Cell Linearity

Genesis · 19th November 2006, 17:25

This is one of those areas in Rebreather diving that became blindingly apparent to me when I started working on the K1 electronics, and ended up driving part of the code design.

My first "lesson" came about when I started modelling and synthesizing various failure modes, including intentionally dunking the head board (un-potted) in water while the unit was on to simulate a failure of the isolation.

As expected the unit freaked out - but what was not expected is that the leakage of current into the cell inputs caused the cells to spaz for a very long time.

In one case the cell was effectively destroyed - it read low forever after that.

In several others, however, I got really STRANGE high readings for a variable amount of time - and not a small amount of time either! One sensor read high for more than 24 hours, although it later returned to normal. (!)

The original design for the K1 had two handsets - a full eCCR one and a second handset that was a monitor only (no solenoid control.) That second handset was scrapped after these tests - it became apprent that after an electronics failure you could not trust anything connected to the failed unit, and that included the O2 sensors! So unless you were prepared to run six sensors in the head you were functionally better off with a computer link (e.g. VR3) with a fourth sensor than redundant handsets, because the VR3 connection was electrically isolated from the other cells.

I also had a sensor in an O2 meter that went EOL on me, and started experimenting with it, finding the voltage-limiting behavior quite easily. That was disturbing as well.... yet another way those infernal things can kill you.

Anyway, all of this led to the "validation" requirement in the K1 code, and the decision to single-point calibrate on air. In reality single-point cal on O2 at the surface with a subsequent removal of the head to validate on air would work just as well. It did, however, make clear that two-point calibration was downright dangerous in the face of a cell fault - you could very easily be led to believe that the cells were working when in fact you had set a slope that was not actually linear at hyperbaric PO2 levels! Using the "implied zero" of single-point cal avoids this issue.

I firmly believe that the only way to be reaosnably certain that the sensors are going to read correctly is to see them doing so in a proven environment with a HIGHER Po2 than you intend to trust them at when running on "automatic." Whether you do that "out of the box" in a dedicated cell checker, you do it voluntarily on the way down as part of every dive, or you are forced to do it by the electronics in order to run a hyperbaric setpoint isn't the important detail. That you do perform the check is, because if you don't and there's a problem here you are quite likely to take an O2 hit without warning and die as a consequence.

rdmmdr · 19th November 2006, 17:37

cool distructive testing is allways fun. now where did i put that can of factory installed smoke for the repairs. nothing works after that pesky smoke gets out.
rick

dave t · 19th November 2006, 17:41

Karl

is it not possible to buffer the cells so they dont react to stray voltage and the like hitting them?

I was one of the early crash test dummies for the HH electronics on one dive I had a primary flood at 80m, the primary freaked and the hud scared the crap out of me with its flashing and vibrating but the secondary po2 readout remained perfectly in line with the other two cells I had in the loop (5 in total) which were hooked to a second passive display. I assumed from conversations with KJ that the "buffering" would stop the crapped out primary affecting the reading. I may have misunderstood what Kev told me but I managed to complete all my deco on the loop (another hour or so) before the secondary readings started to be affected.

I should add had I not had an independant back up po2 meter I would not have trusted the secondary on its own

Dave

Dave

AD_ward9 · 19th November 2006, 18:03

Quote: (Originally Posted by Genesis)

My first "lesson" came about when I started modelling and synthesizing various failure modes, including intentionally dunking the head board (un-potted) in water while the unit was on to simulate a failure of the isolation.

As expected the unit freaked out - but what was not expected is that the leakage of current into the cell inputs caused the cells to spaz for a very long time.

In one case the cell was effectively destroyed - it read low forever after that.

That was because you are using the wrong cells.

It is very important when designing any life critical electronics, to establish a baseline of what performance you require for each component, then test the component to that. It is very good to see you are doing that. As an aid to assessing what tests are needed, the test regime we use for cells is attached.

Our views on the poor performance of the Teledyne cells are well known, based on thorough testing. It never fails to astonish me how many companies just design these in without testing at all. For the cell to be suitable for a rebreather, dumping a cell into water should not change its characteristics, nor should shorting it out (the cells have a low resistance load on them anyway, internally).

I am working on getting released onto this forum, the results of a 5 year study of O2 sensors, which gives all these characteristics.

Alex

AD_ward9 · 19th November 2006, 18:10

Quote: (Originally Posted by am)

It's really very simple. The number corresponds to the year and the letter corresponds to the month.

So for example 5A would be January 2005 and 6D would be April 2006.

