Cell Checker - What am I doing wrong?

dunny

dunny

Member
I bought a mini check from Narked @ 90 and I've just done my first check of the cells. All of the cells manufacture dates are 2015-04 for a JJ rebreather. See below for results:

Results.png


So the good news the cells are appear to be linear. The bad news they fall outside of the expected mV range but I can't figure out what drives this EXP mV data since the sheet is protected. Also Cell 2 is hotter than Cells 1 & 3 (is that correct terminology) but still linear. So based on this I would still dive these since they are linear but I was curious what other peoples thoughts were?

I'm going to phone John at Narked @ 90 tomorrow and ask his views too but in the mean time my curiosity has got the better off me so I thought I would ask the members of CCRX for help :)

Cheers
Phil
 
Are you sure that the cell mv in air at 1ATA is accurate? 12.0, 13.0, and 12.0 seem pretty high to me.
 
Your O2 is not pure (fO2 < 1) ?
Your local atm pressure is above 1 ata (P0 > 1)?
Your cell checker pressure transducer is reading higher than it should?

I know your report says ambient pressure = 999 mbar (= 0.986 ata), so the first and last hypotheses might be worth checking (get another O2 source and compare; put your SW in the pot and try to read out what depth it returns).

In all cases, your cell checker will read low:

V = P/P0 x fO2 x V_air

where V_air is the mV read in air and V the mV read at whatever pressure P you are.
What is surprising is that the relative difference is increasing, which in either situation above, is not expected.
However, if you look carefully at the reported % values, you'll notice that they are not correct: take the P0 value I quoted above, the assume fO2 of 1 and the mV reading in air of the first cell and compute:

(V_meas - V_exp)/V_exp, where V_exp is the "expected" value and V_meas the measured value (which I take to be the ones found in the "white" columns and you'll find the following results for cell 1 (similar results are obtained for the other cells):


0
-8.4
-10.7
-10.3
-9.8
-9.6
-9.6
-10.4
-9.8
-10.2
-10.0

You'll notice that there relative error is constant (within the margin of uncertainty), as expected... that is if you have one of the factors wrong (see above).

The only explanation I can find to the increasing relative error is that they use the strange (and wrong) formula:


(V_meas - V_exp)/(10*V0)

So in summary, you have 3 different possible reasons (and maybe others) to check.
 
I analysed the o2 cylinder before I started and it read 100%. I used the mbar output on the petrel which read 999mbar.

How could I confirm the pressure transducer is reading correct?

Sent from my SM-G925F using Tapatalk
 
Well, my O2 generally read 102% (by that I mean that it is quite difficult to measure O2 unless you first calibrate your sensor with a known 100% O2 source)... If you did calibrate your sensor in air, you might well be off by a few %.

Put a couple of SW (or any dive computers) in the pot and make a dive to 2 ata with them (stay some time at each depth).
Read the logs. If they match to one another but not with the pot, you've got your culprit.
My Petrel and my Predator are off by a couple of ft at depth (~150 ft), which is 1%, so not much of a difference, but I have read about transducers going bad. This could be the case (or not) of your pot.

There could be other factors: humidity, temperature (difference between ambient outside and at depth inside, where you have flushed your pot with hopefully dry O2)...

The bottom line though, is that your sensors are linear up to 2 ata (unlike suggested by the varying percentages of error reported by the pot).
 
I don't ever find the cell-checker very accurate on that chart because I find the gauge on the mini-check hopeless as the mv output is to 1DP and the gauge is "by eye".

How are you reading the values in air? On the dining room table? Doing it with air in the pot is the best way to get the temperature and humidity close to that you will get on entering O2.

The plotting formula used Ambient Pressure, Temp/C and RH% so you need to get those right for the gas in the pot after you have flushed.

I try doing it "upside down" by getting the expected mv first and cross checking the pressure is "fine" on the gauge.

Then you can adjust the RH and Temp to get the dotted line right and see if you believe the values. Unless you can measure them, in the pot.

I don't find the cell checker very useful and have not used it in quite some time. I did have all the problems though so figured out quite a lot about what I thought was going on.

Matt.
 
My shearwater and my ISC handset don't read the same mV on the same cells. The shearwater (via fischer cable) is high by 1 to 1.5mV in air. I don't know why but since its just a monitor and not actually controlling my ppo2 I don't worry about it.

By any chance have you confirmed your cells with a voltmeter? 13 mV in air for a used 9-13 mV cell seems high to me. At least my used ISC/Vandergraphs seem to settle at about 9.9 mV in use. They are also, in theory, 9-13 mV cells. I don't have that many cell changes to based this off of, but I'd be shocked if other voltmeters confirmed the 13 mV in air value on a 4 month old used cell.
 
