Diluent flush and current limited cells

[sheck]

Hi everyone,
First off, I apologize in advanced for my broken English,as I'm not a native speaker, and for any silly assumption since I haven't even finished my Mod 1.
I was just wondering how a diluent flush at depth can be helpful to check cells.
As I understand, some cells may read ok a low ppo of, let say 1.0, but fail to read higher than that. So, if my set point is 1.2/3, wouldn't the controller keep adding o2 rising the ppo2 dangerously without the cells telling me. In other words, the flush would show the ppo2 for that depth which is supposed to be low but wouldn't hint that a cell is faulty because it does read the low ppo but not a higher one.

Dies it make sense?

Thanks a lot

Sheck

 

[DSix36]

A diluent flush at depth will tell you if the cells are reading accurately but not if they are current limited. On oxygen flush at 20' will tell you if the cells are current limited.

A diluent flush is more of a tool to verify voting logic to know which cells to trust if they are not all reading the same.

 

[matt]

There's a few things I do to check cell health, Dil Flush can be one of them. Here's my list:

- check the cell's output with the previous calibration values. I like to do this in air with the lid off and expect to see 0.21 0.21 0.21
- check the end cal-values with the previous end-cal-values. On the inspo this is the final values given in calibration and are standardised mv outputs with the decimal shifted
- check the cell will reach 1.6. I do this with O2 at 8m on ascent

The first shows any issues with the wiring or connections as well as exposing a failing cell. If I don't get close to 0.21 then there is something wrong that needs fixing.

The second is more likely to show a failing cell due to age. Aged cells calibrate high, that is they give more output as they start to die. However any big deviation from the last time is a cause for concern. Just because you can calibrate does not mean the cell is healthy.

The third check for current limiting. This failure mode is treacherous for the reason you described; the controller may keep adding O2 and the limited cell(s) won't rise correspondingly.

In addition I always fit cells from different batches. Here is a good article on cells: http://www.revo-rebreathers.com/wp-content/uploads/2016/02/Understanding_oxygen_sensors.pdf

Personally I only use a dil-flush as a cell check if I suspect a problem at depth. In my experience it is not easy to do on eCCR if the SP is set higher than the dil's PO2. But is easy and a good test if the O2 is off or lower.

It's also worth noting that new cells are just as problematic as old cells and really all cells can die at any time in a range of ways, so it is not safe to assume new cells are better than old ones and you do need to keep your eye on them.

Enjoy the Mod1.

Matt.

 

[Petemes]

Hi Sheck,
This debate has got manys a forum exploding with everyone putting their $0.02c in. Here is mine. Of course any recommendations will be of little help to you as I am sure there are those who will give you 100% conflicting advise. But here goes! People will argue that a dil flush will NOT show current limiting. This, in essence is correct. But what a dil flush can do is tell you a lot. For example. You are on a dive and your cells read 1.28 1.29 2.55. Ok classic example of current limiting. Two cells are controlling the O2 into the unit by voting logic. So the 2 cells cant get to 1.3 so the controller keeps adding O2 into the loop. So - what to do? You do a dil flush. This immediately reduces your liklihood of breathing high PO2 at dangerous levels (if it is the case). It lowers the cell read out. Then you switch the unit to LOW setpoint and watch the cell readout ( all three cells reach the low setpoint easily). Then manually add O2 to bring it up to your 1.3 set point. All three cells are behaving themselves (all keeping within their parameters) up till a little over 1.2. Now you are vigilant. Adding more to bring it up to 1.3. Cells one and two cant get there and 3 gets to 1.3 and over. So now if you reduce your setpoint to 1.0 all three cells will be able to reach that setpoint. This is what I would do.

There are also lots of tail tail signs too. We all know when you are at a constant depth only small amounts of O2 are ever injected to replace matabolised O2. So if you hear your solenoid constantly adding - this is tail tail #1. #2 is you are becoming more positively buoyant doe to more O2 going into your loop.

All I do is change all my cells annually. (I dont go for rotating cells, personally. - (Another hot topic) This is just my personal preference.

Pete Mesley

 

[sheck]

I see. It's just another tool that gives you information on cells health. But can't be trusted 100%.
Thanks for the tips guys.
This is complex OMG

 

[Randy Thornton]

I see. It's just another tool that gives you information on cells health. But can't be trusted 100%.
Thanks for the tips guys.
This is complex OMG

Sheck, I am sure your instructor will cover this topic in detail during your course, but as Mr. Mesley has indicated above, the dil flush is simply a tool to use as part of a diagnostic procedure. When you perform a dil flush correctly, there should be no second guessing about what your 3 cells should be reading at depth. It's simply a matter of mathematics.
Understanding your cells expected behavior and readings is simply part of learning to dive a rebreather. I'm sure it will all make more sense by the time your course is complete. Hang in there! It get's better!