Regards

AnneMarie

This is for Teledyne's cells. Each manufacturer is using their own codes. Good of you to highlight these codes, as a surprising number of divers have forgotten them and are diving on three cells with the same date code (a dangerous practice).

For rebreather use, cells should be marked much more clearly with the date, e.g. "SEPT 06", using an ink that does not vanish if the label rotates (as it does on the Teledynes).

Alex

AD_ward9 · 19th November 2006, 18:28

Just to bring to the attention of those interesting in cell linearity: O2 cells are not linear with pressure.

The cell sensitivity changes with ambient pressure, given constant PPO2, by up to 20%.

Keep PPO2 constant, then pressurise the cell slowly, and watch the output.
The design using the cell, should compensate for pressure in taking PPO2 measurements. Generally the sensitivity increases with pressure, so this is not a critical safety issue (at depth, the PPO2 will be lower than that displayed in most designs, but not low enough to cause hypoxia).

Alex

dive2dive2000 · 19th November 2006, 21:47

Quote: (Originally Posted by jradomski)

martin,
even if the cells are in the right range for 100% oxgen this doesn't guarantee that it will be accurate for anything above this...

I have seen cells that looked fine at a 1.0 po2, but when the controller was showing a 1.3, the real po2 was over a 1.5... These cells were't quite current limited at 1.3, I did manage to get them to just over 1.4 but I needed a po2 of almost 1.8 to do it... (was off loop with 100% o2 in loop and descended down to 9m, at 8m it maxed out)

from general observation, the higher the cells normal output is in air, the earlier it the cell appears to start going non-linear.. And this makes sense since it takes a greater chemical reaction to generate the higher voltages(currents).. This assumes cells of similiar sizes and diffusion rates.. If you go to larger cells where you can increase the electrolyte/lead and increase the diffusion membrane, this behavior would lessen..

The PRISM (as well as other Rebreathers) requires a cell that operates in a mv range 16-22mv. The reason for this is so you will have a enough mv for high ppo2 readings. It saddens me that manufactures would place low mv cells in Rebreather

I took for granted that this was across the board, after all this is not new information and high out put cells have also been around for the last 10 yrs or better.

AD_ward9 · 19th November 2006, 22:27

Quote: (Originally Posted by dive2dive2000)

The PRISM (as well as other Rebreathers) requires a cell that operates in a mv range 16-22mv. The reason for this is so you will have a enough mv for high ppo2 readings. It saddens me that manufactures would place low mv cells in Rebreather

I took for granted that this was across the board, after all this is not new information and high out put cells have also been around for the last 10 yrs or better.

Low output cells last longer and have a better response time. Low output cells have a lower output impedance, which can be useful. If the electronics is designed properly, it does not matter whether the cell output is 1mV or 100mV. "Designed properly" includes measures to eliminate/reduce noise.

Alex

**Gill Envy** · 19th November 2006, 22:32

Quote: (Originally Posted by Louby Lou)

It is still here, although it is closed.

The thread is HERE

the link appears to be dead.

dive2dive2000 · 19th November 2006, 23:12

Quote: (Originally Posted by AD_ward9)

Low output cells last longer and have a better response time. Low output cells have a lower output impedance, which can be useful. If the electronics is designed properly, it does not matter whether the cell output is 1mV or 100mV. "Designed properly" includes measures to eliminate/reduce noise.

Alex

I am getting lost, you are now throwing a whole bunch of information into the mix

It was my understanding that with low output cells there will come a point where you will not be able to read a 1.6 or higher?

I know for my unit and my cells it reads true as long as I maintain cells that produce mv between 16-22 As far as other cells and other manufacturers and instructors go they should be versed in the specs that apply to them and should be teaching this in there courses. The original thread's main theme was basically is a pressure pot a necessity for testing cells. IMHO it is not. A volt meter your cells operating range and a swim to 9m is all you should need. Genesis and you provide very interesting information but it is way over my head

