ppo2 monitor for SCR with fo2 capability

[pegcjs]

Having bought a second hand Ray on ebay, been on the TDI course and done some diving this year on the unit I found myself very curious about how much oxygen was really in the loop. You can't buy drager oxygauge these days so I decided to build my own. This is a data dump from the project in case anyone else is interested in doing something similar.

Just knowing ppo2 which is the common solution and this would have been fairly easy. However for the SCR where the oxygen fraction is (relatively) constant one uses a standard nitrox computer (suunto vyper 2 in my case). You really want to know that the oxygen fraction you set for your gas mix on the surface, from which the computer is calculating all it's tissue compartments and deriving deco status is correct. Thus I decided to build a sensor unit with a pressure sensor to measure the gas pressure allowing a calculation of Fo2 as well as ppo2.

Basis for this was a pic - 16f688 programmed with the free version of microchips hit-tech compiler. I'm using an R22D sensor at present. The pressure sensor I'm using is an MPX5700A which goes up to 7bar and runs of 5v with 2% accuracy over the whole range. It's handy as its completely compatible with the input ranges of this microcontrollers ADC. for the ppo2 I've used an lm358 single supply op-amp setup with a gain of ~40 to match the ADC range of 0-5v. Attached are some pictures of the unit, a circuit diagram and the microcontroller code (its c code but named .doc to get by the uploaders restriction).SChematic shows the 3 chunks of circuit - the power, microcontroller and lcd display sections. In the unit the connector and housing for the cell and microcontroller board is a modified bottle trap from my local hardware emporium, modified by gluing a drager p-port plug (bored out as a tube) into one of the ports. Drager inhale side hose is connected to the other port using a cable tie.

To get fo2 I need to measure pressure so the system also has a depth gauge built in. As I am also still eeling paranoid about scrubber life I built in a simple clock that counts up runtime from when the unit is calibrated on the general principle that this is roughly the scrubber lifetime if I am careful about when I calibrate. I'll add in an NDIR CO2 monitor sometime later in the year to get more of an idea about the scrubber maybe.

The system implements a calibration with cal constants stored in the eeprom so that one doesn't need to re calibrate the sensor with every switch on. This allows one to do a calibration, switch the electronics off assemble the rebreather and then switch on the electronics when ready to dive.

I built the cal switch inside the sensor housing so that one can only push the cal button if the housing is disconnected from the unit making it less likely that one may accidentally cal on high % mix.

Battery life on one durcaell pp9 is so far 6 dives and counting.

If anyone's interested I am happy to provide further info should it be wanted.

[Packhorse]

Very cool project.

How hard was it adding the depth sensor and associated code?

[pegcjs]

Hi there. Thanks for the post. I have certainly enjoyed doing this so far.
For the pressure sensor you just give it a regulated 5v and an earth then it gives a dc voltage back which is 640mV per bar with a small offset of about 200mV.

Then I just had to put in code to do a cal for surface pressure (because the offset was uncertain at first) and then sort out computations for fo2 to stay within the 65536 max for int variables. That's why the code has several lines to archive one calc. Using floating point did work but it used up too much space in the pic hence the pain.

Keeping a bit of space means I can bolt on a buzzer later to give alarm for excessive ppo2 or fo2 below my deco setting.

Still need to do some work on the switch as it did come adrift before one dive but its just a matter of sorting out the magnet and its mount.

I might try to turn this into a dedicated dive computer using the ppo2 to track the deco computation but it would need a much bigger pic than the 4k version used here.

Sent from my HTC Desire using Tapatalk

[tecdiverdude]

Well done.. Im impressed .. thats a pretty good home job...

[Packhorse]

Cool thanks.

I have just added a temperature sensor to mine and that has a 500mv off set then 10mv per deg C. Its nice to know there is a pressure sensor that works the same way. Still, that means I need more analogue pins!!!

[pegcjs]

Hi there,
Hmm - one solution to the too few pins problem is to use an analogue multiplexer but you have to devote some pins to driving it. That's how I've done my tempstick in the past - used an ADG438 8:1 with 3 address pins and one analogue pin ion the pic to allow the little pic to read 8 thermistors in turn. It means adding a great big 16 pin dip to your circuit.

To be honest the whole tempstick wasn't a great help though as the termperature of my scrubber tended to be quite homogeneous after the first 10 minutes or so. My dream of a nice scrubber bargraph failed to survive contact with this discovery. If I had put the sensors down the edge of the scrubber it might have been better but I didn't try it at the time.

[gerstl_ossi]

Well done. Looks pretty nice.
Best Ossi

[blademaster]

looks nice, i used a KISS Ppo2 monitor... and custom maded fitting for the lung!

[pegcjs]

Thanks for the appreciation. I am wondering if you chaps have any experience with using mixes other than 50% in these units ? Specifically I am wondering what range of vo2 rates people find ? Any data apart from my own would be much appreciated.



Sent from my HTC Desire using Tapatalk

[Simon A]

I am wondering if you chaps have any experience with using mixes other than 50% in these units ?

You can get T53274 (32%) and T53276 (40%) orifices to enable you to rebuild the dosing unit for different mixes. Apparently the 32% part is end of life and it's only what people have in stock.
Online Shop | W+S Water Safety Europe GmbH | Seit 1999 | Medizinischer Sauerstoff, Peli Koffer und Lampen, Dräger Rebreather und Ersatzteile have it listed as in stock. The web site is in German, but google translate does a pretty good job. Also the guys speak/write decent English so it's easy to deal with them by email.

Tecme (Tauchen - Cave, Nitrox, Rebreather Ausbildung, Tec Ausrüstung, Reisen) can get hold of most Draeger bits (but don't have any 32% orifices left)
HTH
Simon A