Kind regards,
Randy

 

[ECho]

It's simply a matter of mathematics.

If you happen to have a Shearwater electronics, you don't even need any maths... Two pushes on the right hand button will give you what the dil PO2 should be for your current depth on the bottom row, for easy and instant comparison to you sensor readings. If any of the sensors don't agree with provided value, you know they've gone sideways. It doesn't get much easier than that.

Good luck with the rest of your training, and welcome to the dark side!

 

[paulraymaekers]

There's a few things I do to check cell health, Dil Flush can be one of them. Here's my list:

- check the cell's output with the previous calibration values. I like to do this in air with the lid off and expect to see 0.21 0.21 0.21
- check the end cal-values with the previous end-cal-values. On the inspo this is the final values given in calibration and are standardised mv outputs with the decimal shifted
- check the cell will reach 1.6. I do this with O2 at 8m on ascent

The first shows any issues with the wiring or connections as well as exposing a failing cell. If I don't get close to 0.21 then there is something wrong that needs fixing.

The second is more likely to show a failing cell due to age. Aged cells calibrate high, that is they give more output as they start to die. However any big deviation from the last time is a cause for concern. Just because you can calibrate does not mean the cell is healthy.

The third check for current limiting. This failure mode is treacherous for the reason you described; the controller may keep adding O2 and the limited cell(s) won't rise correspondingly.

In addition I always fit cells from different batches. Here is a good article on cells: http://www.revo-rebreathers.com/wp-content/uploads/2016/02/Understanding_oxygen_sensors.pdf

Personally I only use a dil-flush as a cell check if I suspect a problem at depth. In my experience it is not easy to do on eCCR if the SP is set higher than the dil's PO2. But is easy and a good test if the O2 is off or lower.

It's also worth noting that new cells are just as problematic as old cells and really all cells can die at any time in a range of ways, so it is not safe to assume new cells are better than old ones and you do need to keep your eye on them.

Enjoy the Mod1.

Matt.

as long as everybody understands that current limiting of sensors usually only kiks in when you are already diving for a short time, usually not at the beginning of a dive: there is a misconception about it!
(unless your sensors were already current limited for a few months..)

see: http://www.revo-rebreathers.com/wp-...ygen-Sensors-part-II_the-complete-picture.pdf

 

[Mark Chase]

I see. It's just another tool that gives you information on cells health. But can't be trusted 100%.
Thanks for the tips guys.
This is complex OMG

I dont know why you think it cant be trusted?

I havent read all the responces but ill run you through what you should do:


1: Switch to low set point

2: Tripple flush the unit with diluient which is above low set point

3: Read the PP02 displayed and match that with your known diluient PP02 at your current depth. So lets say your on 18/45 at 40m your display after a triple flush should read 0.9ppo2


If its reads 0.9 0.9 0.9 all three cells are able to read to 0.9

If it reads 0.9 0.3 1.0, cell 2 is screwed cell 3 is reading high and possably burning hot / about to fail Cell 1 is reading spot on.

4. Manualy inject o2 in short bursts and watch the PP02 go up. If it goes up to set point rise the PP02 above set point to say 1.6pp02 If the cells all read up to 1.6 you know they are working. Anything that is "current limited" will crap out at some point before 1.6 and you will be able to see this

5: If the cells are current limited to lets say, 1.3 you have two choices. 1: Bail out and abort the dive (100% safe) or 2: FLy the unit home with what valid info you have. IE in the example above Cell1 is reading OK but you might decide to run 1.2 out to be on the safe side.

This HAS to be manual. You can not allow voting logic to control 02 add on faulty cells.

IF you chose option 2 and fly the unit home yourself I STRONGLY sugest flying it on low set point with manual add 02.

BEFORE you EVER decide on Option2 you must gain suitable experiance on flying the unit home manualy. By which I meen ending the dive and ascending on low set point maintaining high set point by manual injection. You need to do this regularly to have a muscle memory for manual injection which will allert you to a problem in the event of a cell failure if the unit starts to demand a daft amount of 02.


Too much hassle?

OK just bail to OC and go home. There are no awards for stile in the event of a CCR failure only for surviveing it.


Personaly I run low set point on every dive down to the max depth on entry. At that point I do the above cell check EVERY dive before I start the actual dive part. The upside of this is a real world cell check befroe the bottom phase and a routeen that becomes so quick and easy, doing cell health checks later on in the dive is a easy option.



A final note on PP02 at depth.

