Question How to check an analyser O2 cell?

[Manatee Diver]

You pointed this out yourself... The functional linearity is lost from the top of the curve. So in an analyzer when you're only reaching 98% instead of 100% what does that mean for your accuracy at 32%? Nothing, that is less than 0.5% deviation. You get more variability from barometer pressure changes than this.

Yes order a new cell, but while you wait for it to arrive you can continue to use that analyzer

There is also what are you using the analyzer for. I am simply verifying tanks that I know either contain 32% or oxygen. If they were filled with anything else the results would be very different. A bit different requirements from a CCR cell.

 

[Agro]

You pointed this out yourself... The functional linearity is lost from the top of the curve.

That's what I say

 

[Agro]

So in an analyzer when you're only reaching 98% instead of 100% what does that mean for your accuracy at 32%?

It does not say anything about accuracy at 32%. It sais the cell will soon be faulty.

 

[Agro]

You get more variability from barometer pressure changes than this.

A good analyser compansates barometer pressure. Divesoft analysers do so.

 

[Agro]

There is also what are you using the analyzer for. I am simply verifying tanks that I know either contain 32% or oxygen. If they were filled with anything else the results would be very different. A bit different requirements from a CCR cell.

Can you be 100% sure it is 32%, not 30 or 34 or 25 or 47?

 

[rjack321]

A good analyser compansates barometer pressure. Divesoft analysers do so.

Shrugs there is no need to do this and the vast majority of analyzers do not.

Can you be 100% sure it is 32%, not 30 or 34 or 25 or 47?

LOL are you done grabbing at straws? First of all analyzers are to confirm what you already know based on mixing calculations.
Your analyzer deviating 2% at 100% is not going to somehow magically mis-analyze 32% as 47%

If it's 7% off at 32% (inaccurately reading 25% instead of 32) it will have way more than a minor 2% deviation at 100%. The slope of the line will have it down in the 80% range at best and its clearly not useful. They are simple batteries and generally fail in predictable ways. Used up KOH leading to loss of voltage output (which is actually current) at the higher O2 exposures, perforated membrane leading to vastly excess output, and occasionally a failed temperature compensation circuit

 

[Agro]

Shrugs there is no need to do this and the vast majority of analyzers do not.

Use a ATA Analox (without barometer compensation) at only 2'000 meters above sea and you will see what lousy results you get. Yes you can compensate that by yourself:
- use a barometer when calibrating, for ex. 1'010 mbar at 0 meters above sea
- use a barometer at 2'000 m, for ex. 813 mbar
- take the wrong measurement from the analyser and calculate

Even at same altitude you have different barometric pressures, it is simply great if your analyser is compensating.

Yes the majority of analysers do not. That's why I wrote a GOOD analyser has compensation. An analyser without barometric compensation is not good.

 

[Agro]

First of all analyzers are to confirm what you already know based on mixing calculations.

This attitude leads to death.

 

[Agro]

Your analyzer deviating 2% at 100% is not going to somehow magically mis-analyze 32% as 47%

Deviation 2% today means the sensor is not good anymore. It will become worse within the next time. We do not know if it becomes worse tomorrow or in 1 month, we simply know the sensor is dying. And yes after this it can show any other number but 32%. Perhaps tomorrow, perhaps in 1 month.

 

[rjack321]

This attitude leads to death.

This is pretty rich coming from someone diving a liberty without any training. Apparently, you have no mixing cert either.

If you add 14% O2 and top with air you get 32% every time. The analyzer is just confirming that your original math was correct.