Mailing List Archive Mailing List Archive
Scott wrote:
> rick,
>
> i still stick with "hogwash".
>
> physical chemistry class will prove all of this. ... snip
> the gas doesn't stratify in your tanks.
I'm not suggesting that it is. I don't see how it could be unless
the gas was liquified or in a hypergolic state, which it wouldn't be
when near normal temperatures and tank pressures.
But what are the diffusion times of gasses at tank pressures? Even at
atmospheric pressures it takes some gasses quite a while to diffuse.
After all, when someone breaks wind in a room, everyone doesn't smell
it instantaneously. :-)
Another example is propane settling in low areas, like the bilge of
a boat, and causing explosions.
Of course, propane is a much larger and heavier molecule than O2 or N2,
so it will diffuse much more slowly into air.
The propane is also usually cooler because it has just been released from
a container at higher than atmospheric pressure. So the higher density of
the cooler gas also tends to make it settle into low areas.
Does the high temperature and resulting low density of a gas just pumped
into a cylinder cause it to rise to the surface of an otherwise ambient
temperature tank like a hot air balloon rises in cooler ambient air? Seems
reasonable.
The hot tank valve at the top would also tend to keep the gas near it warm.
The high temperature would tend to make it diffuse more quickly, though.
If the tank is turned on it's side, the "top" would be the long side wall
of the tank, spreading the hot gas out so that it has a much greater
interface area with the cooler gas, thus promoting diffusion.
Rolling the tank would induce laminar flow and tend to spread the warmer
gas into an even larger interface area with the cooler gas, further
promoting diffusion.
Laminar flow effects would be enhanced greatly by the high density of the
gases at tank pressures.
Gasses at thermal equilibrium don't stratify in the tank. Gases at higher
temperatures rise to the top and if the components of a mix are heated
differentially they will rise differentially.
It would seem that this effect would only be observed when the gasses are
pumped into the tank separately, as in pumping O2 into a tank then pumping
air in for a nitrox fill.
Also, pumping by compressor versus decanting from other cylinders or a
cascade system would have different effects.
This would seem to suggest that the last gas pumped into a cylinder would
be over-represented in a sample conducted soon after the fill.
Is this what people are seeing?
>
> people are seeing different values for their mix after they wait
> a day, but nobody has done a real cause/effect analysis on why
> this is happening.
Ah, so the effect is real, it's the cause that's in doubt? Have you
ever experienced this effect with the mixing techniques that you use?
If not, it would suggest that the others may be getting it through
mixing or measuring techniques that introduce errors.
> what if you laid it on its side for 6 hours
> and farted on the cylinder three times in rapid succession. then
> you analyze, and the number is different. now everyone assumes
> it was the farting and starts doing it.
No, only someone who falls into the trap of using faulty, reverse logic
in cause-effect relationships.
> please read chapter 9 of robert pirsig's "zen and the art of motorcycle
> maintenance" to get a real understanding of what i am talking about here.
> better yet, read the whole book. (sorry about the farting for whoever
> was offended - i couldn't resist)
Persig's book is humorous and an easy read. It's particularly helpful to
those who have difficulty in grasping concepts in drier texts on logic and
ontology.
"If she's a witch she must be lighter than a duck."
Monty Python and the Holy Grail
Maybe you should go back and read it again, since you seemed to jump to the
erroneous conclusion that I thought the gas was "stratifying" in the tank.
(My apologies as well on the breaking wind analogy). :-)
>
> any residual gas in the valve stem should not be a problem. i
> alway purge that first to blow any funk out of my valve before i
> hook up my flow regulator. you are using a flow regulator with
> your analyzer, aren't you? if not, there is a source of error.
> and again, the analyzer is ONLY good to +/- 2% if you have a
> miniox like me. read the owners manual.
>
No need to get testy. The questions I posed were sincere. They weren't
meant to be rhetorical put downs of what you were saying. I was merely
stating that there seems to be a disconnect between what is known about
gas behavior and what is being observed in the field. This indicates
that there is something that we are missing that we might need to find
out about.
>
> the biggest problem is temperature. the gas has to have reached
> ambient temperature or it will be cooling as you are analyzing.
> also, the gas will be cooler as it exits the tank due to
> expansion (like your a/c freon). the analyzer cannot correct for
> temperature. your calibration gas, unknown gas, and analyzer
> really need to all be at ambient temperature and at equilibrium.
>
Sounds reasonable to me.
OK, to get back to the original question: Are oxygen spikes a danger
in using Nitrox?
What do we know so far?
1. Gas at thermal equilibrium doesn't stratify in tanks. The O2 won't
settle to the bottom of the tank because it is heavier than N2.
2. Some people have observed unexpected test readings after a gas
mix, the same cylinder tests out at the expected values after
the passage of some time.
3. Some people say rolling the tanks helps mix the gas in the tank
after a blend, giving a test result closer to the expected numbers
and closer to test numbers after the tanks sit for some time period.
4. The calibration gas, unknown gas, and analyzer really need to all
be at ambient temperature and at equilibrium to get meaningful test
results.
Does it matter as far as getting an O2 spike at some point in the dive?
It appears that it doesn't matter unless you snatch the tank off the
compressor and immediately jump in the water. Even then, the immersion
in water will tend to move the tank contents to thermal equilibrium and
a uniform blend very rapidly.
The main problem seems to be that it can complicate testing, causing you
to wait for a period of time after the fill to get good results.
Thanks for the reply Scott. Does the above sound reasonable to you?
Rick
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]