Roll, Roll, Roll your tank

[vicp]

I bank trimix into 444 cu ft cylinders. They are bolted to the wall. The gas mixes just fine after an hour or 2 after blending. No rolling, no shaking, no witch doctor dance. What mixes gas is time, not agitation. But agitation doesn't hurt anything. Go ahead and shake, rattle, and roll until you're as tired as you want to be. If it makes you feel better about yourself, go ahead and have a ball.

Not quite sure why you are having a cow over this, but; while molecular diffusion and mixing (aided by agitation) are different, mixing does promote diffusion. To quote "Transport Phenomena: A Unified Approach" by Brodkey and Hershey (found on Googgle Books):

"Another way to consider mixing and dispersion is to realize that in the preparation of a completely homogeneous mixture ..., the first step in the process is to disperse one material in the second by some means (such as agitation). If there is no agitation, the homogeneous mixture will be achieved by the process of interdiffusion ..., which will take a long time. If agitation is present, the eddies formed by turbulence from the agitation unit will speed up the diffusional process."

While the author refers to liquids (water and ethanol), the same applies to all fluids, especially presurized gasses as they become more liquid-like and slower to diffuse.

No one disputes that time is required for complete diffusion (as in your 444 cu ft trimix banks), it is just that agitation can help speed the process, especially if you want a more accurate reading shortly after getting a (blended) fill.

 

[Wookie]

I was at my LDS last week to get a Nitrox fill. The shop owner completed the fill and then placed my tank on an electric rolling machine that slowely spun the tank for 5 minutes to make sure the gases were evenly distributed. Then he checked the concentration with the meter and put the sticker on. Then he had me check the concentration with the meter. After recording the fill, I paid the bill and left. I have never seen tanks spun on a machine after a mixed gas fill. Is there an oppinion either way? Is it overkill?

Not quite sure why you are having a cow over this, but; while molecular diffusion and mixing (aided by agitation) are different, mixing does promote diffusion. To quote "Transport Phenomena: A Unified Approach" by Brodkey and Hershey (found on Googgle Books):

"Another way to consider mixing and dispersion is to realize that in the preparation of a completely homogeneous mixture ..., the first step in the process is to disperse one material in the second by some means (such as agitation). If there is no agitation, the homogeneous mixture will be achieved by the process of interdiffusion ..., which will take a long time. If agitation is present, the eddies formed by turbulence from the agitation unit will speed up the diffusional process."

While the author refers to liquids (water and ethanol), the same applies to all fluids, especially presurized gasses as they become more liquid-like and slower to diffuse.

No one disputes that time is required for complete diffusion (as in your 444 cu ft trimix banks), it is just that agitation can help speed the process, especially if you want a more accurate reading shortly after getting a fill

Not having a cow. To refresh you with the OPs statement,

Is there an oppinion (sic) either way? Is it overkill?

My opinion which is derived from mixing thousands and thousands of cubic feet of trimix and (as near as I can tell) 10 million cubic feet of Nitrox is that agitation of the cylinder is not only unnecessary, it is useless. Rolling a cylinder will not make the mixture become homogenous more quickly. Placing some marbles in the cylinder and rolling it around would, however. Rolling a cylinder does not agitate the gas.

 

[vicp]

Not having a cow. To refresh you with the OPs statement, My opinion which is derived from mixing thousands and thousands of cubic feet of trimix and (as near as I can tell) 10 million cubic feet of Nitrox is that agitation of the cylinder is not only unnecessary, it is useless. Rolling a cylinder will not make the mixture become homogenous more quickly. Placing some marbles in the cylinder and rolling it around would, however. Rolling a cylinder does not agitate the gas.

That is fine. You know what they say, opinions are like belly buttons (or any other anatomical feature you want to use), everybody has one. I am sure in a commercial setting where you bank considerable amounts of gas, it is already blended (as you say after 1 or 2 hours) and fills are consistent.

But I will say, from FIRST-HAND OBSERVATION (MINE AS WELL AS OTHERS IN THE CLASS), partially blended trimix (that is using Partial Pressure (PP) Blending method, not continuous blending (stick) or bank fills). that the immediate readings were false and those tanks that were agitated (maybe not just rolled, but layed flat, inverted, etc.) displayed proper readings in a MUCH SHORTER PERIOD OF TIME THAN THOSE LEFT ALONE.

