Air Quality Sensor

[tantt2810]

Hi all,

I have my library for reading sensor. It is open source and its name is OpenSensor. I think that it's useful for every body.

Besides, I have my a monitoring system in real time using Arduino, MQ2, MQ135, MG811, GP2Y10 sensors and shield wifi ESP8266. It use OpenSensor library to read value from sensor and push it to database of WebServer by ESP826.

This is link: https://github.com/tantt2810

Hoping it helpful for all,

Good job,

[alexsh1]

@tantt2810 Thanks

#define RSRO_CLEAN_AIR_FACTOR        3.7                                    //The value of Rs/Ro in clean air. According chart in datasheet.
#define RL_VALUE                     20                                     //The value of the load resistance on the board, in kilo ohms.
#define GET_RO_SAMPLE_TIMES          10                                     //The times of calibrating 
#define GET_RS_SAMPLE_TIMES          10

Can you please elaborate on how to calibrate the sensor?
Also how do you connect it if it comes on the board like this?

http://i.ebayimg.com/00/s/NjAwWDYwMA==/z/vMIAAOxycD9TTOVW/$_1.JPG

[epierre]

@alexsh1 said:

Also how do you connect it if it comes on the board like this?
http://i.ebayimg.com/00/s/NjAwWDYwMA==/z/vMIAAOxycD9TTOVW/$_1.JPG

easy here, the AOUT is the one you will put on an analog one.

the DOUT (Digital OUT) is just an alarm linked to a level you could change on the board)

[alexsh1]

@epierre said:

the DOUT (Digital OUT) is just an alarm linked to a level you could change on the board)

So DOUT is not used. Just AOUT?

[epierre]

@alexsh1 said:

@epierre said:

the DOUT (Digital OUT) is just an alarm linked to a level you could change on the board)

So DOUT is not used. Just AOUT?

yes !

[alexsh1]

@epierre For the air quality purpose which sensor would you recommend TSG-2600 or TS2602? They are slightly different, but both detect aiir contaminants (based on ppm of H2 and EtOH).

So far, I got MH-Z14A and the Dust sensor working with OLED screen and Domoticz and I just want to add 1-2 sensors to finish the project.

[epierre]

@alexsh1 well any I would think for they are similar technology

[alexsh1]

@epierre What about MICS-4514 or MICS-5524?

I am thinking ordering this board:
https://sgx.cdistore.com/ProductDetail/MICSVZ86F-SGX-Sensortech-Limited-formerly-e2v/556306/#.V2Bw7jVUmec

[Tore André Rosander]

How do i get the Airquality sensor to show in domoticz? I use the AirQuality example from the library and MQ2.
I can see that it works in the serial, and it shows as a node under the mysensors hardware, but nothing in log or the device list when i expose it for gas.

[epierre]

@Tore-André-Rosander said:

ut nothing in log or the device list when i expose it for gas.

huh you're on which version of mysensors ? have you tried the console (in arduino programmer) to see if there are some readings or a good wiring ?

[Tore André Rosander]

@epierre Hi! The dev branch (v 2) and the raspberry pi gateway (nrf24 directly connected to RPi GPIO). This is what i get in the serial console when i test with gas from a lighter:

Starting sensor (RNNNA-, 2.0.0-beta)
Radio init successful.
find parent
send: 255-255-255-255 s=255,c=3,t=7,pt=0,l=0,sg=0,st=bc:
find parent
send: 255-255-255-255 s=255,c=3,t=7,pt=0,l=0,sg=0,st=bc:
read: 0-0-255 s=255,c=3,t=8,pt=1,l=1,sg=0:0
parent=0, d=1
req id
send: 255-255-0-0 s=255,c=3,t=3,pt=0,l=0,sg=0,st=ok:
req id
send: 255-255-0-0 s=255,c=3,t=3,pt=0,l=0,sg=0,st=ok:
read: 0-0-255 s=255,c=3,t=4,pt=0,l=2,sg=0:5

send: 5-5-0-0 s=255,c=3,t=15,pt=0,l=2,sg=0,st=ok:

send: 5-5-0-0 s=255,c=0,t=17,pt=0,l=10,sg=0,st=ok:2.0.0-beta
send: 5-5-0-0 s=255,c=3,t=6,pt=1,l=1,sg=0,st=fail:0
id=5
Init complete, id=5, parent=0, distance=1
0
LPG:0ppm    CO:0ppm    SMOKE:0ppm
0
LPG:8ppm    CO:47ppm    SMOKE:18ppm
send: 5-5-0-0 s=0,c=1,t=37,pt=2,l=2,sg=0,st=ok:95
0
LPG:0ppm    CO:0ppm    SMOKE:0ppm
send: 5-5-0-0 s=0,c=1,t=37,pt=2,l=2,sg=0,st=ok:0```

[epierre]

@Tore-André-Rosander said:

t=37

I don't know, would it be possible to have someone knowing the difference of the gateway with domoticz for 2.0 ? here it seems correct to me, some value are well seend from the sensor, so it is between the gateway and domoticz for me.

[tuimanso]

@epierre I am attempting to apply this code to work with the MQ-8. I am trying to get rid of all of the radio and gw stuff that is in the code. Do you have anything that just calibrates the sensor and reads out the value without of all these other libraries?

[tuimanso]

@epierre Also, what is the purpose of #define MQ135_MAXRSRO 2.428 //for CO2
#define MQ135_MINRSRO? I don't see them being referenced after they're defined?

[epierre]

@tuimanso said:

@epierre Also, what is the purpose of #define MQ135_MAXRSRO 2.428 //for CO2
#define MQ135_MINRSRO? I don't see them being referenced after they're defined?

Datasheet defines min and max values they should guarantee, this is an application of it.

if(validinterval<MQ135_MAXRSRO && validinterval>MQ135_MINRSRO) {

[tuimanso]

Your previous message

Hello,

I have a question about the MQ-8 sensor module . The potentiometer that is soldered onto the board, is it to control the sensitivity of the LED light or for the sensor itself? According to the datasheet for the sensor, the RL (load resistance) for the sensor should be at minimum 10k Ohm, yet I only see an SMD 102 resistor on the board, which is a 1K Ohm resistor. I have used a multimeter to connect the 5V pin and the AOUT pin, and I get precisely a measurement of about 1K Ohm... Does that mean that I should connect a load resistor to this?

This value seems very important since it is referenced in the code, the RL is part of the calculations, yet it is already pre-written 10K ohms...

Please Clarify
Thanks

[AtmosCheme]

Hello everyone,

I have a few of these sensors, but lately I have been concentrating on a MQ131 ozone sensor module from SainSmart. I am building an array utilizing MQ gas sensors and an Arduino that can be attached to a drone to measure vertical concnentration profiles of ground level ozone.

I am using an ozone generator and a 2B Technologies Ozone Monitor for the calibration of the sensor.

I have done the 24 hour burn in period on my sensors.
I have warmed them up to get steady values on my serial monitor.

The problem that I am having is that I am creating nearly 20,000 ppb O3 and the analog output is only around 25 to 30 units. It works out to around 922 ppb per unit.

Example before ozone is added the sensor reading is steady at 41. After raising the concentration of O3 to 18,727.5 ppb the reading drops to around 17-18.

Recently I have added a 22,000 ohm resistor in series with the analog output so that my readings start out at 971. But I am still only dropping around 20 units to 951 when ozone levels rise to nearly 20,000 ppb.

My question is how do I use a larger range of the units. There are 1023 possible values and I would like to see my reading drop from 971 to maybe the 200 area so that I am using more of the range. I would like each unit to be worth about 0.5 ppb. Ambient outdoor concentrations very rarely go higher than maybe 150 ppb so with the setup I have that would not even register on the serial output.

Thank you in advance for any help.

[epierre]

@AtmosCheme the datasheet gives the MQ131from 1 to 1000 ppb, so you reach the limit obviously

http://eph.ccs.miami.edu/precise/GasSensorSpecs/Ozone.pdf

[AtmosCheme]

@epierre

Thank you for the reply epierre.

