O2 Analyzer without decimal

[Tedestam]

Hiya
I just bought an O2 analyzer of the brand NRC and realized it only shows full % numbers. We want to use it for teaching the PADI Nitrox course, however the theory specifically states that an analyzer with one decimal should be used. Are there any major concerns for using this and do we need to be more conservative when using it? I do not know if the analyzer already rounds up to the next full number. Any thoughts on this?
Thanks

 

[The Chairman]

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

 

[dberry]

It's basically telling you that 32% could be anywhere from 31.5 - 32.5%. Have your students calculate MODs for those values.

Isn't there something about measuring with a micrometer and cutting with an axe?

What I'm less sure about is how the uncertainty in calibration range (20.5 - 21.5%) might change the measured pO2... it's probably not linear (i.e. it might not add just 0.5% to the measured value.) Anyone ever do the experiment? I would, but I only have cylinders with air at the moment.

 

[BoltSnap]

It's basically telling you that 32% could be anywhere from 31.5 - 32.5%. Have you students calculate MODs for those values.

Why not 32%= 31.1 - 32%? 32%= 31.5 - 32.5% doesn't make sense.

 

[dberry]

Why not 32%= 31.1 - 32%? 32%= 31.5 - 32.5% doesn't make sense.

I was assuming it measures more accurately and rounds up/down like most electronics. The sensors and basic electronics should give a signal with plenty of decimals, but they can probably save some pennies on the display by showing fewer.

31.1 should should round down to 31%, 32.6 would round up to 33%.

It's a matter of "significant figures". Would it be meaningful to report 32.0001%? That should round to 32, not 33. I think in this case there may be more significant figures from the electronics, but they've decided that the 1 decimal isn't "significant" to divers (and I agree.)

Unless, of course, they really skimped on the sensor and/or electronics, which I doubt, since there aren't that many companies that make O2 sensors out there.

 

[J-Vo]

I would be most concerned about compounding error from not calibrating as accurately, but generally you will still probably land within the margin of error for the cells anyway.

 

[BoltSnap]

I was assuming it measures more accurately and rounds up/down like most electronics. The sensors and basic electronics should give a signal with plenty of decimals, but they can probably save some pennies on the display by showing fewer.

31.1 should should round down to 31%, 32.6 would round up to 33%.

It's a matter of "significant figures". Would it be meaningful to report 32.0001%? That should round to 32, not 33. I think in this case there may be more significant figures from the electronics, but they've decided that the 1 decimal isn't "significant" to divers (and I agree.)

Unless, of course, they really skimped on the sensor and/or electronics, which I doubt, since there aren't that many companies that make O2 sensors out there.

The more conservative way, from O2 point of view, is to round any number with fraction to the next whole number%. Suunto computers do this in their O2 calculations. If you make 31.4% as 31%, then you are working with lesser O2% in MOD and O2 Tox calculations.

 

[dberry]

Good point about being conservative, but I'm not sure it's really significant when you consider all the variabilities and approximations that go into MOD limits and OxTox.

I'm not a fan of normalization of deviance, but I don't see this as an issue with +/- 0.5% pO2.

 

[FreqFlier]

Why not 32%= 31.1 - 32%? 32%= 31.5 - 32.5% doesn't make sense.

Paperwork (in german...) says accuracy is +/- 0,5 Vol.% of O2 at O2 range between 0 – 50%

https://nrc-international.com/cdn/shop/files/PickO2-Flyer-2016_D-low_1.pdf?v=386530234801376598

 

[BoltSnap]

It may not be significant but I'd rather take the conservative approach. I am not an expert by any means so I can't add more here.