19th November 2006, 17:25	#21 (permalink)
Genesis Bubbless Box of Death Current Rebreather/s: Home Build Other Rebreather/s: Home Build Join Date: Oct 2005 Location: Sunny Florida Posts: 1,395	Re: Cell Linearity This is one of those areas in Rebreather diving that became blindingly apparent to me when I started working on the K1 electronics, and ended up driving part of the code design. My first "lesson" came about when I started modelling and synthesizing various failure modes, including intentionally dunking the head board (un-potted) in water while the unit was on to simulate a failure of the isolation. As expected the unit freaked out - but what was not expected is that the leakage of current into the cell inputs caused the cells to spaz for a very long time. In one case the cell was effectively destroyed - it read low forever after that. In several others, however, I got really STRANGE high readings for a variable amount of time - and not a small amount of time either! One sensor read high for more than 24 hours, although it later returned to normal. (!) The original design for the K1 had two handsets - a full eCCR one and a second handset that was a monitor only (no solenoid control.) That second handset was scrapped after these tests - it became apprent that after an electronics failure you could not trust anything connected to the failed unit, and that included the O2 sensors! So unless you were prepared to run six sensors in the head you were functionally better off with a computer link (e.g. VR3) with a fourth sensor than redundant handsets, because the VR3 connection was electrically isolated from the other cells. I also had a sensor in an O2 meter that went EOL on me, and started experimenting with it, finding the voltage-limiting behavior quite easily. That was disturbing as well.... yet another way those infernal things can kill you. Anyway, all of this led to the "validation" requirement in the K1 code, and the decision to single-point calibrate on air. In reality single-point cal on O2 at the surface with a subsequent removal of the head to validate on air would work just as well. It did, however, make clear that two-point calibration was downright dangerous in the face of a cell fault - you could very easily be led to believe that the cells were working when in fact you had set a slope that was not actually linear at hyperbaric PO2 levels! Using the "implied zero" of single-point cal avoids this issue. I firmly believe that the only way to be reaosnably certain that the sensors are going to read correctly is to see them doing so in a proven environment with a HIGHER Po2 than you intend to trust them at when running on "automatic." Whether you do that "out of the box" in a dedicated cell checker, you do it voluntarily on the way down as part of every dive, or you are forced to do it by the electronics in order to run a hyperbaric setpoint isn't the important detail. That you do perform the check is, because if you don't and there's a problem here you are quite likely to take an O2 hit without warning and die as a consequence. __________________ "A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American." http://www.denninger.net http://www.diversunion.org/liability.htm - Fix the Diving Cert racket
(Online)

19th November 2006, 17:37	#22 (permalink)
rdmmdr designer of death Current Rebreather/s: Other CCR Other Rebreather/s: Other CCR Join Date: Mar 2006 Location: kerman,california Posts: 372	Re: Cell Linearity cool distructive testing is allways fun. now where did i put that can of factory installed smoke for the repairs. nothing works after that pesky smoke gets out. rick
(Offline)

19th November 2006, 17:41	#23 (permalink)
dave t Custom Title Allowed! Current Rebreather/s: Inspiration Classic Inspiration Vision Evolution Sport Kiss Classic Kiss Dolphin Ray Azimuth Home Build Other Rebreather/s: Dolphin Ray Azimuth Home Build Join Date: May 2005 Location: UK Posts: 654	Re: Cell Linearity Karl is it not possible to buffer the cells so they dont react to stray voltage and the like hitting them? I was one of the early crash test dummies for the HH electronics on one dive I had a primary flood at 80m, the primary freaked and the hud scared the crap out of me with its flashing and vibrating but the secondary po2 readout remained perfectly in line with the other two cells I had in the loop (5 in total) which were hooked to a second passive display. I assumed from conversations with KJ that the "buffering" would stop the crapped out primary affecting the reading. I may have misunderstood what Kev told me but I managed to complete all my deco on the loop (another hour or so) before the secondary readings started to be affected. I should add had I not had an independant back up po2 meter I would not have trusted the secondary on its own Dave Dave
(Offline)

19th November 2006, 18:10	#25 (permalink)
AD_ward9 So much more to learn Current Rebreather/s: Other CCR Other Rebreather/s: Other CCR Join Date: Jun 2005 Location: Scotland Posts: 1,864	Re: Cell Linearity Quote: (Originally Posted by am) It's really very simple. The number corresponds to the year and the letter corresponds to the month. So for example 5A would be January 2005 and 6D would be April 2006. Regards AnneMarie This is for Teledyne's cells. Each manufacturer is using their own codes. Good of you to highlight these codes, as a surprising number of divers have forgotten them and are diving on three cells with the same date code (a dangerous practice). For rebreather use, cells should be marked much more clearly with the date, e.g. "SEPT 06", using an ink that does not vanish if the label rotates (as it does on the Teledynes). Alex
(Offline)

