DIY "Deluxe" Oxygen Analyzer: cool idea or stupid overkill?

[j_c_a]

I want to get some opinions on a crazy idea I’ve come up w/.

I'm planning on taking a Nitrox course soon and after doing some research it seems like it would be a very good idea to have an O2 analyzer of my own.

After researching many of the commercial offerrings I’ve come to the conclusion that they're pretty simple devices, which led to the idea of DIY'ing one. So, being the obsessive/compulsive that I am, I researched how to make one. There seems to be pretty good information out there for that, and the designs I've seen are very simple.

At this point I'm somewhat disappointed because the designs I've seen lack any real challenge. I've also noticed that at least one brand supplies a chart used to calibrate the %O2 of air using temperature and humidity compensation.

Now's where it gets juicy 'cause it's starting to get interesting. I happen to have a couple microchips laying around doing nothing, so what's to stop someone from measuring temperature and relative humidity to come up w/ a more accurate %O2 for air and then use that for calibration. A digital pot (also lying around) could be used to automatically calibrate the O2 sensor at the push of a button. Now my intellectual curiousity has something to chew on.

After researching components for measuring humidity (temperature is easy) it seems that this is very doable.

OK, so now, I know that an auto-calibrating O2 analyzer is within my grasp but I'm left w/ a very under-utilized microchip, w/ lots of excess memory and spare pins.

So ... what is the value of not only checking your gas mix, but also being able to log the results by gas source? That way you could keep track of the accuracy of the mix you get from various vendors. PC download is a pretty simple proposition and I'm a software developer by trade so a simple program to accept the download is another easy addition to the project.

While I'm at it, I've got a nice 4x20 lcd w/ backlight lying around so what additional info would be nice after analyzing your nitrox mix? Maybe display MOD? Any other ideas?

OK, I admit at this point, it seems like overkill, but I enjoy DIY’ing stuff and find myself looking for decent projtects and this sounds like fun. Considering the parts that I already have on hand, I can probably build something like this for $150 tops and potentially less depending on how complicated I want to get w/ the humidity sensing (ie basic componets vs more modular ones). Does this sound like a decent idea or just plain stupid

?



Just looking for a sanity check before I start ording parts.



Thanks,

JC

 

[MASS-Diver]

Scroll down and check out "El Cheapo Plans" - http://www.oxycheq.com/Oxygen_Analyzer_Main_Page.html

 

[j_c_a]

Thanks MASS-Diver. I've actually seen that one as part of my research. Those plans along w/ some others I've seen are what got me to wondering in the first place. They seem to be simple devices: basically you're measuring voltage across the o2 probe and scaling that using variable resistance to get a meaningfull number. It even looks like the relationship of o2 to voltage is linear. One could pretty easily get the same results using a battery, o2 probe and a DVM w/ a few calculations.

I'm not in any way implying that these simple circuits are not satisfactory. Obviously multitudes of divers use circuits just like these w/o any problem. I'm just wondering if there's anything more interesting that could be done w/ an o2 analyzer. Considering I've already got all the parts except for the o2 sensor (which I'd have to buy anyway) and a humidity sensor.

I guess I'm just looking for an interesting DIY project that I might actually result in something useful.

 

[FredT]

O2 sensors measure absolute oxygen partial pressure modified by temp and humidity. The variation in barometric pressure between a high and low pressure day will cause a significant change in reading. That is one reason the sensors need to be calibrated before each set of readings, and with the same flow pattern you'd use to get your nitrox reading. i.e. Use a bottle of air to calibrate the meter with all the flow lines and passage lengths you'd use to read a bottle of Nitrox or mix.

FT

 

[j_c_a]

Thanks Fred,

My idea w/ the humidity/temperature compensation comes from a chart I've seen that comes from ANALOX. According to the chart, the % atmospheric o2 varies depending on temp/humidity from 20.9 to 18.5. They seem to be suggesting that you can calibrate w/o using a pressurized air source: http://www.analox.net/site/content_o2e2.phpv

W/ what I've read about the physics of %02 measurements, what you've said makes perfect sense; specifically using identical flow rates (which you obviously can't get using the calibration method suggested by ANALOX).

I can see the convenience of not having to calibrate using pressurized air (as shown in the ANALOX link above) but I'm wondering if it's truly accurate, or just a "good enough" thing.

I may be looking at a solution in search of a problem here, but it seems like a decent academic exercise at worst.

 

[Charlie99]

JCA, the sensors are reading ppO2, not percentage of O2. As Fred mentioned, it's best to calibrate with dry air, using the same lines and flow rates that you will be using with the nitrox tank. Any slight pressure differences in the lines directly affect the reading.

I only use ambient air as a coarse check to convince myself that the reference tank is actually air.

 

[j_c_a]

Thanks Charlie for the clarification. It does however, raise another question in my mind. Maybe I'm nuts, but it's my understanding that electo-o2 sensors work by measuring the number of oxygen molecules that react w/ the sensor. Obviously this is going to be dependent on density which seems to imply that pressure, not flow, would be an influencing factor.

Flow restrictors cause in increase in pressure on the upstream portion, before the constriction. Assuming there are no significant restrictions to flow after the constrictor, the pressure at that point would be ambient and not effect the density of the gas. Thus, it seems like you could accurately calibrate using ambient air pressure.

I may be completely wrong here and just showing my ignorance. If I am, feel free to tell me so (just be nice

), but it seems like as long as you kept the flow rate low when analyzing your mix (so as not the create a pressurized sensing chamber) then there would be no legitimate reason not to use ambient air for calibrations. In either case it seems that compensation for temperature/humidity would still be applicable.

Again this is all theory to me so far and I may be completely off-base.

p.s. Further research suggests that you need to have at least some flow over the sensor, because the measuring process depletes o2.

 

[Charlie99]

Yep. It's just absolute partial pressure of O2, not flow.

The amount of O2 used by the sensor is negligible, and that isn't why flow rate needs to be controlled. In real life, changing the flow rate changes the pressure a little bit --- the changes are the same sort of 1% or so variation you get with humidity/temp in free air.

Charlie

p.s. Most humidity sensors will work for a while on DC, but have much longer life with AC drive.

 

[cfelliot]

DIY Nitrox:

http://www.circuitcellar.com/library/print/0105/Smith174/index.htm

 

[j_c_a]

Thanks alot cfelliot. That like is very much along the lines I was thinking. Good stuff.