Oxygen Sensor Fundamentals

Please register or login

Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.

Benefits of registering include

  • Ability to post and comment on topics and discussions.
  • A Free photo gallery to share your dive photos with the world.
  • You can make this box go away

Joining is quick and easy. Log in or Register now!

@stuartv and that again is where if he had paid attention to the mV's, then he would have caught that. You can see the voltage move with pressure. With a CMF though it certainly does increase the risk of an issue with calibration. My Mod1 course included ignoring the actual ISC checklist and making my own as we assembled the unit. After I thought I had it sorted, we compared against the ISC checklist and adjusted as necessary. My checklist changes every once in a while as a function of missing something that I didn't think needed to be on there. Every time I miss something, it gets put on the check list.
It sounds like your buddy didn't have that on his checklist, or didn't actually follow it properly.

If he would have paid attention to the ppO2 reading BEFORE doing the calibration, he would have also caught it - so, again, no need to look at mV. :)

The thing with check lists is, it seems to me, a really good check list doesn't just have all the steps you need. It also has a design that helps you to catch errors - for example, if you skip a step by accident. The guy I was talking to had made his own checklist. And I suspect that "turn O2 off" was on it. I suspect that he simply skipped that step by accident. And the design of the rest of his checklist did not cause that error to become glaringly obvious.

With the rEvo check list, the O2 has to be on. But, the nature of the steps is such that, if you skip that step of turning the O2 on, the next step or two will fail and you'll realize that you forgot to turn it on.
 
If he would have paid attention to the ppO2 reading BEFORE doing the calibration, he would have also caught it - so, again, no need to look at mV. :)

The thing with check lists is, it seems to me, a really good check list doesn't just have all the steps you need. It also has a design that helps you to catch errors - for example, if you skip a step by accident. The guy I was talking to had made his own checklist. And I suspect that "turn O2 off" was on it. I suspect that he simply skipped that step by accident. And the design of the rest of his checklist did not cause that error to become glaringly obvious.

With the rEvo check list, the O2 has to be on. But, the nature of the steps is such that, if you skip that step of turning the O2 on, the next step or two will fail and you'll realize that you forgot to turn it on.

On the mV, I don't have a DiveCAN ccr, but on my Meg it doesn't show ppO2's once you enter calibration menu, only mV.

That is the reason my instructor has me make my own. The one from ISC is ridiculously extensive and is going to make you skip steps. The key is to have the ones that are critical, not necessarily "obvious", or are easy to miss. The ISC has 42 lines on the assembly checklist and 19 on the calibration checklist. Mine has 11. My checklist was however looked over my by instructor who should have caught something like that if it's a glaring error.
 
On the mV, I don't have a DiveCAN ccr, but on my Meg it doesn't show ppO2's once you enter calibration menu, only mV.

When I do the Closed check list, I don't go into the calibration menu unless the ppO2 readings (with the loop filled with O2) are off from 0.98 by at least 0.02. I.e. unless I first determine that it actually needs to be calibrated.

If I do the O2 flush and look at my computers, they will generally be showing right around 0.98 and I don't even do a calibrate. If any sensors are showing 0.95 or maybe 0.96, I will do a Calibrate. If any were showing 1.03 (for example) or 0.90 or something like that, then I would know that I had done something wrong and would not just blindly do a Calibrate. If just one were off by that much, I would then be doing further checks to decide whether to replace that sensor.
 
Stuart,
If I may humbly suggest, start thinking in mV. I had a CCR diver, that I was mentoring, that started getting funky mV numbers and couldn't figure out why. His PPO2 seemed fine however, because I had taught him to think in mV, he knew that something was not right. When we discussed the issue I had him run through his checks with me on the phone. What we found was that he was flushing then closing the DSV and turning off the O2 which resulted in an over pressurization of the loop, as you had mentioned earlier. The problem was that he had already calibrated so it took a minute or so for his PPO2 to display 1.0 however it would not go over that. While on the phone I heard his OPV burp and knew right away that he was calibrated at an over saturated PPO2.

As the presentation, from the OP states, thinking in mV, checking linear deviation with mV, and 1.6 checks for cell limiting with mV will give you a much better understanding of your cells health. It also makes me more comfortable during a dive knowing the health of each of my cells. PPO2 is only a mathematical representation of a divers input, not necessarily an accurate representation.

You should still, IMHO, only calibrate when things are off however always verify by checking the mV output in air, O2, & 1.6 PPO2. If you can't get your mV numbers in air & O2 or your PPO2 is off, then you calibrate before getting to depth to do the 1.6 check.
 
Out of curiosity, when checking linearity, what level of deviation do you feel comfortable with. I.E., my cell is 9.8mV in Air, multiplied by 4.77 it should be 46.75 in oxygen. What amount off would bother you? If that cell showed 47.2 mV I personally wouldn't care. I know others that would.
 