I was going off of the SainSmart website that says the range for their module is 10 ppb to 200 ppm. That would cover 10 ppb to 200,000 ppb.

http://www.sainsmart.com/sensors/eco/sainsmart-mq131-gas-sensor-ozone-module-for-arduino-uno-mega2560-r3-raspberry-pi.html

The values keep dropping down as more and more ozone is added, I'm not thinking it is a problem with range as much as a problem with resolution. I have gone as high as around 50,000 ppb and continue to get response from the sensor.

Ive thought of using mapping functions in my code as well as resolution functions. But I just wasn't sure if there was a component I could add to the circuit to improve resolution.

In my latest sketch I have used a calibration code that analog reads the sensor for 6 minutes, and then remaps the min and max values observed during that time. I though this would get a broader range of values. The problem is that the longer the sensor is on it becomes stable so the min and max values become the same and there is nothing to map. So the longer it runs the smaller the range of values in the serial monitor.

[epierre]

@AtmosCheme said:

I was going off of the SainSmart website that says the range for their module is 10 ppb to 200 ppm. That would cover 10 ppb to 200,000 ppb.

check their datasheet (or ask for it !)

The values keep dropping down as more and more ozone is added, I'm not thinking it is a problem with range as much as a problem with resolution. I have gone as high as around 50,000 ppb and continue to get response from the sensor.

yes it should be possible but you don't know from the datasheet the value. On a power regression you make an estimation with error, but that may change outside of know datapoints.

Ive thought of using mapping functions in my code as well as resolution functions. But I just wasn't sure if there was a component I could add to the circuit to improve resolution.

some as the Particle Photon has a better ADC that goes up to 4096

[jnewton957]

Hi,

I just try to connect 7 sensors to an nano.
MQ-2, MQ-3, MQ-4,MQ-6, MQ-7,MQ-9, MQ-135

With sketch form robosensor https://forum.mysensors.org/topic/1595/mq2-sensor-w-airquality-sketch/2

I get data like MQ-7:1020 ,MQ135: 294, MQ-2:191

So it looks really good.
As i would like to have also the correct names an value names of the sensor and a correct calibration I think I should use a sketch like AirQuality-Multiple_Gas_Sensor1_4.ino
But I only see 3 of my 7 sensors in this sketch.

What would be a good sketch for my 7 sensors ??

Thanks
Jorg

[walk_porto]

Hello everybody,

I am working with an Arduino Uno and a MQ135 module arduino : http://www.sainsmart.com/sainsmart-mq135-sensor-air-quality-sensor-hazardous-gas-detection-module-arduino.html

I can get back data from this sensor, I got 100 ppm when I launched serial terminal after compiling and uploading this code below.

int sensorValue;

void setup()
{
  Serial.begin(9600);      // sets the serial port to 9600
}

void loop()
{
  sensorValue = analogRead(0);       // read analog input pin 0
  Serial.println(sensorValue, DEC);  // prints the value read
  delay(100);                        // wait 100ms for next reading
}

But, I would like to get back data from gases like NH3, NOx, alcohol, benzene, smoke and CO2 . I read a lot of topics and the datasheets of the sensor but I don't see how to do.

If someone could help me, it will be great.

Thanks.

[AtmosCheme]

@walk_porto

I think that the only way you are going to be able to distinguish between different gases is to have a known concentration of that gas and adjust load resistance. This sensor will detect all of these gase because of cross sensitivity to the gases mentioned in the filament. You can not use just one MQ135 sensor and get readings for all of the gases.

[Thanh-Viet]

Hi Guys,

Nice work. I'm based in Hong Kong, and what matters to us is PM2.5, NO2 and Ozone (and indoor CO2/VOC).

Have you guys checked the laser based particle sensors like the Plantower G1-G3-G5, Innovafitness SDS011/021 ?
http://aqicn.org/sensor

[alexsh1]

@Thanh-Viet Hello. Yes, I have Plantower G3 on order from Aliexpress and won't be able to test it before September. I think adopting it for MySensors is going to be easy

[paqor]

Hello, there are statements if the values increase the MQ135 when a room ventilated, without people and is furnished? I I have several rooms here an increase over time. Perhaps this is also by the increase in temperature, sunlight or gases from the furniture. I have not found any literature on this. Is this behavior with other users as well?

