dave t

Below is an email from Martin Parker giving his insight into cells, with his permission I have republished it here for interest

Dave




Apologies to everyone for this long email. Some of this reply answers
your query John but more is aimed at the general oxygen cell issue and
my comments aren't aimed at you or anyone in particular. Apologies for the lack of structure,
if you make it to the bottom, you get a free tub of Sofnolime, ..not :


John wrote:
The Oroborous has a cell health readout which gives the millivoltage as
a percentage of its original , easy to read and gives a good indicator
of cell life, possibly something similar from AP would be a thought.


*******
To be accurate of course, it would need to be making corrections for the
atmospheric pressure - as you'll obviously see a drop in output on a low
pressure day and a recovery on a high pressure day. C'os as some have
trouble understanding - the cell measures oxygen pressure not oxygen %.


I believe for this to be really useful you'd need to be comparing mV
output up at 1.4 to 1.5 bar or above - which we tend to avoid having in
the rebreather anyway. To compare outputs from one year to the next
measured in air is worth nothing other than seeing if the damn thing is
dead and your Inspiration/Evolution will pick that up anyway. To compare
outputs in pure
O2 is of no value to the CCR diver - other than does the output fall
within an expected range? - which the Inspiration and the Evolution do
automatically anyway - which is quite useful - particularly when the
diver is feeding it 90% O2 and telling the machine it is 100% ! You
could compare
outputs up at 1.3 but the warning will come in too late, there would be
a distinct risk that the current limited cells would be affecting the
breathing mix before you get any warning and in any case you're then
relying
on the cells which you are supposed to be analyzing telling you what the
oxygen pressure is - bit of a flaw that.


What is important is - will it get to above the setpoint? - and you can
do
that by just adding a bit of oxygen during the dive - just to make sure
the
display will go above.
Some want to check with oxygen flushes at 6m - if that makes them feel
good - let them do it. But the most important issue, is to check for
current
limiting and that's easy - you just add a tad of O2. They'll claim they
are
checking accuracy and linearity(which is affected by current limiting
which
you check for by adding a tad of O2)- well that's fine if they want to
do
that - it's all good stuff but I've done more accuracy checks on oxygen
cells at elevated pressures and ages of cells than most in controlled
conditions - looking for the "holy grail" - the technical advantage and
I
don't believe there's any mileage in it when diving with an Inspiration
or
Evolution but if it makes you feel good - please go ahead, c'os
understanding and testing your understanding and not relying on what
this
"thing" says is the best state of mind to be in - to a point. And let's
face
it if you are changing rebreather makes to ones where they have no need
to
prove their product to a 3rd party test house then it pays to check!


I can hear Tom screaming across the Atlantic at me re the liability of
my
statements here but we just do the same thing a different way - we make
it
accurate at 1.0 bar through calibration and we check for current
limiting by
ensuring it will display above setpoint and most of all by avoiding the
issue in the first place with some simple preventative measures.


Getting back to John's point:


A cell's output tends to remain up close to it's delivered voltage for
the
majority of it's life and then crash down very quickly at the end.


Now we'll design ourselves a pressure measuring system but using
language
that everyone should understand (this analogy was given to me some years
ago
by John Lamb - he obviously felt he needed to communicate in terms that
I
might understand ! ):
If you imagine a tank of water with a large bore pipe at the bottom -
the
flow out of the pipe will depend on the head of water. And, you'll see a
degradation of output as the tank empties. Now apply pressure to the top
of
the tank and you'll see a big change in the flow out of the big pipe. So
you
have built a fast response system but it isn't really capable of
measuring
pressure applied due to the fluctuations due to the head of water. But,
monitoring the output flow at a known applied pressure gives us a very
good
indication of the life of the pressure measuring system but of course it
won't
be long before the tank's empty.


Now change the design - run a microbore pipe 100 feet down below the
bottom
of the tank, the flow from the microbore pipe will be practically
constant -
right up until the last dregs of water; so measuring the flow out,
trying to
determine the end of the life of the pressure measuring system, is a
waste
of time, but of course the advantage is the measuring system will last a
lot
longer than the first design, c'os the water flows out slower. Apply
pressure to the top and the flow increases very slightly - so we have to
be
able to measure millilitres instead of litres, not a problem in this day
and
age. Now we have a fine tuned output relative to the pressure applied -
it's
clear though you need to alter the length and diameter of your microbore
pipe to get the appropriate response characteristics and appropriate
life of
your measuring system, but it works.


Now dealing with the life of the pressure measuring system : Given that
we
know the flow out for known applied pressures (c'os we calibrated it) we
can
forecast when the water will run out but what we really need is a way to
measure the level of water in the tank - and that's what's missing from
current oxygen cell technology ( we haven't got a way of measuring what
the
lead anode has left).