I thought cells were generated a current based on PP02? So mv reading can be different based on the resistance of the device
 
mick4707 said:
I thought cells were generated a current based on PP02? So mv reading can be different based on the resistance of the device
Click to expand...

Yes but in this case all 3 cells look pretty high. For a slightly used cell in air at very close to sea level.

Perhaps there was some O2 dribbling into the cell checker?
 
matt said:
I don't ever find the cell-checker very accurate on that chart because I find the gauge on the mini-check hopeless as the mv output is to 1DP and the gauge is "by eye".
Click to expand...
The gauge is pretty hopeless as you say and very difficult to eyeball. That said, one gets a feel for it and, if you don't get hung about absolute accuracy, the cell checker is a very useful tool for ensuring that the cells are more or less linear in their response and, more importantly, aren't current limited.

How are you reading the values in air? On the dining room table? Doing it with air in the pot is the best way to get the temperature and humidity close to that you will get on entering O2.
Click to expand...
You take the air reading prior to installing the cells into the pot. I do this in my workshop or on my dining room table, whichever is most convenient. Unless you introduce some humidity into the checker, the gas inside is irrelevant. Also, the temperature is only going to change as a result of the increasing pressure within the pot, unlike the loop in 'real life' which will be affected by the sea temperature and the effects of the reactions taking place within the scrubber.

The plotting formula used Ambient Pressure, Temp/C and RH% so you need to get those right for the gas in the pot after you have flushed.
Click to expand...
I have no insight into the formulae used in the spreadsheet and the humidity/temperature setting are very granular but, having experimented with adding/subtracting from both parameters it doesn't make much difference in my experience.

I try doing it "upside down" by getting the expected mv first and cross checking the pressure is "fine" on the gauge.
Click to expand...
I do this too, also cross-checking by doing the measurements 'in reverse'.

I don't find the cell checker very useful and have not used it in quite some time. I did have all the problems though so figured out quite a lot about what I thought was going on.
Click to expand...
I still think it's a great tool to 'sanity check' my cells but it's a tool that needs to be used intelligently - just like a rebreather ;)
 
Thanks for the help guys with a bit more practice I've managed to get the mV output to within 4% of expected values which is good enough for me!

Thanks again

Sent from my SM-G925F using Tapatalk
 
nickb said:
I have no insight into the formulae used in the spreadsheet and the humidity/temperature setting are very granular but, having experimented with adding/subtracting from both parameters it doesn't make much difference in my experience.
Click to expand...

I don't have the sheet on this machine but if you fiddle with the values you can adjust the position of the baseline a bit - as you say, not a lot. Probably enough in this case to remove the perceived error.

I still think it's a great tool to 'sanity check' my cells but it's a tool that needs to be used intelligently - just like a rebreather ;)
Click to expand...

I used it quite a bit during the cell-fiasco, but no longer use it routinely as fresh cells can easily be obtained and I typically test them in the water on first a check dive and then at the end of the dives for linearity.

I also took the advise and started rotation so I have cells form all different dates and ages.

I also find it useful to browse through the pO2 data from all 3 cells across the dive and look for oddities post dive. Sometimes turns up a problem ahead of it being a problem. The APD Viewer is not bad for that, and you can also paste it into XLS - I guess Shearwater etc. can do the same.

Cheers
Matt.
 
matt said:
I used it quite a bit during the cell-fiasco, but no longer use it routinely as fresh cells can easily be obtained and I typically test them in the water on first a check dive and then at the end of the dives for linearity.

I also took the advise and started rotation so I have cells form all different dates and ages.

I also find it useful to browse through the pO2 data from all 3 cells across the dive and look for oddities post dive. Sometimes turns up a problem ahead of it being a problem. The APD Viewer is not bad for that, and you can also paste it into XLS - I guess Shearwater etc. can do the same.

Cheers
Matt.
Click to expand...

I agree but its nice sitting on the boat knowing when you prepared the unit the cells where linear to 2bar the day before... rather than 3weeks before when you last dived (for example) I have also seen signs of drop off at high pressure which has directly pointed to the cell which later fails to be linear in the usable range.

FWIW I have a full fat N@90 cell checker which has built in MV gauge but it tracks pretty close to what the SW reads, if I made one improvement it would be to the pressure gauge and O2 add, I would prefer a bigger gauge and needle valve to get the pressure dead on.
 
You must log in or register to reply here.
Back
Top