[paqor]

Has the type of installation (horizontal or vertical) have an impact on the values of MQ135? I have found no photos of built-in sensors already exist in the data sheet a statement.

[alexsh1]

@paqor yes, in fact my MQ-135 was so unreliable that I had to move on to a different sensor

[paqor]

Those interested in a 4 sensors Sketch FHEM (CO2, Temp, Hum, Motion) has ...

[A Former User]

@paqor
I'm interested, but i'm using mysensors 2.0. Can you update your sketch?

[paqor]

@gieljnssns said:

@paqor
I'm interested, but i'm using mysensors 2.0. Can you update your sketch?

Yes, sure but I need time. I'm not a good programmer and need a lot of testing. My Englich is not particularly good. As some patience please, because I have to house, garden, woman and taking care of animals.

[paqor]

@gieljnssns said in Air Quality Sensor:

@paqor
I'm interested, but i'm using mysensors 2.0. Can you update your sketch?

@paqor
https://gleisnetze.de/2016/08/07/sensoren4sketch/

• v0.8 korrigierte Lib von RSTOICA
• EEPROM set R0Cor
• amended mq135.h required!
• Temp, Hum --> CO2Cor
• Api 2.0
• Motion is not yet tested
• Auto base value set r0corr if Co2cor <390
• R0Cor set by TextField runs (incoming message)
• Redesign

[paqor]

@paqor
Probably one should not speak of a sensor CO2 value as well as responding to many other particles at the MQ135. The most accurate CO2 measurement we certainly be possible only with a MH-Z14. I now have to buy me a part and will then calibrate the MQ135 so. With the MQ135 can z. B. only say something about the air quality in relation to the outside air. It is rather a relative value.

Here the comparison of MH-14T (yellow) and MQ135 (red). One sees the calibration and as yet are both curves to each other.
[https://smarthome.gleisnetze.de/38-mhz-14-erste-erfahrungen-mit-dem-co2-sensor-fuer-mysensors-arduino-fhem]

Wahrscheinlich sollte man beim MQ135 nicht von einem CO2 Wert sprechen da der Sensor auch auf viele andere Partikel reagiert. Die genaueste CO2 Messung wir sicher nur mit einem MH-Z14 möglich sein. Ich habe mir jetzt so ein Teil bestellt und werde dann den MQ135 damit kalibrieren. Mit dem MQ135 lässt sich z. B. nur etwas zur Luftgüte im Verhältnis zur Außenluft sagen. Es ist also eher ein relativer Wert.

[ysinh]

@tuimanso
There seems a general misunderstanding about how those gas sensors really work: the "heater" must be switched on (5v) and "off"(1.4v) for the air sample going in and out of the sensor, when heating the air sample expands and goes out of the sensor, then in the cool down period the remaining air contracts and sucks in a new sample.
Those boards with mq... sensors only provide fixed 5v heating, so this can never work reliable
the 1.4v is needed to keep the sensor on a minimum temperature,example 40'c for the catalyst to react, the 5v will heat it up to example 80'c, to clean the catalyst and air flows out of the sensor
http://i49.tinypic.com/2rzdqgn.png something like this will work
Soon I will report my results together with a modified sketch for mysensors 2.0

[bezeefly]

@epierre I am working on a dust sensor project using Samyoung DSM501A. I am confused with the identical calculation formula used in the Samyoung DSM501A and Shinyei PPD42NS dust sketch.

concentration = 1.1pow(ratio,3)-3.8pow(ratio,2)+520*ratio+0.62;

My understanding is that the formula is based on the specific product's data sheet curve. Is the formula for DSM501A, PPD42NS, or both?