Stuart,
If I may humbly suggest, start thinking in mV. I had a CCR diver, that I was mentoring, that started getting funky mV numbers and couldn't figure out why. His PPO2 seemed fine however, because I had taught him to think in mV, he knew that something was not right. When we discussed the issue I had him run through his checks with me on the phone. What we found was that he was flushing then closing the DSV and turning off the O2 which resulted in an over pressurization of the loop, as you had mentioned earlier. The problem was that he had already calibrated so it took a minute or so for his PPO2 to display 1.0 however it would not go over that. While on the phone I heard his OPV burp and knew right away that he was calibrated at an over saturated PPO2.

As the presentation, from the OP states, thinking in mV, checking linear deviation with mV, and 1.6 checks for cell limiting with mV will give you a much better understanding of your cells health. It also makes me more comfortable during a dive knowing the health of each of my cells. PPO2 is only a mathematical representation of a divers input, not necessarily an accurate representation.

You should still, IMHO, only calibrate when things are off however always verify by checking the mV output in air, O2, & 1.6 PPO2. If you can't get your mV numbers in air & O2 or your PPO2 is off, then you calibrate before getting to depth to do the 1.6 check.

Hey, Bobby, thanks for taking the time. No need to "humbly suggest"! I know you, respect what you have to say, and I'm always open to hearing advice from people who know what they're talking about.

And I do keep an eye on mV. I'm a numbers guy. I do things like calculate ppO2s, MODs, Best Mixes, etc. in my head, just to keep my brain from getting any duller. So, checking mV readings on sensors is a normal thing for me to note and cross-check.

But, it's not part of my formal, official process. And I still don't understand what value it would give to make it so.

By following my check list, if I made the mistake that your mentoree made, when I got to the step where I do a dil flush (during my Closed check, after assembling the unit at home), I would have seen that the ppO2 of my dil was not reading what it should have been. Calibrating to O2 in a pressurized loop would make my dil read as a lower FO2 than it should (I believe).

And, to me, the whole point of all of this discussion is that you are able to identify when something is wrong. Whether you identify it by seeing a mV reading that isn't right, or whether you identify it by seeing a ppO2 reading that isn't right doesn't seem to be so important. What's important is that your process ensures that you catch it when something is wrong.

In the case of your mentoree, once he saw that something was wrong, it seems like it should have been easy to walk back through his check list and realize where he'd made a mistake, then start over, do it again, correctly, and be good to go. The problem - it seems to me - is if following his check list would not have informed him that something was wrong. If the only way he would have known something was wrong was by looking at his mV output and doing a calculation to confirm the value, then that seems like his check list is failing him. Unless, I suppose, it's part of his check list to do the calculation and compare against the mV reading. But, that seems way more complicated than it needs to be.

Once I'm into a dive, the calibration is set. It's not going to change during the dive. So, looking at a ppO2 reading is the same as looking at a mV reading. It's just scaled to a different actual number. But, the ppO2 number is one that I'm much better equipped to easily recognize as "good" or "problem". If I see ppO2 = 1.5 when I'm on the bottom, I know that that's bad. If I see a mV reading of 75, well, I don't know if that's good or bad except by doing some math or looking at a chart or something. And why would I do that?

If my ppO2 is reading 1.1 and my other sensors say 1.3, then looking at the mV is not going to tell me anything new. That sensor is going to be reading the mV equivalent of 1.1 - whether it's current-limited to that value or not. The only way to know whether that sensor is right and the other 2 are wrong, or vice versa, is to do a dil flush. And with the dil flush, I'm not going to be looking at mV. I'm going to be looking at the Dil PO2 on the computer and the ppO2 readings from the sensors. Those PO2 numbers are what's going to tell me which sensor(s) are right, not mV readings.

When I get to the end of my dive, ascent to 20', and do an O2 flush, my sensors are going to read 1.55 - 1.6 or so - or they're not. Again, I won't need to look at mV readings to know if I have a problem. Again, the ppO2 # and the mV # are the same, just scaled differently.
 
Stuart,
Everything you said is spot on except one issue, which may or may not be in play for you and the types of dives you do. Yes you can check and verify with FO2 of dil and today a lot of our dive computers will even compute that for us. When doing deeper dives though and wanting to raise the PPO2 deeper to accelerate of gassing we need to verify that cells are not limited and/or humidity limited. Even then you would not "have to" use mV, I do gas switching with my CCR so that I have a FO2 at an appropriate decompression stop. A solid dil flush will show me where my cells are and how high they will show, then I simply keep a couple of points below that threshold and I know that the PPO2 is right.

Where learning to think in mV really helps is in tracking and knowing your cell health. All cells will change in linear deviation as they age. How much linear deviation is unacceptable for the brand of cells you are running and the type of diving you are doing? When is it time to replace them? I have heard things like "just rotate them out every six months".

I view cells like the brakes on my car. I get over 100,000 miles on a set of brakes however my wife gets far less. Why should we both replace our brakes every six months? That would be crazy. I pull 18,000 pounds with my truck regularly, have an exhaust brake and heavy duty brakes, and drive mostly on the highway. She drives mostly in town and has a mid size car.

