxygen, maximum pressure, gas velocity, temperature, material compatibility, equipment design, contamination and ignition sources are all inter linked. If the fraction of oxygen is high, insure that proper equipment design is utilized, the oxygen service is current, keep pressures and temperatures low and open valves slowly.

Agreed. I thought the summary - 'Where does this leave us' - was excellent.

My personal 'take-away' from the article included the above text in the first paragraph, AND the third paragraph (formatting added for emphasis): 'If you approach gas blending from the position of controlling the variables that you can, such as system design, oxygen service, gas velocity, air purity, pressure and temperature, then the outcome will be based on science, not guess work.'

So, oxygen service is one - of several - elements. It is not the be-all / end-all by itself (hence the title of the article). But, it is one important element, and one that we can influence.

Making OCA is easy. Stop the idea of having 2 air standards and that "some" hydrocarbons are ok in "air only" cylinders. If everything is OCA there's less contamination to share around cylinders and whips. Charge people a fair price to O2 clean their valves frequently (annual would be good). Tanks don't need annual cleanings or overpriced "nitrox visuals" usually and tank cleanings could be biannual (still an annual crack/corrosion vip just knock off the stupid idiotic upcharges)

Agreed! Frankly, any competent fill station has (or should have) procedures and equipment (filters) in place to produce OCA. As an aside, what is a 'nitrox visual', anyway? I get the impression from several posters that some shops are charging a bit extra for doing visuals on nitrox-wrapped cylinders, and I cannot understand the basis for doing so, or what they would do differently.

Either the valve and cylinder are cleaned for oxygen service, or they are not (what happens to the cylinder afterward is another discussion, entirely). But, I can think of nothing that can really be done, short of that, that would constitute a 'nitrox visual'. When we wrap a cylinder with an enriched air sticker, we do just that - we wrap the cylinder. We don't clean it in any way (or if we do, we are fooling ourselves, and the customer who pays an extra fee).

But really if you go into any gas supplier 99% of the guys handling ten of thousands of Ls of O2 per day have no clue about this stuff. . . . . So there's little or no chance of any practical changes happening unless they are at the individual level.

Yes! I can clean a cylinder and valve for oxygen service, test for contaminants, etc. (I won't even go into the possibility of false negatives associated with that testing. ) But, I have zero influence, much less control, on owner handling of the cylinder, or fill station practices.

Now, that realization comes with a dark side, to which I alluded earlier. If such cluelessness is pandemic, why are there so few fires / explosions / catastrophes?

 

[lowviz]

Now, that realization comes with a dark side, to which I alluded earlier. If such cluelessness is pandemic, why are there so few fires / explosions / catastrophes?

IMHO, those who do handle pure O2 regularly treat it with a bit of respect.

If anyone needs a visual, then take a 9 V battery and touch it to some 0000 steel wool. Next, drop it into a bucket that you filled with pure O2. Things are very different in a pure oxygen atmosphere.

 

[rjack321]

So, oxygen service is one - of several - elements. It is not the be-all / end-all by itself (hence the title of the article). But, it is one important element, and one that we can influence. Agreed! Frankly, any competent fill station has (or should have) procedures and equipment (filters) in place to produce OCA. As an aside, what is a 'nitrox visual', anyway? I get the impression from several posters that some shops are charging a bit extra for doing visuals on nitrox-wrapped cylinders, and I cannot understand the basis for doing so, or what they would do differently.

I haven't had a shop visual in many many moons but around here they charge more for a nitrox visual. Waving a black light around is not doing anything useful even if they charge $5-10 more

Now, that realization comes with a dark side, to which I alluded earlier. If such cluelessness is pandemic, why are there so few fires / explosions / catastrophes?

Luck? O2 isn't "that bad" unless you really get all the pieces of the fire triangle in place?
NASA had years of successful shuttle launches, until they didn't. They also lost the crew of Apollo 1 to an O2 fire decades before that but after many successful Gemini launches.​

 

[TrimixToo]

Yea well if you don't understand the fire triangle and that to a certain extent handling high pressure oxidants is rolling the dice you have no business in fill business.

