Mailing List Archive Mailing List Archive
on 1-30 Paul Ostby wrote:
<cut>
- Even if your oxygen meter has an offset error, you can still get
accurate measurements with it. As long as the sensor itself is linear.
Here's how:
Start with two calibration gasses, for example air and 40%. Take pO2,air =
Pair = 20.8
Take pO2,40 = P40 = 40.0
Take the meter measurement for air. Call it Mair. Take the meter
measurement for 40% O2. Call it M40.
Take the meter measurement for the gas you wish to check. Call
it Munknown.
Calculate the desired result: Punknown = Pair + (P40 - Pair) *
(Munknown - Mair) / (M40 - Mair)
This technique will adjust for both offset and scale errors.
- The above technique may be more trouble than it is worth, even aside
from possible human error.
- If you get one of the meters David Drake was talking about, you
should
be careful about storing and using the sensor. Apparently the meter
is adjusted for the sensor's offset when the sensor is equilibrated
with air. If you store the sensor in inert gas for a long time, it
will have finished off-gassing O2, and its offset will have changed.
Take the sensor out of storage some time before using it. This may
not be a problem since it should probably come up to temperature
anyway.
[David: Do you know if this is correct? If so, how long does it take
to reach equilibrium? I am assuming that the offset error is mostly
due to O2 absorption and release by the surrounding material.]
Paul Ostby
<end>
I don't know about how long it would take the cell to reach
equilibrium in a non-O2 environment or when you bring it from there to
an air environment. The assumption is that you will have O2 in the
material simply by handling it in air. This may screw you up if you
have the sensor stored in a O2 tight place for a long time where it
uses up all the O2 that comes out of the surrounding material. Then
you'd have a void of O2 in the material, pulling O2 out of the gas
that is passing thru. If you keep the cell in the frig., you'll want
to warm up the cell before using it. This thermal equilibrium should
give it enough time to equalize the O2 concentrations. If you don't
allow for thermal equilibrium, you could see a lot of drifting as the
sensor warms and the reaction rate increases. Is any of this
significant enough to worry about? I'd expirement with air and
another known gas measurements when you bring it out of the
frig/non-O2 environment and then later. If you get bad readings on the
known after calibrating to air early, then good ones later...
Regarding the compensation equation, without a proof supplied, I had
to guess at the logic and do a lot of body part scratching. I assume
that you are calibrating with air and not touching the meter
calibration when you check the 2nd or third gas. Given this, you
might want to think of the second known gas as simply a reference gas.
W/o mathematical handwaving, the bottom line is yes, the equation
theoritically looks good to me. However, I would doubt that the three
measurements with their error factors and the division and
multiplication by the differences between them (which magnifies the
error) will give you a reliable result. Given that the cells
themselves will not be *perfectly* linear, you're REALLY stretching
the linear model here. Buy or make a better meter.
The name of this company is Pequod Diving Service. They do tech
charters, instruction, research, salvage, etc. The phone is in the
home of Capt. Bud George (904) 863-2205. I got a cell for about $150
at TEK price. If you call, please tell them you heard of them from
techdiver and that they should get on the net. I'd like to see more
manufacturers online. I'm not getting anything from this except a
longer list of things to do.
If you have interesting Q's, you might want to call them and send a
post out to the rest of us. I know if I'm interested, there's usually
a few more people interested too.
David Drake EDS/SATURN Infrastructure 8-320-4190 on GMnet
Spring Hill, TN USA Internet: saturn.ddrake05@gm*.co*
Navigate by Author:
[Previous]
[Next]
[Author Search Index]
Navigate by Subject:
[Previous]
[Next]
[Subject Search Index]
[Send Reply] [Send Message with New Topic]
[Search Selection] [Mailing List Home] [Home]