O2 Analyzer without decimal

[BRT]

In this whole mess does anybody check their air and adjust the DC according?
Oh wait, that's the calibration point.

I am quite sure in cities o2 in the air could be lower....

I can tell you that the calibration depends on the atmospheric pressure. I doubt very much if the readings are accurate to a full percentage point, i.e. that you can be sure that a reading of 32% is between 31.5 and 32.5%. When I mix nitrox I calibrate all 3 sensors to air. I check the mix into the compressor, the mix out of the compressor and then check the tanks. I get close, but not that close.

 

[tursiops]

I can tell you that the calibration depends on the atmospheric pressure. I doubt very much if the readings are accurate to a full percentage point, i.e. that you can be sure that a reading of 32% is between 31.5 and 32.5%. When I mix nitrox I calibrate all 3 sensors to air. I check the mix into the compressor, the mix out of the compressor and then check the tanks. I get close, but not that close.

Maybe we mean different things by "calibration?"
If I have a tank of air next to a tank of nitrox, and calibrate the analyzer (with a flowrate limiter) on the air so it reads 20.9, and then the analyzer on the nitrox reads 32.2, you are saying (1) atmospheric pressure is somehow affecting this, and (2) the 32.2 might be wrong more than a full percentage point?

 

[BRT]

Maybe we mean different things by "calibration?"
If I have a tank of air next to a tank of nitrox, and calibrate the analyzer (with a flowrate limiter) on the air so it reads 20.9, and then the analyzer on the nitrox reads 32.2, you are saying (1) atmospheric pressure is somehow affecting this, and (2) the 32.2 might be wrong more than a full percentage point?

No, I am saying that if I calibrate the sensor at one elevation it reads differently at a different elevation. And yes, calibrating 3 sensors to air at the same time will not make them read the same on 32%. They simply are not that accurate. Worse than this is attempting a calibration after measuring several tanks of nitrox. Try that and go back to the tanks you just checked and see what you get.

 

[berndo]

Hi
Not really. You can force open it and change the cell.
Once, I broke the housing and used a small transparent plastic food container to hold the parts...not sexy but it worked
As for the precision, people should not confuse accuracy and precision

Did you pry it open with a screwdriver? I'll try that, I still have an old one of these and a bag od old JJ cells.

 

[jborg]

The Palm O2 Sensor manual specifically calls out

That's fascinating as it says not to measure EAN32 after calibrating air, which is probably the most common procedure in the world? With "never" in bold caps, even. I wonder how they intend one should use this device on a random dive boat.

 

[jale]

Did you pry it open with a screwdriver? I'll try that, I still have an old one of these and a bag od old JJ cells.

Yes, I just unglued the top and removed the internals. But was a long time ago (I have a Divesoft now).
I am sure you can make something out of your old unit and JJ cells

PS: if you go to Analyser and scroll down, the picture illustrating "ALL PARTS 100% MADE IN GERMANY HIGHEST QUALITY AND STANDARDS" shows an opened unit

 

[boulderjohn]

No, I am saying that if I calibrate the sensor at one elevation it reads differently at a different elevation.

How often do you calibrate the sensor at one elevation and then use it at another? I'm trying to imagine when that could happen.

 

[iointerrupt]

Look, this isn't complicated enough. Page 128 of the Oxygen Hacker's Companion, has a big table for adjusting for relative humidity. Followed by a discussion whereby even the oxygen sensor manufacturers disagree about it.

So after accounting for sensor accuracy, sensor response time, sensor age, altitude, and relative humidity, you should have a reading which you can then bet your life on. Maybe.

 

[PR_SMITH]

How often do you calibrate the sensor at one elevation and then use it at another? I'm trying to imagine when that could happen.

We have a home in Cyprus, at 650 meters above sea level. If I were to analyze at home, then at sea level, there may be a difference? Haven't actually tried it yet...

 

[jadairiii]

Not really. You're going to round the decimal anyway, so what's the big diff? Nada, in my opinion.

When marking you cylinder, you should not "round up", it should be the reading to that one decimal point. Why you may ask?

People are lazy, and if you see a cylinder that is marked just "32" or "36", there is a good chance the person that marked it was "told" what was in the tank and marked it so. With it marked to the decimal there is a better chance the person marking it (including you) analyzed it, read it, and marked it. I've had analyzed plenty of tanks marked 32 or 36 that were way off.

That said, never use a tank you haven't analyzed yourself. But it you are diving with a group or a team, and they maybe relying on your analyzation, the other diver(s) wants to see the reading to that decimal. That is why you see cylinders marked that way.