ScubaRon:
The only independent study on this subject that I am aware of was conducted by NASA. Unfortunately this was not representative of the situation that recreational/technical divers are in. So my advice is to O2 clean your reg before switching from standard air to any nitrox mix.
You obviously didn't read the NASA study.
I did.
NASA tried to force ignition in GROSSLY contaminated regulators, by hooking them up to high pressure gas sources with extremely-fast-acting (pizeo-actuated) electrical valves that could pressurize the line leading to the regulator in a few milliseconds.
This is much faster than you can do so with a manual valve, even if you slam it open, and is HUNDREDS of times faster than you will do so if you have a brain and open valves slowly with the purge depressed when you do so.
They were very interested in this for all the obvious reasons - NASA uses O2 all over the place, from aircraft to spacecraft to training purposes and on and on and on. They were caught offguard during the Apollo 1 fire - nobody thought it was as dangerous as it was to have 100% O2 in there, even at 1 ATM, and they found out exactly how bad it can be, as the materials that killed the astronauts would not have burned in such a fashion in atmospheric air. Now they're anally careful about such things, and for good reason - three men died as a consequence of not understanding where the line was, and what the risks were.
The speed of cycling is important because adiabatic heating is dependant on the
rate of rise of pressure at any given location. Anywhere that there is a constriction (e.g. quick change of direction, orifice, etc) in a gas path you can get localized adiabatic heating. Basic chemistry of fire tells us that you need heat, fuel, and an oxidizer to get a fire.
There is
always fuel in a regulator in the form of seats and O-rings, and those little nooks and crannies are GREAT little places to get quick changes in direction of the gas (specifically, a nice dead end!) Think those "oxygen compatable" O-rings won't burn in pure O2? Think again. Even at 1 ATM, they will. Try it with a gas welding outfit if you'd like - make sure you have somewhere safe (e.g. ceramic that is rated for that level of heating, a welding backstop, etc, and a fume hood!), turn on the Oxygen, bathe the piece in it (atmospheric pressure only!) then try to light it with a common long-stick butane lighter (e.g. a BBQ lighter). Poof goes the O-ring and/or seat!
Next contestant please!
(BTW, brass, the common regulator body metal, DOES NOT burn in pure O2 at pressures up to about 10000 psi. That's one of many reasons why its a great metal to use for regulators!)
The only difference between the O2-compatable parts and non-O2-compatable ones is the ignition temperature. Its HARDER (requires more heat) to get the O2-compatable ones lit, but once they are.....
Oh, BTW, adiabatic heating is quite capable of producing several hundred degrees of temperature in small localized places. More than enough to touch it off.
Anyway, what NASA found was:
1. Below 50%, there was almost no difference between air and EANx in terms of ability to force ignition. The common 40% recommendation was drawn from this, in part, as its a decent safety margin over the 50% tests.
2. At or above 50%, the risk was essentially identical to that of 100% O2. So those who think they're increasing safety by using an 80/20 deco bottle instead of 100%, in terms of combustion risk, are deluding themselves. Both will light with almost the same effort.
3. Even GROSSLY contaminated regulators were difficult, but not impossible, to get to light off with mixes under 50%, but remember these were EXTREME conditions. Also remember that the risk was almost the same for say 32% and 21% - the PPO2 is very high in a scuba tank!
4. It is a LOT easier to get a reg to light off on high (over 50%) FO2s, to the point that care is definitely called for in that case, contaminated or not. Slamming valves on those tanks is just plain asking for it. Also, regs made of Titanium and Aluminum (e.g. the SP "UL" series, and all the titanium regs) are EXTREMELY unwise on high FO2 mixes, as the METAL will burn, not just the O-rings and seats! Get one of THOSE lit and it will NOT be a flash fire that is self-limiting due to exhaustion of fuel - it will instead be a conflageration. Indeed, one medical regulator company that made med-grade O2-regs out of Aluminum was forced to recall them after a number of them ignited in use (!)
From the data there is no particular reason to believe that there is significant risk from mixes in the recreational venue (< 40%) used with conventional regulators. There IS a risk with non-O2-clean tanks and valves if you are PP mixing, since you must put in 100% O2 before you top the tank with air, and during that process you can initiate a flash fire with disasterous results. However, that risk is easily controlled - no heat, no fire, and the way to avoid heat is to put that O2 in S.L.O.W.L.Y.
The CYA language is still there, but its like the language that warns you that you can fall off a ladder and break your neck. There are things in this world that are best met with a big fat "DUH!" when we hear 'em.
