Mailing List Archive Mailing List Archive
On Wed, 12 Apr 1995, Ronnie Bell wrote:
> In response to the "safest" way to mix nitrox in the field,I'd suggest the
> following. Calcluate how many PSI of O2 is needed, put in as much air as
> you can FIRST and still be able to put the O2 in. For example if you are
> mixing a 32% o2 mix and plan on ending up at 3500 psi you need to put in
> about 487 psi of O2 and the rest air. So if your O2 tank has 1800 psi you
> could easily put in 1000 psi air, put in 487 O2 (ok 480) then top off
> with air to 3500.
I used to do this when mixing nitrox out in the field, but I stopped
because I don't think it's the best way to do it.
I'm assuming we're dealing with storage cylinders of oxygen, an O2-clean
oxygen decanting hose, O2-clean tanks, and air that is contaminated with
hydrocarbons, but would otherwise be safe to breathe as compressed
air...right?
The contaminants we're worried about are the hydrocarbon molecules in the
compressor air. The reason we're worried about them is that hydrocarbons
have a relatively low ignition temperature in high O2 environments, and
are thus more likly to combust. Now: it's my understanding that we're not
terribly worried about the hydrocarbons themselves combusting, because
even in "dirty" air there are not enough of these molecules to cause big
fire problems. What we *ARE* worried about is the hydrocarbon combustions
creating point-source temperatures high enough to ignite other "fuels" like
aluminum and steel (The components of the tank and valve). That's when bad
things happen.
We'll assume the tank and valve are already contaminated with hydrocarbons
from previous fills with "dirty" air. These hydrocarbons are deposited on
the internal walls of the cylinder and valve, and are floating around in
the airspace of the cylinder, but have not contaminated the O2 decanting
hose. Where is the combustion likely to take place when we start putting
oxygen into the cylinder (at any pressure)? It won't likely happen in the
atmosphere of the tank when oxygen mixes with contaminated air, because
there is nothing to generate temperatures high enough to ignite even
hydrocarbons. Even if the airborne (vapor?) hydrocarbons did ignite,
the mini-combustions wouldn't be near any fuels to start a "real" fire.
I suppose the combustion could take place on the walls of the cylinder,
but again, what would cause the temperatures needed to ignite the
hydrocarbons? The gas blowing around in the cylinder is probably not
moving fast enough to cause a flake of metal to scrape against the wall of
a cylinder hard enough to generate high enough temperatures to ignite (at
least that's what the guy from NASA at the 1989 Nitrox Workshop in Houston
told me). The place where the combustion is MOST likely to happen is
inside the valve of the cylinder, where gas velocities are high, and where
you would get the brunt of the adiabetic combustion if the valve was shut
and you rapidly pressurized the decanting hose.
So...if we take an empty cylinder and decant 480psi of oxygen (using
Ronnie's example), then we will have a maximum of 480psi pure oxygen in
the cylinder and valve. If the cylinder already had 1000psi of air in it
when we started decanting pure oxygen, then yes, the oxygen would be
diluted immediately upon entering the cylinder, and the probability of a
combustion in the *cylinder* would be reduced. HOWEVER, you would also
have up to 1480psi of pure oxygen inside the tank valve, the place where a
combustion is probably most likely to occur. Given that the probability of a
combustion increases with increased pressure/concentration of oxygen, I
believe it is always safer to put the oxygen in the tank first (minimizing
the total pressure of oxygen exposed to contaminants), then top-off with air.
Another reason I only like to put oxygen into *empty* cylinders is this:
Most oxygen fires are ignited by the "Diesel Effect" (usually slamming
high pressure oxygen against a shut valve), so I always like to make sure
all the valves in the decanting pathway are open before I crack the valve
on the oxygen cylinder. If the cylinder contained contaminated,
pressureized air, then the decanting hose would be pressurized with this
air, and thus the decanting hose would have to be treated as contaminated
itself. I do not have this problem if I always start with an empty cylinder.
Finally, I would make these recommendations to anyone who blends nitrox
themselves:
Get a good, oxygen-clean decanting hose with a monel or stainless steel
needle valve inline, upstream of a oxygen pressure gauge.
Sort of like this:
SS or Monel
needle valve
_ /
_ | _
X-<_|===T-=---------------------=_=-= --Attachment to O2 cylinder
O
^^^^^ ^ ^^
Yoke Pressure High
For Gauge Pressure
Tank Quick-Disconenct
(With
bleed
valve)
To decant oxygen, connect hose to empty scuba tank and full O2 cylinder.
Open scuba tanlk vale ALL THE WAY, open the needle valve in the decanting
hose only a small amount. SLOWLY crack the valve on the O2 cylinder until
oxygen starts to flow. Shut the needle valve and wait until the hose is
fully pressurized. Open the O2 cylinder valve ALL THE WAY. Finally,
slowly crack the needle valve and monitor the tank pressure with the
pressure gauge on the decanting hose.
When the pressure in the cylinder is where you want it, follow this sequence:
1) Shut needle valve
2) Shut Scuba tank valve
3) Shut oxygen valve
4) Open needle valve
5) Bleed the pressure in the hose using the yoke bleed valve
The point is, you never want to "slam" pressurize anything, and you never
want to use the scuba tank valve or the O2 cylinder valve to regulate the
fill rate. A monel needle valve is ideal, but expensive. I use a SS
needle valve which I O2-cleaned, and it seems to work fine. I
periodically disassemble it, and have never seen any evidence of
deterioration or combustion activity in the valve. However, some friends
of mine who used to use scuba or oxygen cylinder valve to regulate the
fill rate found that in a short time the teflon valve seats were burned away.
By the way, the stainless steel high-pressure quick-disconnect is made by
Aeroquip, and is very useful to plug on and off several cylinders easily.
Female part (with F 1/4" NPT) is $38 and male part (also with F 1/4" NPT)
is $11. Let me know if anyone wants a part number.
Ok, that's all....
Aloha,
Rich
deepreef@bi*.bi*.ha*.or*
Navigate by Author:
[Previous]
[Next]
[Author Search Index]
Navigate by Subject:
[Previous]
[Next]
[Subject Search Index]
[Send Reply] [Send Message with New Topic]
[Search Selection] [Mailing List Home] [Home]