Go Down

### Topic: Interfacing with a CO2 sensor (Read 3528 times)previous topic - next topic

#### ekiss

##### May 07, 2010, 02:27 pm
Hey,
I have been asked to make a simple carbon dioxide monitoring system. My only issue is that I am really only a programmer and have limited knowledge about electronics. In the data sheet, they have a diagram for interfacing with a external micro controller (http://cdn.shopify.com/s/files/1/0019/5952/files/Spec070426-K22-PWM_ed2-2000.pdf?1254409988). Although I have an idea of what to do, I do not want to damage the expensive sensor in anyway and would feel more comfortable if someone reaffirmed me.

I have already poked around online and seem to have stumbled on some working code already:

int  pin=14;
unsigned long duration;

void setup()
{
pinMode(pin, INPUT);
Serial.begin(115200);
}
void loop()
{
unsigned long ppm; // adding for clarity

duration = pulseIn(pin, HIGH); // get the raw reading
ppm = duration - 177000UL; // subtract the 'zero point'
ppm /= 500UL; // divide by the gain
ppm += 350UL; // add the zero point back in

Serial.print("  PPM Co2: ");
Serial.print(ppm,DEC);
Serial.print("  Raw Sensor: ");
Serial.println(duration,DEC);
}

I am really hoping someone can give me a detailed explanation of what I need to do but any help is appreciated.

Eric

#### RIDDICK

#1
##### May 07, 2010, 04:30 pmLast Edit: May 07, 2010, 04:48 pm by RIDDICK Reason: 1
1.
the co2 sensor and the arduino need to share a common ground voltage level...
how does ur power supply look like?

2.
the PWM output varies between 0V and V+ of the CO2 sensor...
the arduino needs below 30% of its supply voltage for a digital LOW...
the arduino needs above 70% of its supply voltage for a digital HIGH...
but the arduino doesnt like voltages below 0V or above its supply voltage...
maybe u need a voltage divider if the PWM output voltage can be higher than the arduino supply voltage...?

3.
i would use the following method to convert the PWM into the ppmCO2 value, because it doesnt depend on an accurate time (on my arduino i have an error of +0.3% IIRC... 0.3% of 1004ms (~=~3ms) corresponds to 6ppmCO2):
Code: [Select]
`uint16_t pwm2ppm(uint8_t pin) {  while (digitalRead(pin) == HIGH);  while (digitalRead(pin) == LOW);  uint32_t h,l;  for (h=0; digitalRead(pin) == HIGH; h++);  for (l=0; digitalRead(pin) == LOW; l++);  // h+l <=> 1004msec  // 1004*h/(h+l) <=> duty cycle D in msec  // (D-177)/0.5 + 350 = ppmCO2  return ((2*1004*h+1004)/(h+l)-2*177) + 350; // +1004 for proper rounding...(?)}`

maybe u should even disable interrupts...
http://arduino.cc/en/Reference/noInterrupts

-arne
-Arne

#### RIDDICK

#2
##### May 07, 2010, 05:35 pm
btw:
1.
do u know where pin 14 is?
i think it is analog pin 0...

2.
after a reset all analog pins r configured as inputs...
so that u can be quite sure, that the sensor isnt damaged by them...
but maybe u want to put a resistor (2kOhm or so) between the arduino pin and the sensor PWM output... just to be sure... :-)

-arne
-Arne

#### ekiss

#3
##### May 09, 2010, 03:13 am
Thanks for the reply. Here is a link to where I got the code from: http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1257112824

In the post, he seems to have gotten correct input using just two resistors and the code provided but I wanted to be sure of where exactly he put them.

#### RIDDICK

#4
##### May 09, 2010, 11:46 amLast Edit: May 09, 2010, 11:47 am by RIDDICK Reason: 1
according to the K22 datasheet (page 6 & 5) u dont need ur own pull-up resistor (4.7k to 5V), because the K22 already contains its own on-board...

the 10k is just for the case, that u re-conf pin #14 as output pin and give 5V on it... :-)
everything clearly above 125 Ohm should be safe here, because arduino pins can take 40mA...
according to the K22 datasheet there is already a 120 Ohm resistor on the K22 board...
the BC847 on the K22 can take even more than 40mA...

i like my approach (without pulseIn()) still better... :-)

-arne
-Arne

#### ekiss

#5
##### May 10, 2010, 12:39 am
hmm so you are basically saying i can just connect the sensor without any resistors (unless i want to be on the safe side)?

#### ekiss

#6
##### May 10, 2010, 12:44 am
also i think ill try the method described in the other post first because i know it has been tested and works. after i get it working ill be sure to try the method you described.

#### Andrew Robinson

#7
##### May 21, 2010, 06:04 pm
Hello,

Hopefully you have gotten this interface working!

I am a representative from CO2Meter and have noticed this forum post. I'd just like to add that we have a guide written specifically for Arduino boards that will guide you through interfacing with the sensor over I2C:

cdn.shopify.com/s/files/1/0019/5952/files/Senseair-Arduino.pdf?1264294173

Using a purely digital form of communications will generally provide a higher degree of accuacy.

Let me know if you have any further questions!
Andrew Robinson
andrewbrobinson.com

#### windhamrules

#8
##### May 21, 2010, 11:55 pm
WOW very nice, a rep helping you out! Fantastic! Rare, but awesome!
Chris
Its hard to put the smoke back in the wires

#### ekiss

#9
##### May 26, 2010, 02:12 pm
Hey Andrew,
Thanks for the post! I did get the CO2 sensor working via PWM. The guide you linked seems very helpful. I thought about using I2C in the beginning but I wasn't too comfortable with it.

One question I have is what do you think a average house CO2 level should be? I read in around 600-750 ppm on average. I want to make sure I did my math right and it is calibrated.

Thanks, Eric

#### gwen

#10
##### Jun 30, 2010, 06:24 pmLast Edit: Jun 30, 2010, 07:02 pm by gwen Reason: 1
Hi Andrew,
I just finally logged back on today and got your message.. yes 600-800 ppm is where most residences sit.
If you have your space filled with green plants like a greenhouse then the co2 can go lower as it is consumed by the respirating plants.

gwen
ps good luck with that project

Go Up

Please enter a valid email to subscribe