Nitrox: How to calibrate/analyze

[matt_unique]

...
But if I am checking a tank of 50% that I will be using for deco at a PO2 of 1.6, the size of the error is potentially greater and I am also much more concerned with knowing if it is fact 45% or 55% rather than 50% due to the effects this will have on MOD and PO2.

Unless I missed something in your post, the calibration with ambient air vs. pressurized air will not create the large discrepancies you reference above. NWGrateful diver posted earlier about a .1 difference between the two methods. Even if you are checking your 80% deco gas with a planned 1.6pp02 stop the difference is still dramatically less than the accuracy of the device. Were you saying you would not use ambient air calibration for deco gas measurements?

--Matt

 

[MXGratefulDiver]

Last night I repeated the experiment in the Nitrox class I'm teaching. I had the students take several measurements, first from ambient air, then off a cylinder filled with air at 3,000 psi. I had them do it both using the flow restrictor (via the inflator hose), and via the more commonly used method of holding the analyzer in front of a flowing cylinder valve.

We took a total of six measurements. The greatest variance we got, using both methods, was 0.3 percent between ambient air and cylinder (pressurized) air.

Since this is still only one data point with respect to the fact that all of these measurements were taken with the same analyzer, I would be interested if those of you out there with analyzers could repeat this experiment and post your results here ... both the type of analyzer you're using and the difference between ambient air measurement and pressurized air measurement.

... Bob (Grateful Diver)

 

[AtomicWalrus]

When you're calibrating your nitrox analyzer, the most important factor is to get enough flow across the sensor to avoid oxygen depletion during calibration. The sensor is an electrochemical cell that reacts with oxygen to produce voltage, and as it operates it will consume the oxygen in the air around the sensor. If the volume in the sensor is relatively small and the air diffusion rate in is low, you'll get a noticeable change in concentration during calibration with the result that your readings will be too high when you analyze your tank. That's the reason for the flow rate guidelines provided by analyzer manufacturers.

The pressure of the sample isn't really an issue. The effect of pressure on the calibration can be explained with the Nernst equation, which shows that the effect of pressure is logarithmic - you'd really need to almost double the pressure to have a big effect on the output of the sensor. This will vary with the sensor, of course, but again this is taken into account by the flow rate guidelines. Keep the same flow rate when calibrating as analyzing, and pressure effects really won't be an issue.

All of these effects are observed in hydrogen fuel cells as well - I work in fuel cell R&D, which is where I get my background in electrochemistry. These nitrox analyzers are reasonably accurate when they're maintained & used properly, but they're really dependent on the quality of the accompanying electronics. That's why I bought my own - I just don't want to trust the shop's analyzer...

Of course, if you wanted to get a REALLY good O2 analysis, you could use a paramagnetic analyzer like the Rosemount NGA2000 - of course, they cost $10,000 and aren't exactly portable!!

 

[padiscubapro]

The pressure of the sample isn't really an issue. The effect of pressure on the calibration can be explained with the Nernst equation, which shows that the effect of pressure is logarithmic - you'd really need to almost double the pressure to have a big effect on the output of the sensor.

not true ask ANY CCR diver..

we calibrate our meters for PO2 readings but its just as aplicable for setting them up for FO2..

If I flush with oxygen at 6m (1.6 atas) I'll get a 1.6, go up to 3 m, I only have a 1.3, then up to the surface I have 1.0, that is a tremendous change over much less than doubling the pressure..

with a 30 % increase in pressure a the same ( scaled) increase is seen.. Oxygen sensors are linear in output (well almost) the output rises linearly whether its a pressure OR concentration rise. Without this relationship RB electronics would be muchmore complivcated than they are.

if I take a sample of air to 5atas (a po2 of 1.05) I would expect to see the same voltage across the cell as bring 100% oxygen to just over 1 ata..

if I get 7mv in air off the sensor I expect approx 33mv in 100% oxygen
If I was seeing 7mv (in air at 1 ata) I would expect to see 5 times that at 5 ATAs..

This is easily verified and is the basis for meaurement using thesesnesors.

 

[Keysdrifter454]

I recently passed a Nitrox course; at the shop I was taught to analyze by attaching the analyzer to a tank of regular air, adjusting the dial on the analyzer until it reads "20.9" or thereabouts, and then remove the analyzer and place it on the tank containing the EANx mix.

However my friend was telling me that you can just leave the analyer in the open air and calibrate it that way--would anyone be able to give me a step by step procedure for doing it that way? thanks!

The way you were taught was the way I was taught, which is the only way I do it.

The reading is particular to flow rate.

I suppose waving the analyzer in the air probably works, just as long as no lines or adapters are attached, but I don't do it.

Waving anything but the bare sensor in the air is a roll of the dice.

 

[AtomicWalrus]

I was referring to pressure effects for calibration of a nitrox analyzer at ambient pressure. You're talking about the effect that a significant pressure change has on the sensor. Same difference. The effect in question is described by the Nernst equation:

E = Eo + (RT/nF)*ln(([reactant activity]/[product activity])*P^0.5)

where
E = sensor voltage
Eo = ideal sensor potential
R = universal gas constant
T = sensor temperature
n = mols of electrons involved in reaction
F = Faraday's constant
P = sensor pressure

That pressure term is what I was talking about, and it's why when you're calibrating a sensor to measure oxygen concentration that you need to make sure your sensor is at the same pressure during calibration as during sample analysis.

This equation also shows where the non-linearity in sensor output comes from: over a small range, the logarithmic output is relatively linear. Over larger ranges, the non-linearity becomes apparent and sensitivity decreases substantially.

not true ask ANY CCR diver..

we calibrate our meters for PO2 readings but its just as aplicable for setting them up for FO2..

If I flush with oxygen at 6m (1.6 atas) I'll get a 1.6, go up to 3 m, I only have a 1.3, then up to the surface I have 1.0, that is a tremendous change over much less than doubling the pressure..

with a 30 % increase in pressure a the same ( scaled) increase is seen.. Oxygen sensors are linear in output (well almost) the output rises linearly whether its a pressure OR concentration rise. Without this relationship RB electronics would be muchmore complivcated than they are.

if I take a sample of air to 5atas (a po2 of 1.05) I would expect to see the same voltage across the cell as bring 100% oxygen to just over 1 ata..

if I get 7mv in air off the sensor I expect approx 33mv in 100% oxygen
If I was seeing 7mv (in air at 1 ata) I would expect to see 5 times that at 5 ATAs..

This is easily verified and is the basis for meaurement using thesesnesors.

 

[MXGratefulDiver]

Dang!

Hope you guys aren't planning to give us all a test ...

... Bob (Grateful Diver)

 

[padiscubapro]

Dang!

Hope you guys aren't planning to give us all a test ...

... Bob (Grateful Diver)

The clearest paper on the subject of oxygen sensors that I have seen can be found at

http://www.teledyne-ai.com/pdf/lauer.pdf

 

[MXGratefulDiver]

Thanks ... actually, I understand how it works. What the typical diver wants to know, however, is simply how to accurately analyze the contents of their cylinder.

Empirically, the stuff you guys are discussing may be of interest to the technical diver ... but to the majority of the folks in this (Basic Discussions) forum, it amounts to measuring with a micrometer, marking with chalk, and cutting with an axe.

... Bob (Grateful Diver)

 

[Upwelling]

I'm a little confused now--when you rent a nitrox tank do you get to do the analysis yourself with the shop's analyzer or does the shop worker do the analysis for you? thanks