Palm O2 analyzer reading higher than fill station/other analyzers?

[Rearviewmirror]

Same here, but in that case I would set my computer to 31% (lowest of the readings rounded down to the nearest integer). I choose to be on the safe(r) side of things.
In any way practically 1-2% doesn't make any difference since in any way we should avoid going near NDL/MOD especially for repeated dives.
YMMV

I'm a bit confused by that advice? I deliberately set to the higher reading given that would be a shallower MOD (never dropped below an NDL of 5mins regardless).

 

[grf88]

I'm a bit confused by that advice? I deliberately set to the higher reading given that would be a shallower MOD (never dropped below an NDL of 5mins regardless).

Set for the lower percentage of O2 but be aware of the MOD. You want your computer to give you the safer NDL and if you do get delayed and exceed your NDL a suitable deco.

 

[inquis]

I'm a bit confused by that advice? I deliberately set to the higher reading

Use the higher reading for MOD, but use the lower in the computer for tissue tracking (more conservative).

 

[tursiops]

The O2 galvanic cells are sensitive to pressure, temperature, and humidity. You mitigate the pressure sensitivity by using as low a flow rate as possible, and by calibrating against "air" at the same flow rate (i,e, pressure on the cell) as the flow rate you use for analyzing. You mitigate the temperature and humidity sensitivity (see table in post number 9 above (reproduced here for your convenience) by calibrating and analyzing at zero % relative humidity and 32 deg F/0 deg C.

Note the the analyzer can be calibrated to 20.9 only for low humidity and low temps.
Air from an air cylinder is often used for calibration because it is dry....unlike the air in the dive-center room or on the dive boat...where the relative humidity can easily be 70% or more. Suppose the room temp wehe you are analyzing is 90 def F and the RH is 90%; the analyzer ought to be calibrated to 20.0, not 20.9. But when you push the CAL button, it assumes 20.9, so all your readings will be 0.9% high. Additionally, using the little tube analyzers held up against a tank valve invariably entrains some room air into the gas mixture, which lowers the reading done that way. How much? Hard to say, but it WILL happen and is a source of anomalously low readings.

So, in theory, the shop/dive-center analysis and your analysis should be the same. It never will be...
The advice above is excellent: use the lowest analysis for your FO2 setting, and the highest analysis for your MOD.

 

[wnissen]

The O2 galvanic cells are sensitive to pressure, temperature, and humidity. You mitigate the pressure sensitivity by using as low a flow rate as possible, and by calibrating against "air" at the same flow rate (i,e, pressure on the cell) as the flow rate you use for analyzing. You mitigate the temperature and humidity sensitivity (see table in post number 9 above (reproduced here for your convenience) by calibrating and analyzing at zero % relative humidity and 32 deg F/0 deg C.
View attachment 731740
Note the the analyzer can be calibrated to 20.9 only for low humidity and low temps.
Air from an air cylinder is often used for calibration because it is dry....unlike the air in the dive-center room or on the dive boat...where the relative humidity can easily be 70% or more. Suppose the room temp wehe you are analyzing is 90 def F and the RH is 90%; the analyzer ought to be calibrated to 20.0, not 20.9. But when you push the CAL button, it assumes 20.9, so all your readings will be 0.9% high. Additionally, using the little tube analyzers held up against a tank valve invariably entrains some room air into the gas mixture, which lowers the reading done that way. How much? Hard to say, but it WILL happen and is s source of anomalously low readings.

So, in theory, the shop/dive-center analysis and your analysis should be the same. It never will be...
The advice above is excellent: use the lowest analysis for your FO2 setting, and the highest analysis for your MOD.

This is really an excellent overview of what happens and why. For better or worse, the laboratory instrument most people are familiar with is the temperature probe. Since it is measuring molecular activity, there are many robust methods that can be used correlate molecular activity with a signal, and only require a one-time calibration. Thermometers just work. Ditto electrical measurements, they are measuring an electrical property so it's easy to make an accurate one. Multimeters just work.

But chemical instruments, as you note, are not like that. The "oxygen sensor" is a fancy circuit, one that produces a current. That's it. There's a lot of processing and assumptions to turn that current into a percentage of oxygen. I would argue that it's imprecise to even display a tenths place when the instrument isn't accurate to that. It would be more clear to the user if it displayed something like: 31-33%. And that's before we get into all the calibrations, including altitude. Even 1000 ft. / 300 m will put you at 20.1. The tenths place is a lie! You could produce a device that automatically adjusts for temperature, humidity, altitude, etc. (someone probably already has) but it would cost a ton and the extra accuracy would be irrelevant for almost everyone.

