Nitrox/Trimix & CO analyzer

[stepfen]

Welcome Lidiano.

if you could put a more detailed list of the components you use,

A list of components is given in the very first post of this thread. The only thing I see missing there is a DC/DC converter to convert whatever your battery output is to 5 or 3.3V your circuits need.
I like to use these ones:
3pcs 3.7V 9V 5V 2A Adjustable Step Up 18650 Lithium Battery Charging Discharge Integrated Module Module Board from Electronic Components & Supplies on banggood.com
They work with 18650 lithium batteries and include a charger.
The Helium sensor according to the datasheet needs 3.0V DC. Hence on top of the previous converter I also use something like this:
3Pcs LM2596 DC-DC Adjustable Step Down Power Supply Module Module Board from Electronic Components & Supplies on banggood.com
to get these 3V. I know using 2 DC/DC converters is not optimal but it works. I am pretty sure the He sensor should work fine with 3.3V too hence one could just use 3.3V for everything (with a different board - arduinos need 5V) but I haven't tried this.

As for the display @Miyaru uses an ST7789, 240x240 SPI display. I use 0.96" screens
Geekcreit® 0.96 Inch OLED I2C IIC Communication Display 128*64 LCD Module Geekcreit for Arduino - products that work with official Arduino boards Module Board from Electronic Components & Supplies on banggood.com
They are much cheaper BUT they are very fragile ) I have broken 2, and 2 came broken during transport.

BTW you can find additional similar info in this thread:
DIY Nitrox Analyzer (Arduino based)

I don't I am expert, it is possible to use a bigger Touch Screen monitor so you could also insert the two buttons inside the display

You can do whatever you like - even attach it to a nuclear power plant and put everything in spacecraft and send it to the moon or above. The thing is how much you will to spend (touch screens are expensive let alone nuclear power plants and spacecrafts) and how much experience you have with these things - the more complicated things become the more difficult it is to keep track.

I would suggest to take it easy and start say with an O2/nitrox sensor, then add one step at a time , see how it goes and improve your "product" on the go. Believe me there are a lot of very important "details" that need to be addressed and usually other people have different needs and resources hence usually you can't just copy/paste what others did.

another thing I order the components and I hope if I have problems you help me,

As for help - you can ask here or in the Internet specific questions and usually you will get very good info and suggestions, but please don't expect anybody to do the thing on your behalf.

Enjoy your stay in scubaboard and all the best!

 

[Miyaru]

...
The Helium sensor according to the datasheet needs 3.0V DC.
...
I am pretty sure the He sensor should work fine with 3.3V too hence one could just use 3.3V for everything (with a different board - arduinos need 5V) but I haven't tried this.
...

I wouldn't try that. Others did and damaged the MD62 or MD61.

Hi everyone, I'm new to the forum, I found you because I was looking for a circuit to build just like this, I wanted to ask the author @Miyaru of this if you could put a more detailed list of the components you use, and then in another thing, I don't I am expert, it is possible to use a bigger Touch Screen monitor so you could also insert the two buttons inside the display, another thing I order the components and I hope if I have problems you help me,
thank you

I think I mentioned in the beginning that the code supplied was meant as a starting point. It's not copy-paste-analyzer. You gotta understand at least the basics of programming, and to understand the calculations for fractions and gas density, I assume you can apply all the theory that you learned during the trimix course.

Reason number 2 it's not copy-paste-analyzer:
I analyze gasses that I dive myself. I trust my life to this analyzer, which may sound dramatic, but at trimix depths there's no second chances. The analyzer has no brain, so you must use your own.

I have one of those touch screens, but the resolution is horrible, calibration is unstable and requires too much memory. I'm already using nearly every available bit, so in my opinion: bad idea.

 

[stepfen]

Hello again. As I am experimenting with this and I was looking at "heVcal". I assume this is the voltage you read at the sensor at rest (with air) and which (in theory) one should set to zero by adjusting the potentiometer.

The problem is that in my case I get something like this over time:

That's just with air. (Don't worry about the small spikes around 90 minutes - I had to move my prototype a bit).

And that's after I had to wait yesterday for few hours for it to "settle down" and set the potentiometer so that (yesterday) it was reading around 0mV (actually <0.1mV for half an hour) at rest. Today without changing anything it rests close to 1mV.