SOme computers like my Sheerwater have a display which will indicate PP02 at current depth on current diluient

If not then have wet notes in your pocket with PP02 at say 3m intervals arround max planned dive depth

HTH

Mark

 

[sheck]

Sheck, I am sure your instructor will cover this topic in detail during your course, but as Mr. Mesley has indicated above, the dil flush is simply a tool to use as part of a diagnostic procedure. When you perform a dil flush correctly, there should be no second guessing about what your 3 cells should be reading at depth. It's simply a matter of mathematics.
Understanding your cells expected behavior and readings is simply part of learning to dive a rebreather. I'm sure it will all make more sense by the time your course is complete. Hang in there! It get's better!

Kind regards,
Randy

Thanks Randy.
Coming from oc tech diving this is proving to be a whole new world!

 

[sheck]

I dont know why you think it cant be trusted?

I havent read all the responces but ill run you through what you should do:


1: Switch to low set point

2: Tripple flush the unit with diluient which is above low set point

3: Read the PP02 displayed and match that with your known diluient PP02 at your current depth. So lets say your on 18/45 at 40m your display after a triple flush should read 0.9ppo2


If its reads 0.9 0.9 0.9 all three cells are able to read to 0.9

If it reads 0.9 0.3 1.0, cell 2 is screwed cell 3 is reading high and possably burning hot / about to fail Cell 1 is reading spot on.

4. Manualy inject o2 in short bursts and watch the PP02 go up. If it goes up to set point rise the PP02 above set point to say 1.6pp02 If the cells all read up to 1.6 you know they are working. Anything that is "current limited" will crap out at some point before 1.6 and you will be able to see this

5: If the cells are current limited to lets say, 1.3 you have two choices. 1: Bail out and abort the dive (100% safe) or 2: FLy the unit home with what valid info you have. IE in the example above Cell1 is reading OK but you might decide to run 1.2 out to be on the safe side.

This HAS to be manual. You can not allow voting logic to control 02 add on faulty cells.

IF you chose option 2 and fly the unit home yourself I STRONGLY sugest flying it on low set point with manual add 02.

BEFORE you EVER decide on Option2 you must gain suitable experiance on flying the unit home manualy. By which I meen ending the dive and ascending on low set point maintaining high set point by manual injection. You need to do this regularly to have a muscle memory for manual injection which will allert you to a problem in the event of a cell failure if the unit starts to demand a daft amount of 02.


Too much hassle?

OK just bail to OC and go home. There are no awards for stile in the event of a CCR failure only for surviveing it.


Personaly I run low set point on every dive down to the max depth on entry. At that point I do the above cell check EVERY dive before I start the actual dive part. The upside of this is a real world cell check befroe the bottom phase and a routeen that becomes so quick and easy, doing cell health checks later on in the dive is a easy option.



A final note on PP02 at depth.

SOme computers like my Sheerwater have a display which will indicate PP02 at current depth on current diluient

If not then have wet notes in your pocket with PP02 at say 3m intervals arround max planned dive depth

HTH

Mark

Thanks a lot Mark.
I have a much more clear picture of the whole process now. I am in fact asking my instructor to focus on manual 'piloting' with low set point and plan to use it in the future for the reasons you just described.
This is really a great place to look for answers to the many questions that arise with CCR diving.
Thanks to every one for all the good advice.

Sheck

 

[Ben M]

2: FLy the unit home with what valid info you have. IE in the example above Cell1 is reading OK but you might decide to run 1.2 out to be on the safe side.

This HAS to be manual. You can not allow voting logic to control 02 add on faulty cells.

The other issue to consider with this option, is your deco information can be incorrect. If you aren't running deco from a standalone computer, the computer that you have decided voting logic to be incorrect to control your setpoint will still use the same incorrect logic to compute your deco.

 

[matt]

You can also get the current Dil PO2 on a Vision.

Matt.

 

[sheck]

The other issue to consider with this option, is your deco information can be incorrect. If you aren't running deco from a standalone computer, the computer that you have decided voting logic to be incorrect to control your setpoint will still use the same incorrect logic to compute your deco.

Which rises another question for me... My unit comes with a predator controller and a fisher cable to add something else. I'm still deciding what to do. a HUD? a EXT pretel or a satand alone petrel?
For the reason you described above, would I be better off by getting a standalone petrel with no cell info but just set point and correspondent deco info.

Thanks

Sheck

 

[Jeff Pack]

I dont see the reasoning behind lower setpoint and manually adding, instead of just letting the unit kick back in. Whether the unit adds, or you add, its still being added, and without the hassle of changing to manual or a lower setpoint. Manual add to get cells up to 1.6 though.