So here's the synopsis from the advanced diver article:
We must understand that oxygen clean is important, but it plays only a partial role in the big picture. The fraction of oxygen, maximum pressure, gas velocity, temperature, material compatibility, equipment design, contamination and ignition sources are all inter linked. If the fraction of oxygen is high, insure that proper equipment design is utilized, the oxygen service is current, keep pressures and temperatures low and open valves slowly.

Fraction of O2? How do you propose to reduce that? I proposed to cease PPing recreational nitrox to cut the numbers of cylinders exposed to >40% way down in the first place. CB all recreational nitrox (membrane or stick doesn't matter)

Maximum pressure? I don't know anyone filling O2 over 3,000 psi even if their booster can do it

Gas velocity? Long term we need a new scuba valve seat design

Temperature? Reduce the number of >40% fills which are the only ones PPed in the first place and you don't have to rush so much with the remaining cylinders. Temp is directly related to fill rate, but the allowable fill rate (in L/min) is cylinder size dependent. This requires a knowledgable human

Material compatibility? Stop agonizing over Orings and cleanings in a 32% cylinder, they aren't at fire risk. Cease using non O2 lubricants in dive shops altogether. If only O2 compatible lubricants are sold and used in the shop there's less risk of mistakes in lube choice. Regs do fine with cristolube/triolube.

contamination sources? Making OCA is easy. Stop the idea of having 2 air standards and that "some" hydrocarbons are ok in "air only" cylinders. If everything is OCA there's less contamination to share around cylinders and whips. Charge people a fair price to O2 clean their valves frequently (annual would be good). Tanks don't need annual cleanings or overpriced "nitrox visuals" usually and tank cleanings could be biannual (still an annual crack/corrosion vip just knock off the stupid idiotic upcharges)

ignition sources? Comes back to equipment redesigns and operator knowledge per above

But really if you go into any gas supplier 99% of the guys handling ten of thousands of Ls of O2 per day have no clue about this stuff. The gas industry is even more moronic than dive shops in their "rules" and pricing. (see john's example of greasing the helium supply bottle nipple) So there's little or no chance of any practical changes happening unless they are at the individual level.

Great summary. I was going to write something pretty close to this earlier today but I didn't have time and you probably did a better job anyway.

On my part, I try to control these variables as best I can.

Of all these things, temperature is my greatest concern. Not for the tank or the valve, because my O2 fills are slow, but for the outlet of the booster. When it gets warm to the touch, I pause and let it cool. Even at cycle rates of 30/minute there is significant heating due to compression when there is a significant differential between inlet and outlet pressures at the booster.

I haven't blended nitrox except for 50% for quite some time--I'd rather just do the deco. All my blending is PP blending and I fill my own 100% tanks too. I'm certified both as a blender and O2 tech. The air from my compressor system is always tested for OCA, and when getting fills away from home (rare; I have a lot of tanks) I get OCA. All my O2 cylinders are LP, and every single valve and cylinder I own except for dry suit inflation tanks is O2 clean. There are flow restricters on my O2 tank and my fill whips. The only lubricant in my valve/reg rebuild box is Crystolube. The only O-rings that are anywhere near my dive stuff are Viton. (At the other end of the garage are BUNA-N assortments and silicone grease for the cars and such.) I buy UHP (aka Grade 5.0) helium because it's the lowest grade with a guaranteed hydrocarbon impurity specification that meets the O2 clean standards I was taught available locally. Many of these things are not necessary, really, but they reduce the probability of me screwing up. I'm all in favor of that.

Nonetheless I worry about handling 100% O2. The tank being filled is always outside, along with the booster if it's in use. "Just because you're paranoid doesn't mean it's not out to get you." I try to stay paranoid. So far, that's served me well.