Cells age depending on the brand (type of vehicle) and how they are used (type of driving). By tracking cells linear deviation and limiting, over time, with mV you will know how your cells are aging and their health. If you simply calibrate when things are off you don't know the percentage of linear deviation. By learning to think in mV you will be able to verify how a cell is acting at any time during a dive, I wish that I could leave the screen on mV, it would be easier for me. Cells dance and react completely different than the PPO2 that is displayed. The cells actual mV output is a much more accurate representation of what is happening with the gas PPO2. We already do a ton of rounding in diving and PPO2 is yet another layer of rounding to "smooth" out what we are seeing. We become comfortable and reliant on that rounding which leads us into a false sense of security. By thinking and watching mV we can see more granular detail and have a more clear picture of what is happening at the moment and over time.
 
Out of curiosity, when checking linearity, what level of deviation do you feel comfortable with. I.E., my cell is 9.8mV in Air, multiplied by 4.77 it should be 46.75 in oxygen. What amount off would bother you? If that cell showed 47.2 mV I personally wouldn't care. I know others that would.
Rddvet,
There are a lot of variables involved that decide what amount of linear deviation is acceptable for me. This is why thinking and knowing mV is so important. The brand of cell, the dive depth, dive duration, and other factors affect the amount of deviation I am or am not comfortable diving. Different cells fail in dramatically different ways. By tracking the linear deviation, checking for cell limiting, and seeing how your particular cells age and fail, for your type of diving, is the only way that I know to be comfortable with a cell. I have decades of experience on a vast variety of cells in a vast variety of diving conditions. That experience has given me a frame work to decide how much is too much. Also my level of comfort is likely different than everyone else. Cells behave differently in different CCR's as well, knowing what to expect and how they act in your CCR is critical.

Bottom line is that I can't, won't, and shouldn't say what is acceptable for anyone else. What I am saying is that by tracking linear deviation with simple math using mV we can all gain the experience and knowledge to make a well informed and educated decision on the health of our cells and when to replace them. I like that much better than using some arbitrary rule that is not based on math and logic.

It is not a matter of being cheap and running cells longer than the manufacturer recommends (though I personally do). Over the years I have seen bad batches come out and have replaced cells after just a few dives or a few hours. I've seen cells come out of the pouch and fail, yet they would calibrate and if I had not been looking at the mV could and have had some really odd outputs during the dive. I have run cells that I knew were a little off just to learn and understand what they were doing, which has helped when doing longer and/or deeper dives that result in similar issues coming up that I then understood what was going on and more easily dealt with it.
 
one thing to add to what @Bobby said to @rddvet question about acceptable levels of deviation
For me at least, the amount of tolerable deviation is directly proportional to the ppO2's that I'm diving. The closer you are to the calibration point, the more deviation you can tolerate. 80% linearity at the calibration point is still exactly the calibration point of 1.0. The farther away you choose to run your setpoints from the calibration point, the less deviation I would be willing to accept. That's a choice you have to make.

@stuartv remember also that calibration is inherently inaccurate on almost all units. With head calibration kits it is a lot less bad because you can be more confident in your flush, but if you're calibrating at a ppO2 of 0.98 but you are putting distilled O2 into the unit which is 99.999% O2, you have a built in 2% deviation from the calibration alone. The main takeaway points I have from many hours of discussions on this subject is that tracking mV's allows you to get a medical history of your cells over time, and it allows you to validate your calibration. Secondary beneficial point is that there is a built in time delay in the controllers to smooth the screens out from dancing about. Combine that with the scale factor and only reading to 3 sigfigs that are scaled at a multiple of ~5, you get a much clearer picture how the cells are behaving when you look at the mV's displayed. You can see them dance during a dil flush and see if one is starting to get sluggish or unstable much easier than by looking at the ppO2's which may hide what's going on.
 
I really learned a lot from Tbone1004 and Bobby about o2 sensors and linearity, and practicality of how to apply. I'm in the rotate every quarter camp, but that doesn't exclude knowing and checking cell health.

I like checklists, I like tracking stuff, and I like simple. The simple part helps ensure there is no draw to skipping steps. My dream was to get this checklist into an app (like google sheets, survey monkey, jotform) where I could enter the values easily, and get the data plugged into a spreadsheet. Then it would be possible then to flag automatically weird answers. For example, cell 1 in air at 10.5mv, but when o2 flush 40.5, could display an error message "Cell1 fail mv validation". Would make it easy to make a graph showing cells performance over time. Alas, all of those apps so far require internet connection, and TMobile doesn't have internet at 99.9999% of the caves, so paper it is.

I have a slot to record air mv, and cal (really o2 flush, I don't re-cal each dive) mv, and then I can jot the expected 20' mv for each cell somewhere to reference.

This condensed the dive rite checklist from a full page, to a quarter page, and captures more information. I still am tweaking it--will remove "solenoid hose and electronics connected" and "dsv direction checked" if I find a way to do that and maintain symmetry. Still searching for the electronic log option....

I'd also really like to build a pressure pot and Arduino/raspberry pi to graph the cells output at each po2, but pulling the cells from the op2tima is a tiny bit of a pain. Maybe someday there will be a milled cap to twist on to pressurize and pull data from the pod connector...

modified op2tima checklist.png

o2 mv expectation.png
 
https://www.shearwater.com/products/peregrine/

Back
Top Bottom