[epierre]

@bezeefly said:

Samyoung DSM501A and Shinyei PPD42NS

This is formula for Shinyei, you're right, I'searched back but don't know if a formula was made out : https://groups.google.com/forum/#!topic/airqualityegg/A1F9padoW10

strangely here it is used too : http://learn.linksprite.com/arduino/shields/use-dust-sensor-dsm501-with-arduino/

a formula by Chris Nafis here: https://github.com/OSBSS/Samyoung_Dust_Sensor/blob/master/Samyoung_Dust_Sensor.ino

[bezeefly]

@epierre I obtained the following equation by polynomial fit the low ratio vs particle (pcs/283ml) graph in the DSM501 datasheet.

concentration = (0.1 * pow(ratio, 2)) + (619 * ratio) + 50;

Please help to check whether the equation is correct.

[epierre]

@bezeefly said:

@epierre I obtained the following equation by polynomial fit the low ratio vs particle (pcs/283ml) graph in the DSM501 datasheet.

concentration = (0.1 * pow(ratio, 2)) + (619 * ratio) + 50;

Please help to check whether the equation is correct.

looks correct for the red curve

using http://www.xuru.org/rt/PR.asp with:
4 2500
8 5000
12 7500
16 10000
20 12500
22 15000
gives =7,613212955POWER(B3;2) + 467,3617082B3 + 631,9862425

a bit better

and on the previous one datasheet http://www.mouser.com/ds/2/744/Seeed_101020012-838657.pdf:
2 1000
3.9 2000
5.8 3000
7.2 4000
9 5000
10 6000
11.3 7000
12.6 8000
y = 19.09464953 x2 + 383.1041412 x + 174.9189116

[jumping]

Hello,
I bought the classic MQ sensors (MQ2, MQ4 and MQ135) and I'm building a multisensor node with arduino nano v3:
@epierre can I use sketch https://github.com/empierre/arduino/blob/master/AirQuality-Multiple_Gas_Sensor1_4.ino, is it suitable?
thanks

[epierre]

@jumping you will need to scale down the script obviously, in term of power it will be too short to have more than one.

[jumping]

@epierre said:

@jumping you will need to scale down the script obviously, in term of power it will be too short to have more than one.

thanks for your answer! please, can you explain how the following code in your sketch is related to datasheet?

• float CO_secCurve[2] = {726.7809737, -4.040111669}; //MQ135

I understand the sketch in the "Gas Detection - Example", but not the previous.
sorry for my poor english

[epierre]

@jumping see just above on power regressin curves, this is the result from a power regression based on the data sheet provided by one maker. Sometime I leave the data points in the sketch, sometime not (based on history

[ambuj]

@epierre Hi I am working on the MQ2 sensor, I want to know CO level. Did you do it before with this sensor?

http://www.amazon.in/Butane-Hydrogen-Sensor-Detector-Arduino/dp/B01N0XSQ9J?tag=googinhydr18418-21&tag=googinkenshoo-21&ascsubtag=9d5b3107-037a-4d7b-a8b9-722d0a4bcfea.

I am using this module and just start work with this sensor.

[Lukács Attila]

@epierre is it possible that the lastest github for Samyoung 501a is not correct? Flashed it and the results are way too high (157000 mg/m3)?

[surepic]

Ppm percentage calculations in the code seems wrong. Instead of providing rs/ro ratio to the function is transfered rs value not the ratio.

[epierre]

@ambuj please see the begining of the discussion regarding MQ like reliability

[epierre]

@lukács-attila check wiring first, seems there are several sub version around thay may need other values from datasheet

[alexsh1]

@surepic Which sensor? Which code? Would you please be more specific?

[Rosmel]

@alexsh1
Hello, I am working recently with tgs8100 sensors, and I am developing a project, I would like to know how it was in the results? How did you acquire them? Any arduino code? Thank you

[alowhum]

@rosmel said in Air Quality Sensor:

tgs8100

That's also quite a non-precise sensor if I understand correctly. Useful for general 'how's the air', but because it will respond to so many different 'polutants' you never know which of those polutant is the culprit.

You will find code in the forum for good CO, CO2 and fine dust sensors.

[rayzyzy]

Hi,
Working with tgs2602 , tgs 2620 and tgs2611 . I've read the replies from top to bottom but haven't understood how
exactly i take data from the graph in datasheet and use it with power regression .