Should I buy a nitrox (O2) analyzer?

[Centrals]

To trust you club isn't the point that he is making, but that you need to cross check to make sure that your analyzer is correct.

How difficult it is to check the accuracy of an O2 sensor?
Nitrox 21 is readily available as standard reference at no extra cost.
1% deviation is perfectly acceptable for recreational diving.

Don’t Trust Your Gas Blender - Analyze Every Tank - SDI | TDI | ERDI | PFI

If the operator does not have an analyzer of their own for anyone to use then the chance is that the nitrox tank is probably from somewhere else.

I can only speak for my own experience : I have yet to come across any operator in SE Asia could not provide me with an working analyzer to analyse the tank.

If anyone believe he/she would need an O2 analyzer then go ahead.
Analyse the content of the tank by yourself ONLY.

 

[KentB]

Your trust to your club is your own business but I would not use any nitrox tank that I have not analysed myself.
I have been using Nitrox since 1997 and never ever dived a nitrox tank without being analysed by myself first.
I have no control on others, just keep it simple.

should check every tank not just nitrox.
The only time I wasn’t checking tanks was when I was brand new to diving and before nitrox and didn’t know better.

and again when I first bought my compressor filling my own tanks without any oxygen on site. As soon as I got a nitrox stick and O2 everything got analyzed even if I was confident it was just air.

FWIW I have the divesoft Trimix analyzer and it’s absolutely fantastic so far

 

[Centrals]

Analyzing your own nitrox tanks

 

[Storker]

Your trust to your club is your own business

I don't trust my club. But they provide three other data points (two cells in the inline mixer, one external analyzer). I don't trust my own analyzer either, but if all analyses agree plusmimus one percent, I trust them. If they don't and I can't find a plausible explanation, I sit the dive out.

I'd never, ever trust one single analysis or one single analyzer. Cells die, and we can't know when they die unless we have a rather elaborate analysis and calibration scheme (which most of us don't)

 

[Storker]

How difficult it is to check the accuracy of an O2 sensor?
Nitrox 21 is readily available as standard reference at no extra cost.
1% deviation is perfectly acceptable for recreational diving.

Yeah, and a failing cell may well be able to give the correct signal for 21%, but fail majorly on 32%, 36%, 40% or 100%. If you calibrate against anything leaner than your richest mix, YOU. CAN'T. KNOW. Unless you calibrate against at least two points, one of them noticeably richer than your mix, YOU. CAN'T. KNOW. Using multiple cells/data points mitigates that risk. As does theoretical mix calcs vs analysis numbers if you're PP blending.

I'm a chemistry PhD, and analytical chem is one of my favorite pastimes on paid time. You want to learn about calibration and trust in your analysis? Insert a penny and budget some time, because I'm gonna talk for quite a while.

 

[BRT]

How difficult it is to check the accuracy of an O2 sensor?
Nitrox 21 is readily available as standard reference at no extra cost.
1% deviation is perfectly acceptable for recreational diving.

Don’t Trust Your Gas Blender - Analyze Every Tank - SDI | TDI | ERDI | PFI

If the operator does not have an analyzer of their own for anyone to use then the chance is that the nitrox tank is probably from somewhere else.

I can only speak for my own experience : I have yet to come across any operator in SE Asia could not provide me with an working analyzer to analyse the tank.

If anyone believe he/she would need an O2 analyzer then go ahead.
Analyse the content of the tank by yourself ONLY.

Just because the analyzer will calibrate to air does not mean it will read a higher PO2.

 

[Storker]

Just because the analyzer will calibrate to air does not mean it will read a higher PO2.

O2 cells fail by not being able to deliver a proportional voltage.

A failing cell may fail to deliver the voltage corresponding to 100%, but deliver the correct voltage for 80%. So, if you're calibrating at 80%, you're screwed.

A failing cell may fail to deliver the voltage corresponding to 80%, but deliver the correct voltage for 40%. So, if you're calibrating at 40%, you're screwed.

A failing cell may fail to deliver the voltage corresponding to 36%, but deliver the correct voltage for 32%. So, if you're calibrating at 32%, you're screwed.

A failing cell may fail to deliver the voltage corresponding to 32%, but deliver the correct voltage for 21%. So, if you're calibrating at 21%, you're screwed.

Bottom line, unless you have multiple cells confirming your analysis, or unless you have calibrated your analyzer against a much richer mix than you're using AND a leaner mix, you can't trust your analysis.

 

[Storker]

should check every tank not just nitrox.

Even with an inline mixer, I believe that this is overkill. Unless someone opens an O2 tank, what you get is air. More or less filtered, but air.

If my tanks have been filled with air and no-one has opened an O2 tank during the filling, I'll trust that what is in my tank is air (or EAN21). Except for my OCD, which requires me to analyze all my tanks.

 

[boulderjohn]

should check every tank not just nitrox.

You have to use a little common sense.

There are two dive shops in Boulder, Colorado, and neither one has the ability to put anything other than air in a tank. One of those shops provides nitrox for its instructors to use on busy certification days when they are doing multiple dives, but when they do, I am the one doing it, and I only put it into specific tanks I O2 cleaned myself.

At the Blue Hole in Santa Rosa, NM, the person doing the fills only has the ability to pump air. She does rent nitrox sometimes, but that will be in special tanks delivered to her from a shop far away. If you use any of the tanks she provides, or is she refills one of yours, it will be air.

There are many situations like that. On the other hand, yes, if you get a tank filled by a shop capable of making nitrox, then you should check.

 

[BRT]

O2 cells fail by not being able to deliver a proportional voltage.

A failing cell may fail to deliver the voltage corresponding to 100%, but deliver the correct voltage for 80%. So, if you're calibrating at 80%, you're screwed.

A failing cell may fail to deliver the voltage corresponding to 80%, but deliver the correct voltage for 40%. So, if you're calibrating at 40%, you're screwed.

A failing cell may fail to deliver the voltage corresponding to 36%, but deliver the correct voltage for 32%. So, if you're calibrating at 32%, you're screwed.

A failing cell may fail to deliver the voltage corresponding to 32%, but deliver the correct voltage for 21%. So, if you're calibrating at 21%, you're screwed.

Bottom line, unless you have multiple cells confirming your analysis, or unless you have calibrated your analyzer against a much richer mix than you're using AND a leaner mix, you can't trust your analysis.

Oxygen sensors are actually a current source that varies with the oxygen level.