 

[tursiops]

Thermometers just work.

One of my pet peeves is when somebody jumps in the water, looks at their dive computer, and announces, "The water temperature is X." There are two things wrong with that: (1) they haven't given their computer time to equilibrate to the ambient water temperature, which might take as much as 5-20 minutes, and (2) they have no idea what accuracy their dive computer has. The spec sheet for many dive computers typically says +/- 2 deg C (4 deg F) (Mares, Suunto), or doesn't say at all (Oceanic, Shearwater, ScubaPro). Given that most folks can easily feel 1 deg C differences, the dive computer readings might be quite misleading. No way to know without actually calibrating one's own dive computer.
FYI, the little Reefnet Sensus is spec'ed at +/- 0.8 deg C, that's +/- 1 deg F. Much better,

 

[wnissen]

One of my pet peeves is when somebody jumps in the water, looks at their dive computer, and announces, "The water temperature is X." There are two things wrong with that: (1) they haven't given their computer time to equilibrate to the ambient water temperature, which might take as much as 5-20 minutes, and (2) they have no idea what accuracy their dive computer has. The spec sheet for many dive computers typically says +/- 2 deg C (4 deg F) (Mares, Suunto), or doesn't say at all (Oceanic, Shearwater, ScubaPro). Given that most folks can easily feel 1 deg C differences, the dive computer readings might be quite misleading. No way to know without actually calibrating one's own dive computer.
FYI, the little Reefnet Sensus is spec'ed at +/- 0.8 deg C, that's +/- 1 deg F. Much better,

Hah, see, even the thermometer doesn't work! You made me curious, my Aqualung computer says the range is "-18 to 60°C (0 to 99°F)" and the resolution is "1°". That actually raises more questions than it answers, but I don't want to get too far off topic.

Short conclusion, the accuracy (and sometimes precision) of an instrument is the worst of the quality of the instrument, calibration, operator, and maintenance.

 

[Tracy]

There are a few reasons, most have been it on above. It practice, you will find that it really doesn't matter much.
As you use your analyzer and get used to it, you will find what it measures in different environments. Calibrate from a known air tank on the LPI and compare the swing from holding it in ambient air. Heat and humidity effect this.
For instance, My personal analyzer in a normal room temp room needs to display about 20.6 in ambient and it will read 20.9 while analyzing air from an LPI. Different analyzers react differently and to make it even more confusing, it will vary as the sensor ages.
The good news from all of this? It doesn't really matter. They are designed for + or - 2% accuracy. That is close enough for any diving you will be doing. You are talking minutes of difference between a few percent. If you are diving that close to the line doing recreational diving, it probably won't be the oxygen percentage error that catches you first.
Always analyze your gas, if it doesn't make sense, analyze it again with another analyzer. Set your computer for what it reads and dive your dive.

 

[PBcatfish]

In my experience, I typically see a few tenths of percent difference from one tester to another in the same environment. If the difference gets bigger than that, a new sensor is probably needed in at least one tester. The sensors do go bad after a few years. They are considered to be a wear item from a maintenance perspective.

I also see a few tenths of a percent difference if I take a reading in my air conditioned house, compared to stepping out into my back yard on an 80 degree, 80% humidity day.

The humidity compensation charts lead me to expect about a half percent error between summer outside air calibration & regulated air tank calibration. Similarly I am led to expect a few tenths difference between air conditioned room air & regulated tank air.

The BIGGEST difference I have seen is between a freshly partial-pressure filled nitrox tank and the same tank after it has been rolled around to mix the contents better. I have seen several percent difference there, not just fractions of a percent. I have measured a tank at the shop, taken it home, gotten a different reading, taken it back to the same shop & then gotten a reading on the shop tester that was very close to my home reading, but not so close to the earlier reading on the same shop tester. Mixing your gas properly matters enough to make a significant difference. I am now very careful to roll all my tanks at the shop before testing them & logging the O2 percentage.

 

[Johnoly]

One of my pet peeves is when somebody jumps in the water, looks at their dive computer, and announces, "The water temperature is X."

I'm guilty of that......and even after 10mins at the bottom I still know my computer's temp is off by 2 degrees.
My dive reports must drive you crazy when I say it's a 1.5 knot current but I have no laboratory measurement instrument. I just use the length of my speargun, counting off 10 seconds, watching how far I've traveled and guestimate my speed.