Bottom line is:
Throughout a cell's life the mV output drops very very slowly.
We've decided that it's not worth tracking that.


Whether Kevin will learn anything by tracking it? - who knows(?), and
good
on him for having a crack at it, but with what we know at this time I
wouldn't
place any importance on it.


*******
In the meantime, if you want to know what the cell's output is, take a
note
of the cell outputs when you calibrate. Make a note of the three cell
outputs at the end of "flushing", they will most likely have a different
reading for each cell.
Due to the constant calibration factor used during "flushing bag" the
value
reached equates to the cell's output in air, assuming you've flushed it
with
fairly pure O2.
i.e 0.98 shows us that cell's output in air is 9.8mV; 1.25 shows us that
cell's output in air is 12.5mV.
(a 10mV cell in air should have an output of about 47.85mV in O2.
Multiply
47.85 x the constant calibration factor(0.0209) and you'll see the
display
will be 1.0 (i.e it's a 10mV cell in air).


Obviously, once it calibrates all three cells will agree - showing the
PO2,
as calculated from Pamb x O2%. E.g 1010 mbar (1.01 bar) x 98% =
0.99(rounded
up).


Then if you want to get more accurate, reference the output back to a
standard pressure so you compare apples with apples.
********


I know of a rebreather maker that convinced it's customer that the
oxygen
cells were good for 4 years, ( which they are if you need to just
measure
oxygen around 0.21 all it's life). The customer soon ran into problems
of
current limiting and then went down the road of designing and building
an
oxygen cell test rig with automated pressurisation, recording and
automated
(SLOW) pressure drop to test the cells, presumably before they dived
them.
We take the simple approach and I believe they have done the same now-
they
are a consumable item - change them at 12 -18 months.


We change them all at the same time.
Our cells are tested for dynamic response, not just static values. The
only
batch problem we ever had was about 6 years ago when Teledyne was a bit
too
heavy handed with the board coating, sealing up the pressure balancing
holes, affecting the dynamic response of the cell (cell's reaction to
changing PO2 and changing ambient pressure). Since then of course we
check
on a production basis the cells response under changing ambient
pressures
too.
We do get cell failures, our QA system finds them before they go out of
the
door but on occasion you get a cell pass our tests only to play up when
used
further, to be picked up by the "cell warning" feature. I believe most
others don't have a Cell warning feature, perhaps they regard it as a
nuisance?


I think if and when they've got a few units out there and gone a few
years
down the line they'll probably wish they had one. I know one make
doesn't
supply oxygen cells - you have to buy them from a cell vendor and then
it
appears the liability for the cell working or not lays on the lap of the
cell vendor??






*****


Since the launch of the Vision electronics we've been able to see from
the
dive files that quite a few divers don't calibrate often enough. Here at
the
factory - and there's about 8 of us dive the units - we all calibrate
just
before every dive- always have. Some divers outside the factory
calibrate
once per day and I think that's fine but I wouldn't do it less than
that.


I actually think that many of the divers that seem prone to cell errors
are
the ones that don't calibrate too often.




I've seen it where the guy hadn't calibrated for about 8 dives and got
cell
warnings. To be fair to him, it's what he'd been told by his instructor
and
it's obviously been taken up with his instructor.


I would like to believe that very few divers run their cells to 2 years.
If
they do they are very likely to see a failure while they are diving -
fine
if they are aware enough to notice the solenoid is open longer than
normal
and aware enough to question why a cell isn't moving when they breathe.
Ok
these are things that divers should be looking for anyway and one cell
misbehaving isn't necessarily a problem but when you get two go down on
the
same dive then it starts getting serious.


I have covered this before: cells fail for a number of reasons but
assuming
you've got a good one and it's going to last into old age:
The life of the cell is dependent on the pressure of oxygen it is
exposed
to. The manufacturers say - if you expose it to air it'll last about
four
years, if you expose it to oxygen it'll last 3 months? ( What do they
mean? - what PO2 will it measure at those dates? Not 1.3 bar that's for
sure!)


Diving with it every day at 1.3 bar will reduce it's life by a tad. What
it's
life really depends on is what you store it in.


For us, we don't want divers taking the cells out of the lid- if it's
not
broke leave it alone, so the most important issue is ensuring the loop
is
returned to air. Don't go out of your way to flush it back to air just
make
sure you don't leave it at 0.7 to 1.0 bar at the end of the day/weekend.


The 12-18 months is purely based on our experience. We haven't had a
current
limited "problem" reported to us with cells in that age range. We have
with
older cells.


Again, apologies for the long email.


Best regards,




Martin Parker
Managing Director
Ambient Pressure Diving Ltd
MartinParker@xxxxxxxxxxxx
www.apdiving.com
Tel: +44 (0)1326 563834
Fax: +44 (0)1326 565945