Is your sensors that unstable?

Or maybe I am just overthinking it ??? We are talking about few mV here while full scale (100% He) is in my case about 530mV - hence we are talking about less than ~0.5%.

BTW I have primed the sensor once I got it for 24 hours or so, and I am generally experimenting with it for few hours a time now and then (but not every day).

Any input welcome.

Cheers.

 

[Miyaru]

I'll setup the same, with a MD61 and a MD62 sensor. Once I have the results I will post them.
Did you run this in a closed environment? Otherwise the sensor will react to CO2 changes in the air (like breathing out).

 

[stepfen]

Did you run this in a closed environment? Otherwise the sensor will react to CO2 changes in the air (like breathing out).

I am using a plastic cylinder like this one as a "probe":

I've opened 4 small holes on the soft cover (top part in the image) for the He sensor pins (the whole sensor is inside the cylinder - only the pins extend outside). I have drilled a hole for the O2 sensor on the side of he cylinder. Another small hole (less than 1mm) at the bottom is used for mixture input (I place the whole thing directly in front of the DIN valve) and a bigger hole (3mm diam or so) on the side of the cylinder is used as exhaust to prevent pressure built up in the "probe".

During the measurements mentioned above the "probe" was left sitting on my desk (not directly in front of me). For now we don't use AC nor heating. Hence temperature and humidity in my house change a bit over time but nothing drastic.

 

[MikeinCA]

This is a super cool project! I don't have much to add, but am following along .

What's the ballpark figure on cost for materials? I imagine that it is substantially less than a cheap O2 analyzer plus a cheap CO analyzer.

^What he said. This looks awesome! Nice work

 

[stepfen]

Hello again. As I am experimenting with this and I was looking at "heVcal". I assume this is the voltage you read at the sensor at rest (with air) and which (in theory) one should set to zero by adjusting the potentiometer.

The problem is that in my case I get something like this over time:

View attachment 622109

That's just with air. (Don't worry about the small spikes around 90 minutes - I had to move my prototype a bit).

And that's after I had to wait yesterday for few hours for it to "settle down" and set the potentiometer so that (yesterday) it was reading around 0mV (actually <0.1mV for half an hour) at rest. Today without changing anything it rests close to 1mV.

Is your sensors that unstable?

Or maybe I am just overthinking it ??? We are talking about few mV here while full scale (100% He) is in my case about 530mV - hence we are talking about less than ~0.5%.

BTW I have primed the sensor once I got it for 24 hours or so, and I am generally experimenting with it for few hours a time now and then (but not every day).

Any input welcome.

Cheers.

As an update, I fount were the problem with the variable DC offset at rest was in my case.

I don't want to have the sensor soldered (as yet). Main reason for this is that I am still prototyping and things change quite often. The sensor has 4 pins and hence once soldered it is a huge PITA to unsolder (if/when needed). Hence initially I chose to use a 7 pin female pin header like this:

so that the sensor can be easily plugged in and out. Problem was that this "plug" didn't offer good conductivity hence it was sensitive to movement causing this DC offset.
Now I have replaced the header with a 16pin dip chip socket like this one:

and things are much more stable.

Now the sensor has similar response every time I use it. It starts with ~2mV at t=0, goes up to ~2.7mV about 5 or so minutes later and then it slowly drops down to 0mV after about 40 minutes or so.

I haven't got any helium left (and thanks to the lockdown I can't get any more for the time being) to experiment with how this curve affects the accuracy i.e. how a (real trimix) reading done say at 5 minutes (peak of DC offset) compares to a reading of the same gas 60 minutes later i.e. after the sensor has been stabilized (DC offset zero).

Anybody tried this?

Thanks a lot

 

[Wouter D'Haeseleer]

Just wanted to let you know I also created my own nitrox analyzer.
Have a look here: netdata-be/nitrox_analyser

It's using a custom PCB instead of a breadboard

 

[Miyaru]

Just wanted to let you know I also created my own nitrox analyzer.
Have a look here: netdata-be/nitrox_analyser

It's using a custom PCB instead of a breadboard

Linear drift of the O2 cell.........?

 

[Wouter D'Haeseleer]

Linear drift of the O2 cell.........?

You mean that I should calibrate on 2 mixes ( air and 100% o2 for example ), right?