This is a well discussed topic on the forum. If you search for "diffusion" you'll find a lot of threads circa 2007 discussing this.

It's simply that at 200 bar gases start to get viscous and diffusion velocities are very (surprisingly) slow so there is a long spontaneous mixing time. Various posters ran calculations on this.
By rolling them you really do speed up the mixing (diffusion) process.

Some people talking about smooth walls not mixing anything. Once the gas gets viscous at 200Bar there will be a lot of drag on the walls.

One can argue if rolling is the most effective method of agitation, but I agree completely with miketsp above. Fluid dynamics is something I studied fairly thoroughly in my Ocean Engineering undergraduate work. Depending on method and flow rate of the fill, the different molecular weights of the gasses (Nitrogen, Oxygen, and especially Helium) may cause them to stratify (layer) until diffusion runs its course. Anything that upsets this stratification (high flow rate - not good for O2, rolling, inverting, shaking, etc.) increases the interaction surface area between the gasses (mixing) and that promotes diffusion.

But then again, it is your opinion.

 

[Akimbo]

This is probably a remnant from the early days of commercial and military mixed gas (HeO2) diving. Industrial gas suppliers would partial pressure mix by weight (which is very precise), but readings were not always as-ordered on the O2 meter calibrated against a certified cal gas. We would roll these K-bottles (230+ Ft³) up and down the pier and they would read spot on. Of course the different densities of Helium and Oxygen is much more likely to produce stratification in a large cylinder than Trimix with a small percentage of Helium in a small cylinder.

We found that standing the cylinders in an East-West line with a shade (tarp) over the top third worked well in warmer climates and was much less work. I have never needed to do it on Scuba-size cylinders.

The proof in is the reading. The only reason I can see rolling before taking the first reading is their fill station has produced unreliable mixing so often it isn’t worth the trouble. Rolling without ever testing first is to see if it is necessary would be a waste.

 

[jacksatlanta]

The question wasn't about stratification of Helium and Oxygen found in Trimix but was about Nitrox fills. While I agree stratification can occur in Trimix tanks and needs to be considered I've never seen a Nitrox tank that was not adequately mixed through filling alone, particularly at the fill rates most shops use, even in a partial pressure method and I've been at it awhile.

 

[Akimbo]

The question wasn't about stratification of Helium and Oxygen found in Trimix but was about Nitrox fills...

Agreed, I was only speculating on what motivated them to bother with the additional step of rolling.

---------- Post Merged at 12:22 PM ---------- Previous Post was at 12:17 PM ----------

Love to see them tumble a set of doubles......

Isn’t that what Laundromat dryers are for?

 

[vicp]

The question wasn't about stratification of Helium and Oxygen found in Trimix but was about Nitrox fills. While I agree stratification can occur in Trimix tanks and needs to be considered I've never seen a Nitrox tank that was not adequately mixed through filling alone, particularly at the fill rates most shops use, even in a partial pressure method and I've been at it awhile.

Agreed.

I never noticed this to any degree with Nitrox (but definitely with Trimix) other than when filling manifolded doubles and having to wait a couple of minutes for the O2% to creep up before settling.

 

[Rogersea]

I have never seen tanks rolled to ensure the gas is mixed well, like Wookie alluded to with Partial Pressure(older method of mixing) the gases are not perfectly mixed/distributed... They used to let the tanks sit for a while before analyzing.... This is not really necessary with continuos blending methods... But again if it makes the gas blender and shop owner feel good there is no harm....

Cheers,
Roger

 

[dumpsterDiver]

I have never seen tanks rolled to ensure the gas is mixed well, like Wookie alluded to with Partial Pressure(older method of mixing) the gases are not perfectly mixed/distributed... They used to let the tanks sit for a while before analyzing.... This is not really necessary with continuos blending methods... But again if it makes the gas blender and shop owner feel good there is no harm....

Cheers,
Roger

Yeah but he said he waits a few hours, didn't he? I fill my tanks and then want to check them and then immediately load them in the truck. I've seen the difference rolling makes.

 

[Glenn08]

How much gas (volume) is consumed during the analyzer testing ? If only a small volume is drawn, it will be the gas in the extension tube of the tank valve, not the mixed gas in the tank. So, instead of tumbling, blow off enough to clear that tube, and tell us if the analysis is accurate then.