You just dill flush, knock the PPO down to known PPO of depth, and watch the cells come back up. Lag, or limited, or both will show as the O2 is added back in. Then you may need to switch to manual or a lower SP, depending on outcome.

During the dive, sometime we'll hot drop, I let the PPO spike 1.4, or 1.5 just so that I know its not current limited above my setpoint either.

 

[matt]

Which rises another question for me... My unit comes with a predator controller and a fisher cable to add something else. I'm still deciding what to do. a HUD? a EXT pretel or a satand alone petrel?
For the reason you described above, would I be better off by getting a standalone petrel with no cell info but just set point and correspondent deco info.

Thanks

Sheck

Add a NERD? If you have 2+ bust cells leading to voting logic errors (low readings extend deco so are less of an issue) then this is a corner case; and you have bigger issues than deco to worry about.

Matt.

 

[sheck]

Add a NERD? If you have 2+ bust cells leading to voting logic errors (low readings extend deco so are less of an issue) then this is a corner case; and you have bigger issues than deco to worry about.

Matt.

Matt,
Why would a nerd help in that situation over a normal HUD? Sorry If I don't see something obvious, but everything is new to me and I'm still trying to get up to speed!
Thanks
Sheck

 

[matt]

It wouldn't help in that situation, I was answering 2 questions in one post. Nothing helps in that situation other than being able to diagnose WTF.

NERD is a type of HUD, I guess. NERD is a computer with a Head-up Display, rather than a PetrelEXT which is a computer on your wrist. If you have a NERD and a Predator Controller you don't need a LED-HUD.

Matt.

Matt,
Why would a nerd help in that situation over a normal HUD? Sorry If I don't see something obvious, but everything is new to me and I'm still trying to get up to speed!
Thanks
Sheck

 

[Mark Chase]

I dont see the reasoning behind lower setpoint and manually adding, instead of just letting the unit kick back in. Whether the unit adds, or you add, its still being added, and without the hassle of changing to manual or a lower setpoint. Manual add to get cells up to 1.6 though.

You just dill flush, knock the PPO down to known PPO of depth, and watch the cells come back up. Lag, or limited, or both will show as the O2 is added back in. Then you may need to switch to manual or a lower SP, depending on outcome.

During the dive, sometime we'll hot drop, I let the PPO spike 1.4, or 1.5 just so that I know its not current limited above my setpoint either.

If your dill is .9 at depth and your running 1.3 set point when you dill flush the unit will start to inject 02 to try and get you back to 1.3 ASAP so your PP02 readings will be wrong instantly.

So you have to be on low set point of say 0.7 to prevent the unit injecting 02 whilst you properly confirm the cells

Voting logic doesent take into account cell failure so if your cells read 1.2 2.0 1.2 but your dill flush concludes Cell 2 is the only one not current limited the unit will continue to inject 02 trying to get 1 & 3 up to 1.3

Didnt someone die recently from doing just that?

Once in the position of knowing you have a cell or cells that are current limited taking over manual control of 02 injection completly avoids any possibuility of the unit putting you in danger of 02 tox by adding too much 02 without you knowing. It also makes you focus on the display more and in doing so your much more likely to spot further odd cell behaviour

If you have a current limited cell its pretty unlikley it will be current limited to the point it cant read a life sustaining 0.16 pp02

Running a lower PP02 will add a bit of deco and you might get out of the water unknowingly 15mins earlier than you were suposed too for you given deco profile and yes if you didnt think about this and have a back up deco plan you might get bent.

02 tox and you are prety much assured of being dead

Id bet the house on being able to do a manual inject deco assent totaly blind without ever dropping below life sustaining PP02 or going above 1.6pp02 for any significant period of time

Below is a dive where I had a total solinoid failure on min 18 of the dive.

You can see the 0.9pp02 cell check at the start of the dive and then for the remaining 2 hours of the dive I just ran the unit manualy. Its realy not that hard to do

 

[matt]

So you have to be on low set point of say 0.7 to prevent the unit injecting 02 whilst you properly confirm the cells

This. And the reason that this is not always easy is that you need to be deeper than your low SP to do it properly. [i.e. shallower than 25m for an air-dil]. If you try doing it at high-SP it really doesn't tell you much at all.

Didnt someone die recently from doing just that?

Try this: https://cognitasresearch.files.word...the-inquest-into-the-death-of-philip-gray.pdf

There have been others, especially one I recall diving with very old cells believing they were fine, but actually being current limited exactly as you said. If you have 2 or 3 cells limited at less than high-SP and you don't notice and fix the high reading (2 cell) or the solenoid activity then you are in grave danger. This is what persuaded me to move from 3 new cells a year to 1 cell every 3-4 months.

Matt.