Gas Sensor module problem

I use the sensor Gas Sensor module with MQ6 sensor.
but the output voltage keep decreasing , what is the stable voltage?

public partial class Program
    {
        // This method is run when the mainboard is powered up or reset.   
        GT.Timer Readingtimer;
        GT.Timer Heattimer;
        void ProgramStarted()
        {

            Debug.Print("Program Started");
            Readingtimer = new GT.Timer(10000);
            Heattimer = new GT.Timer(20000);
            Readingtimer.Tick += new GT.Timer.TickEventHandler(Readingtimer_Tick);
            Heattimer.Tick += new GT.Timer.TickEventHandler(Heattimer_Tick);
            Heattimer.Start();

        }

        void Heattimer_Tick(GT.Timer timer)
        {
            Debug.Print("Heating...");

            gasSense.SetHeatingElement(true);

           

            Readingtimer.Start();


        }

        void Readingtimer_Tick(GT.Timer timer)
        {
            Readingtimer.Stop();
            Debug.Print("Start Reading...");

            double gs1 = gasSense.ReadVoltage();

            Debug.Print("Gas Voltage:"+gs1.ToString() );


        }
    }

output voltage

Program Started
The thread ā€˜ā€™ (0x3) has exited with code 0 (0x0).
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.6722344322344322
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.5690842490842489
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.461904761904762
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.3950183150183149
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.3652014652014652
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.3466666666666667
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.3063736263736261
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.2749450549450549
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.2604395604395604
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.2539926739926739
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.2273992673992673
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.2169230769230768
Heatingā€¦
Start Readingā€¦
Gas Voltage:2.2104761904761903

Is it possible that the voltage decreasing means temperature increasing? Did you check the datasheet?

YES, the datasheet is poor, donā€™t mention much.
https://www.sparkfun.com/datasheets/Sensors/Biometric/MQ-6.pdf

It seems stable?

Start Readingā€¦
Gas Voltage:1.9888644688644685
Heatingā€¦
Start Readingā€¦
Gas Voltage:1.9767765567765565
Heatingā€¦
Start Readingā€¦
Gas Voltage:1.9824175824175823
Heatingā€¦
Start Readingā€¦
Gas Voltage:1.9888644688644685
Heatingā€¦
Start Readingā€¦
Gas Voltage:1.9993406593406591
Heatingā€¦
Start Readingā€¦
Gas Voltage:1.991282051282051
Heatingā€¦
Start Readingā€¦
Gas Voltage:1.9888644688644685
Heatingā€¦
Start Readingā€¦
Gas Voltage:1.9824175824175823
Heatingā€¦
Start Readingā€¦
Gas Voltage:1.9977289377289376

Yes, looks stable enough to me.

1 Like

and you have given it time to ā€œburn inā€, right? From memory, you need to turn the heater on for something like 15 minutes before trying to use it for the first time.

1 Like

Iā€™ve heard everything from 12 to 48 hours for burn in (ie to clean off any manufacturing residues from the sensor). Typically when I burned in my sensors I turn on the heater and left it on for a couple of days.

Note temperature and humidity affect the readings. Iā€™ll get my gas sensor demo on ThingSpeak updated as there is a network fix that I have to add to close connections similar to the keep alive issue I reported earlier.

I Just see a doc said that for heating 48 hours.
Thank you.
But the hardest thing is the relation between the voltage and the concentration.
How do you know the danger value?

If I fall down dead, I guess that would be a good level to use as a warning, but really these sensors are really only for relative type readings (ie its trending up or down) and even then with a pinch of suspicion given they are a cheap sensor (sensors I use when my life is on the line cost a whack more $$$ and are calibrated frequently which costs even more $$).

Which sensors are you interested in as I might have some spec sheets for them?

Yes, calibration is very difficult for this kind of sensor.
I have developed a anesthetic gas sensor,so I know how to calibrate.

But we need a standard Gas, and good Ventilation equipment.
Now I leave the Lab, canā€™t do calibration anymore.

I was testing the MQ6, I am thinking to detect the Gas Leaking when a Big earthquake happened, the Gas Pipe will damaged and the Gas will leaking.

I also bought a NDIR CO2 Sensor, can output ppm directly, convenient and cheap(near 67 USD). Still trying it.

Thanks.

Iā€™ve always intended to use mine as a ā€œburner is outā€ warning, luckily not something thatā€™s in an enclosed space or that will cause me an issue if it doesnā€™t detect it fast enough.

I honestly think that to be considered ā€œaccurateā€ (or even ā€œnot inaccurateā€) you would need to calibrate it with a known concentration of gas, and I donā€™t know that this would be easy (or cheap) at all to do. I suspect, like @ Duke, is that theyā€™re cheap and that they couldnā€™t be calibrated accurately enough for instance to be useful as an alcohol breathalyser telling you if youā€™re likely to have exceeded the limit to drive.

I had some conversations with a couple of gas sensor manufactures about building similar modules to the current gas sensor (ie where you could plug the gas sensor of choice into it) for their higher end gas sensors where you could get a parts per billion type reading, but these sensors were a minimum of $100 each and went up quickly from there depending on what gas your trying to detect, so I doubt there is much interest in sensors that are this accurate at that price.

One thing that I did was bought a handful of each sensor so I could try running multiples of the same sensor at once so I could compare readings to see how consistent the sensors are. Iā€™ll set that experiment up, which sensor are you most interested in and if I have it, Iā€™ll start with that one.

Iā€™m now running 3 MQ-6 Gas Sensor so we can check for consistency between sensors given they are all within 2 inches of each other, I would expect very similar readings. The first couple of days however I would be cautious as Iā€™m not sure as to the burn in status of these sensors so Iā€™ll assume they havenā€™t been burnt in yet and so after a couple of days running I would expect very similar reading.

Dear Duke:
Thank you very much.