How To Open The Tank Valve

[Angelo Farina]

I recently heard a long time PADI CD say that through the 1970's, all agencies firmly believed that everything you really needed to know about scuba was in the OW class as it was taught by all agencies. There was no real need to teach anything more. This was all before tech diving and all the other stuff we know about now.

I do not know about how PADI was behaving in the seventies, but when I made my first OW course with CMAS, in 1975, the course was lasting 6 months, and did include training with CC rebreathers (ARO, pure oxygen, but just in the pool), twin air tanks with double complete regulators, and full deco theory with multiple stops (using US Navy tables). There were no specialties, and at the end of the course each student was evaluated and given either a one-star or a two-stars (AOW) certification. I got the two-stars, which did enable me diving down to 30m, with a buddy, within NDL.
The following year I made the "second level" course (3 stars), lasting "just" 3 months, which enabled me to dive CC pure-oxygen rebreathers down to 10m and air tanks down to 50m, with deco (not exceeding the 9-meters stop) and with more than one dive per day (which was considered far more dangerous than deco, also requiring the use of the very complex tables for repeated dives). This was fully recreational at the time.
Tech was considered below 50m and with deco starting at 12m stop or more.
CMAS was not providing training for tech, you had to go to a professional school for commercial tech divers, such as Istituto Rossi in Vicenza, etc.. In those institutes, the tech course was one full year!

 

[ToneNQ]

Do you feel the same about O2 tanks? "turn mine tank valves on slowly, but I'm a long way from convinced it matters". I'm honestly curious if you think it matters.

I've never felt a regulator get warm, but to me it makes sense to turn slow as opposed to fast. Gives a little time for everything to expand / react / adjust for the added pressure, and maybe ourselves a split second to quickly turn things off should something pop.

TL;DR - Deco bottles I certainly turn on gently. Partly because that is the teaching, partly because its crude to beat on your gear. I'd love to understand the precise issue/s but its not an orphan idea floating about the industry with no basis in any teaching material. I think cleanliness is by far the most critical thing, so much evidence on this issue compared to the others.

Long answer - from the training course literature I have, both DAN and TDI teach to turn O2 valves slowly. Both are quite clear on this, see below for references. If I recall high school and other industry training, flammable things have an auto-ignition temperature. In the case of vapours there are also flash points and upper and lower explosive limits. Changing the oxygen concentration moves those numbers, so what is barely flammable at standard temperature and pressure can become quite flammable. Apollo 1 is a horrific example. Whilst it reduces risk to turn it on slowly, both sources allude that the change in the denominator may be small. Dedicated O2 clean gear is best practice, where there is no fuel there can be no fire.

The DAN 'Oxygen First Aid' manual (p77) does say to that "as a safety precaution, it is a good idea to have the flow valve open prior to pressurising the regulator. This will reduce the initial pressurisation and lessen the (already small) risk of ignition". Towards the end of the paragraph it says "the cylinder valve is opened slowly so that the regulator is pressurised slowly". Also (p74 and 75) does have pictures of a 1st stage reg with fire damage on the HP side. It is a yoke reg with two LP and one HP hose, its hard to know if its a deco reg or an air reg used of O2. The paragraph says to 'not use scuba regulators for administering oxygen since, if not properly designed and machined, they can cause excessive heat to be generated when oxygen flows rapidly through them'. To me that suggests the failed reg was not due to valve opening speed but improper use, but that is speculation. Oxygen is more dense that air, and pipe friction is the (square?) of velocity, so heat generation increases exponentially with flow. Considering the fire tetrahedron, high O2 concentration plus heat and all we get a chain reaction where all sorts of things become flammable.

TDI say (Adv Nx, Ch 5, cylinder cleaning) that "the use of oxygen and high oxygen concentration mixtures does carry some increased risk of equipment related issues. Always turn valves on slowly. Make sure to maintain the level of cleaning present in dedicated gear." In the same chapter they also state that "titanium and aluminium regulators should never be the choice for this application."