19th November 2006, 18:28	#26 (permalink)
AD_ward9 So much more to learn Current Rebreather/s: Other CCR Other Rebreather/s: Other CCR Join Date: Jun 2005 Location: Scotland Posts: 1,864	Re: Cell Linearity Just to bring to the attention of those interesting in cell linearity: O2 cells are not linear with pressure. The cell sensitivity changes with ambient pressure, given constant PPO2, by up to 20%. Keep PPO2 constant, then pressurise the cell slowly, and watch the output. The design using the cell, should compensate for pressure in taking PPO2 measurements. Generally the sensitivity increases with pressure, so this is not a critical safety issue (at depth, the PPO2 will be lower than that displayed in most designs, but not low enough to cause hypoxia). Alex
(Offline)

19th November 2006, 21:47	#27 (permalink)
dive2dive2000 PRISM DIVER & LUVIN IT! Current Rebreather/s: Prism Topaz Other Rebreather/s: Join Date: Feb 2005 Location: Long Island, NY Posts: 788	Re: Cell Linearity Quote: (Originally Posted by jradomski) martin, even if the cells are in the right range for 100% oxgen this doesn't guarantee that it will be accurate for anything above this... I have seen cells that looked fine at a 1.0 po2, but when the controller was showing a 1.3, the real po2 was over a 1.5... These cells were't quite current limited at 1.3, I did manage to get them to just over 1.4 but I needed a po2 of almost 1.8 to do it... (was off loop with 100% o2 in loop and descended down to 9m, at 8m it maxed out) from general observation, the higher the cells normal output is in air, the earlier it the cell appears to start going non-linear.. And this makes sense since it takes a greater chemical reaction to generate the higher voltages(currents).. This assumes cells of similiar sizes and diffusion rates.. If you go to larger cells where you can increase the electrolyte/lead and increase the diffusion membrane, this behavior would lessen.. The PRISM (as well as other Rebreathers) requires a cell that operates in a mv range 16-22mv. The reason for this is so you will have a enough mv for high ppo2 readings. It saddens me that manufactures would place low mv cells in Rebreather I took for granted that this was across the board, after all this is not new information and high out put cells have also been around for the last 10 yrs or better. __________________ Safe Diving, Martin
(Offline)

19th November 2006, 22:27	#28 (permalink)
AD_ward9 So much more to learn Current Rebreather/s: Other CCR Other Rebreather/s: Other CCR Join Date: Jun 2005 Location: Scotland Posts: 1,864	Re: Cell Linearity Quote: (Originally Posted by dive2dive2000) The PRISM (as well as other Rebreathers) requires a cell that operates in a mv range 16-22mv. The reason for this is so you will have a enough mv for high ppo2 readings. It saddens me that manufactures would place low mv cells in Rebreather I took for granted that this was across the board, after all this is not new information and high out put cells have also been around for the last 10 yrs or better. Low output cells last longer and have a better response time. Low output cells have a lower output impedance, which can be useful. If the electronics is designed properly, it does not matter whether the cell output is 1mV or 100mV. "Designed properly" includes measures to eliminate/reduce noise. Alex
(Offline)

19th November 2006, 22:32	#29 (permalink)
Gill Envy Shearwater Copis Divers Current Rebreather/s: Other CCR Other Rebreather/s: Evolution Other CCR Join Date: Nov 2005 Location: seattle Posts: 1,314	Re: Cell Linearity Quote: (Originally Posted by Louby Lou) It is still here, although it is closed. The thread is HERE the link appears to be dead. __________________ Gill Envy ...Because I wasn't born with gills! ><(°>><(°>><(°>><(°>><(°>><(°>><(°>><(°>
(Offline)

19th November 2006, 23:12	#30 (permalink)
dive2dive2000 PRISM DIVER & LUVIN IT! Current Rebreather/s: Prism Topaz Other Rebreather/s: Join Date: Feb 2005 Location: Long Island, NY Posts: 788	Re: Cell Linearity Quote: (Originally Posted by AD_ward9) Low output cells last longer and have a better response time. Low output cells have a lower output impedance, which can be useful. If the electronics is designed properly, it does not matter whether the cell output is 1mV or 100mV. "Designed properly" includes measures to eliminate/reduce noise. Alex I am getting lost, you are now throwing a whole bunch of information into the mix It was my understanding that with low output cells there will come a point where you will not be able to read a 1.6 or higher? I know for my unit and my cells it reads true as long as I maintain cells that produce mv between 16-22 As far as other cells and other manufacturers and instructors go they should be versed in the specs that apply to them and should be teaching this in there courses. The original thread's main theme was basically is a pressure pot a necessity for testing cells. IMHO it is not. A volt meter your cells operating range and a swim to 9m is all you should need. Genesis and you provide very interesting information but it is way over my head __________________ Safe Diving, Martin
(Offline)

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode