Oxygen Analysis (Nitrox)

[SM Diver]

The three tanks I have sitting here for tomorrow's trip were all analyzed immediately after their fill, without so much as a roll on the floor. I am thinking about going back to the shop and asking for a re-analysis before tomorrow's dives.

Would this be wise, or a waste of time? Also, does the oxygen percentage tend to be higher or lower after having a chance to settle and mix? Input appreciated.

--------------------------------

Scott in Tampa

 

[Rick Murchison]

Does the shop do partial pressure blending in the tank? If so, the tanks need rolling and waiting before analysis. If they're premix there's no need to roll or wait.
Most shops are doing premix these days.
The FO2 doesn't change over time, but you can get a bad reading high or low (usually low but no guarantees) on a sample taken right after a partial pressure fill (in tank mixing).
Rick

 

[FredT]

Were they premix or PP fills?
How close to MOD were you planning to go?
Are you going to be pushing the maximum bottom time for the mix?

If premix getting another read is a waste of time. PP probably is too if the meter was run correctly the first time.

If doing square profiles to MOD exact mix numbers can be necessary, but the practice of pushing MOD and exposure times is normally a Bad Idea unless using a surface supplied hat or full face mask.

Simply use appropriate fudge factors to account for possible misreadings at the O2 gauge.

As an example for rec diving I prefer to figure O2 exposure at 1 to 2% higher than the guage reading, and crunch deco numbers for at least 2% lower O2 than gauge. When doing bounce hunting on the rigs I'll normally use a 28% mix. I assume it's 21% for figuring deco, and 1.6ATA allowable on a 30% mix for short term O2 exposure. Be aware that a long rig bounce dive may last all of 10 minutes deck to deck.

FT

 

[The Chairman]

If yu calibrated on room tamperature air (most likely) and then measured hot air from a newly filled tank, the measurement variation could be significant. Redo the analysis.

 

[JDostal]

Even if it is a PP fill, you don't need to wait on the analysis. You don't need to wait for the gases to "mix" inside of the bottle.

Envision a bucket of water, and you have a very small, very high pressure water hose - now spray that hose into the bucket of water. This is what happens when you do PP fills. You don't need to wait for the gases to mix, you don't need to roll the tanks or shake them around...unless you are looking for a workout.

As far as analysing them goes...we do it before we hook up the tanks. An analyser isn't too expensive, and it just gives us peace of mind.

 

[JimC]

For all intents and purposes a scuba tank will be mixed almost instantly to a near homogenious mixture. A tank will never need to be rolled. It may need time to cool down so you get an good reading, but thats the fault of the sensor, not the mix.

Read up on The kinetic theory of Gases, Maxwell-Boltzman distribution and GasLab.


The water bucket thing if a neat analogy, but liquids and gases mix very very diferently. Gas will never settle, and mix them selves. Having each gas "shot" into the tank does mix them faster but we are already talking tiny fractions of a second.

 

[SM Diver]

It was actually kind of an "unusual fill", because the shop had just sold their compressor and had a temporary station set-up. They normally do a partial pressure fill, but on this occassion had a large tank of pre-mix at 36% that they had saved, and then just topped it off with air to get me one 34% and two 32%. It was kind of a "crap shoot." However, given that it was a pre-mix topped off with air, would it still be advisable to get it reanalyzed?

Also, the comment about buying the analyzer is well taken. I may just go get one myself. Any recommendation on good brands?

 

[SM Diver]

FredT once bubbled...

How close to MOD were you planning to go?
Are you going to be pushing the maximum bottom time for the mix?FT

The deepest dive tomorrow will be 90ft. I was planning on using the tank that analyzed at EANx 34.1. That should leave plenty of room for error, leaving me well within the max PPO.

 

[MikeFerrara]

Temperature changes will effect the reading some. If you are not red lining the parameters on your dive it shouldn't be a big deal. Analyzers are only garanteed to within a percent or two anyway. I carry my analyzer and analyze before a dive even though I analyzed after blending. However I do so out of habit and as a matter of procedure (insurance). I'm not as concerned about the reading changing a little as much as I am incorrect labeling. When mixing and labeling my own gas, analyzing is just a formality as far as I am concerned. I would be comfortable diving my mix without analyzing because I trust the procedures. The worst I could do is be a couple percent off and that won't hurt me given the way I plan a dive. However, if the labels are swapped on a bottle of 50% and a bottle of 100% I will almost certainly die.

Back when we used 80% and topped air very slowly I did get some readings that were not even close way way low like 35%. As a new blender I even dumped a entire bottle of 80% once because I couldn't make sense of the readings. After remixing I got the same results. After rolling the readings were dead nuts on. It is possible to mix in such a way that does not give a homogenous mix immediatly. In that case very little air was added and it was added very very slowly. I know what some say about how the gas should mix but I think I could repeat those results if I wanted. With normal mixes enough air is going in on top that there is plenty of mixing going on.

When PP blending the most important analysis takes place in the brain of the blender. I once witnessed a freak situation that envolved multiple failures. The blender used software to calculate the mix. Incorrect numbers were entered and the mix given was way off. The blender, relying on and trusting the software blended the mix. It was like when you hit the wrong button on a calculator and get an answer that isn't even close. If you haven't estemated the answer in your head you won't notice. No big deal right? The analyzer should catch it. But...a reg that was full of water was used to cal the analyzer so it too gat full of water. Just by chance the reading was just what it should have been even though the FO2 was way high. The bad mix was discoverd because in rechecking the numbers it was clear that either the mix wasn't done according to the software output or the analyzer was fried. The point is that in the off chance that there is more than one failure in the process a single analysis can not relyably catch it because the analysis may be one of the failures. This is why correct procedure is so important when PP blending. We analyze to catch mistakes but the idea is to not make any.

 

[sheck33]

JimC once bubbled...

The water bucket thing if a neat analogy, but liquids and gases mix very very diferently. Gas will never settle, and mix them selves. Having each gas "shot" into the tank does mix them faster but we are already talking tiny fractions of a second.

That a gas mix will never settle out like a mudd/water mix is true but gasses WILL mix themselves due to the Brownian motion of the gas molecules. It will just take a while depending on the gasses involved.