 

[NothingClever]

Honestly if your regulator is so far gone that opening a tank valve quickly will fail it, you really should not be using it for life support. Given the condition of some shops training/rental regulators, it may have been taught so they could use there junk one more time.

I was trained to “soft start” my regs but your point above is brilliant. No more.

 

[rob.mwpropane]

TL;DR - Deco bottles I certainly turn on gently. Partly because that is the teaching, partly because its crude to beat on your gear. I'd love to understand the precise issue/s but its not an orphan idea floating about the industry with no basis in any teaching material. I think cleanliness is by far the most critical thing, so much evidence on this issue compared to the others.

Long answer - from the training course literature I have, both DAN and TDI teach to turn O2 valves slowly. Both are quite clear on this, see below for references. If I recall high school and other industry training, flammable things have an auto-ignition temperature. In the case of vapours there are also flash points and upper and lower explosive limits. Changing the oxygen concentration moves those numbers, so what is barely flammable at standard temperature and pressure can become quite flammable. Apollo 1 is a horrific example. Whilst it reduces risk to turn it on slowly, both sources allude that the change in the denominator may be small. Dedicated O2 clean gear is best practice, where there is no fuel there can be no fire.

The DAN 'Oxygen First Aid' manual (p77) does say to that "as a safety precaution, it is a good idea to have the flow valve open prior to pressurising the regulator. This will reduce the initial pressurisation and lessen the (already small) risk of ignition". Towards the end of the paragraph it says "the cylinder valve is opened slowly so that the regulator is pressurised slowly". Also (p74 and 75) does have pictures of a 1st stage reg with fire damage on the HP side. It is a yoke reg with two LP and one HP hose, its hard to know if its a deco reg or an air reg used of O2. The paragraph says to 'not use scuba regulators for administering oxygen since, if not properly designed and machined, they can cause excessive heat to be generated when oxygen flows rapidly through them'. To me that suggests the failed reg was not due to valve opening speed but improper use, but that is speculation. Oxygen is more dense that air, and pipe friction is the (square?) of velocity, so heat generation increases exponentially with flow. Considering the fire tetrahedron, high O2 concentration plus heat and all we get a chain reaction where all sorts of things become flammable.

TDI say (Adv Nx, Ch 5, cylinder cleaning) that "the use of oxygen and high oxygen concentration mixtures does carry some increased risk of equipment related issues. Always turn valves on slowly. Make sure to maintain the level of cleaning present in dedicated gear." In the same chapter they also state that "titanium and aluminium regulators should never be the choice for this application."

Thank you for that detailed response. I haven't ventured into O2 for scuba yet, but we deal with a lot of it at the shop. It's hard to rummage through "by the book" and "real life". It's good to see the pictures and hear the stories and what people actually do. Thanks again.

 

[bakodiver391]

We had a deco regulator explosion on board a liveaboard when a diver cracked the valve quickly. The explosion put out any fire but there was smoke and heat, enough to cause a slight burn on the owner's leg. Some very experienced people have told me to hold the purge down while opening O2 cylinders to lessen adiabatic heating, and this has been my practice for years. Some valves don't operate smoothly, making it difficult to pressurize slowly, so anything to mitigate the danger of O2 fire, especially on a boat, makes sense to me.

 

[Miyaru]

Some valves don't operate smoothly ...

In that case you don't frack around with pressurizing a regset while pressing the purgebutton, but you have that valve either serviced or replaced, especially with oxygen.

 

[bakodiver391]

Problem being you won't know if it's going to be a gradual opening until you have opened it, already too late. I'll stick with my method, thank you.

 

[ToneNQ]

We had a deco regulator explosion on board a liveaboard when a diver cracked the valve quickly. The explosion put out any fire but there was smoke and heat, enough to cause a slight burn on the owner's leg. Some very experienced people have told me to hold the purge down while opening O2 cylinders to lessen adiabatic heating, and this has been my practice for years. Some valves don't operate smoothly, making it difficult to pressurize slowly, so anything to mitigate the danger of O2 fire, especially on a boat, makes sense to me.

