A (stupid) gas blending question

[dberry]

The rate of gas mixture equilibration (not equalization of the total pressure) depends on several things. Diameter of the "path" between the two tanks is very important. I know the I.D. of most high pressure hoses (for an SPG) is very small, and that's why cutting a low pressure hose is much worse than cutting a high pressure hose. I'd imagine a transfill whip has a larger I.D., but how big is it?

What surprised me from the thread on this last year was how long it takes for gases to mix in a single tank. If I recall correctly, some people reported that a tank with He added on top of nitrox took days to give a consistent analysis.

 

[broncobowsher]

If you take a bottle of 100% Nitrogen filled to 2000 PSI
And connect that to the same size bottle of 100% Oxygen filled to 2000 PSI.
There is no driving force to move one to the other. For a molecule out of one cylinder to just naturally wonder through the valve, and the whole length of the whip and through the other valve and never randomly turn around and come back at any time. Yes, a few will. Not enough to ever matter. After a week you could test the gasses and you might get a trace amount in a lab test. Most of that would be from air contamination hooking everything up.

It would be like putting 2 glasses of water on a table, expecting a molecule of water to turn to vapor, move to the other glass and condense back into liquid. Yes, it happens naturally. But not enough to ever matter, unless there is an external driving force.

But if you were to take those two bottles and tee them together and transfer to a third (empty) bottle, the third bottle would be 50/50. Each one would drop the same pressure, and being the same size then the same amount would leave.

(yes, there are a few flaws when you get deep into theory. But for this general question it works good enough)

 

[taimen]

I'd imagine a transfill whip has a larger I.D., but how big is it?

Maybe slightly larger. Cylinder valve i.d. may be limiting anyway.

If I recall correctly, some people reported that a tank with He added on top of nitrox took days to give a consistent analysis.

I get very consistent analyzis next day, even when blending late at night. Letting the tanks cool down (laying on side, not upright) for a couple of hours in my car and some driving around seems to give consistent mixing.
I also usually top up helium containing mixes keeping tanks horizontal because this seems to speed up mixing.

 

[rjack321]

Years ago someone did actually this on decostop(?). In a set of doubles they put air in one side and O2 at the same pressure in the other side. Then opened the manifold connection. Drove around with them in their truck for weeks or even a month. Still had 21% and 100% respectively at the end. For all practical purposes they will "never" mix.

 

[RayfromTX]

When you consider that trimix that is PP blended will need to sit for a period of time for a uniform mix to occur and the molecules are all in the same vessel and then consider the journey that individual molecules would have to take down a long and torturous path in which each collision with the inner walls of the whip and the valves causes a change in direction which would as likely as not be a reversal, I doubt that anything like mixing would occur in the whip itself and absolutely never in the tanks. At some point the seals will fail and the mixing will occur with the big air tank in the sky.

 

[davetowz]

No energy driving the gases to "blend" = static values. Objects at rest tend to stay at rest unless a force drives them.

 

[Storker]

No energy driving the gases to "blend" = static values. Objects at rest tend to stay at rest unless a force drives them.

You're a bit off. See @taimen 's post on the previous page.

If you connect two tanks with different mixes there is a driving force: Nature's dislike for differences, which is better known as "entropy". It's just that if you don't have a pressure differential it happens by the random movements of the gas molecules, which is better known as "diffusion". And diffusion through a narrow HP flow line is so slow that you'll have to wait a very long time for the mixes to equalize.