Heat is generated when gas is compressed, that is resolved during the tank fill process. The only source of heat once the tank is full, (in my thinking) are: friction from the gas flow, which increases with gas density and is highest with O2 that any sort of blend with N2 or He; and combustion of a fuel in the oxygen rich environment.

There is no heat transfer in an adiabatic process, only once heat is transferred to the regulator body and components that it affects the materials. Adiabatic heating to superheat a gas requires substantial energy, but you still only generate heat. You need to inject a fuel for adiabatic heating to become an explosion, or you need to transfer that heat to the surroundings, being an inadequate vessel which ruptures. As the gas leaves the tank it expands to fill the 1st stage volume, so adiabatic cooling is the process. Once it goes through the HP valve it expands again into the IP system and the LP hoses out to the second stage, at which point there is further adiabatic cooling. There should be friction from the gas molecules on the regulatory body, hoses and valves, due to the rate gas flow generating heat in those components. To me the only factor controlled by slow valve opening, is reducing the gas velocity on the first fill of the regulator system. Good materials engineering should negate this requirement, but its a safe habit to have if you are not always using your gear and certain that you have all the contributing factors under control.

Occum's Razor may suggest the most likely cause of the regulator explosion was the regulator being fit-for-purpose. A stage reg being repurposed as a deco reg, not O2 clean perhaps. Burns aren't likely from pressure rupture. Science is lovely because its so consistent, the illogical stuff comes from the human factor. I've seen people use silicone grease on HP spools for their deco reg SPG becuase "it was all they had". The unavoidable piece is that Oxygen does not burn, it is the oxidising agent that supports combustion. Find the fuel and you find the root cause.

 

[Angelo Farina]

Heat is generated when gas is compressed, that is resolved during the tank fill process. The only source of heat once the tank is full, (in my thinking) are: friction from the gas flow, which increases with gas density and is highest with O2 that any sort of blend with N2 or He; and combustion of a fuel in the oxygen rich environment.

There is no heat transfer in an adiabatic process, only once heat is transferred to the regulator body and components that it affects the materials. Adiabatic heating to superheat a gas requires substantial energy, but you still only generate heat. You need to inject a fuel for adiabatic heating to become an explosion, or you need to transfer that heat to the surroundings, being an inadequate vessel which ruptures. As the gas leaves the tank it expands to fill the 1st stage volume, so adiabatic cooling is the process. Once it goes through the HP valve it expands again into the IP system and the LP hoses out to the second stage, at which point there is further adiabatic cooling. There should be friction from the gas molecules on the regulatory body, hoses and valves, due to the rate gas flow generating heat in those components. To me the only factor controlled by slow valve opening, is reducing the gas velocity on the first fill of the regulator system. Good materials engineering should negate this requirement, but its a safe habit to have if you are not always using your gear and certain that you have all the contributing factors under control.

Occum's Razor may suggest the most likely cause of the regulator explosion was the regulator being fit-for-purpose. A stage reg being repurposed as a deco reg, not O2 clean perhaps. Burns aren't likely from pressure rupture. Science is lovely because its so consistent, the illogical stuff comes from the human factor. I've seen people use silicone grease on HP spools for their deco reg SPG becuase "it was all they had". The unavoidable piece is that Oxygen does not burn, it is the oxidising agent that supports combustion. Find the fuel and you find the root cause.

Exactly: no fuel, no fire!

 

[Dody]

I am surprised that nobody talks about it: do you close/ open the valve clockwise or counter clockwise? The fact that there is no clear rule is for me very hazardous especially with this 1/4 back turn where you may think it is almost fully open when it is almost fully closed and can't realize before you can't breath due to the water pressure. Why isn't there a standard way to make sure that we can't get it wrong?