I am trying to use TGS 2602 sensor of air contaminants. I would like to test if my circuit is working as intended. For that purpose I need to generate fumes of one of the gases the can be detected by the sensor.

TGS 2602 sensitivity characteristics

I thought that I would go for ethanol. I tried holding vodka soaked cotton ball near the sensor - to no avail. Later on I tried spilling vodka around the circuit and placing more vodka soaked cotton balls around. I also covered the whole thing with a cardboard box (with gaps at the bottom between table and the box). Again there was no considerable effect. Vodka is essentially ~40% ethanol and water mix (by volume).

Can you say by that short description that I managed to generate between 1 and 100 ppm ethanol gas? If not, can you propose a robust and safe method of producing such a gas?

Please be aware how the sensor looks. It is enclosed in a metallic case with small holes on the top surface.

TGS 2602 sensor

  • $\begingroup$ Are you certain that the wiring and power supply 5 V DC for sensing and for heating are correct? You might want to look at the datasheet. $\endgroup$ Feb 3 '17 at 12:54
  • $\begingroup$ @Klaus Warzecha - that is what I am trying to establish. I am using a somewhat more complicated circuit, because I am using digital rather than analog reader. Anyway when I replace the sensor with resistors (~57 ohm for the heater and ~10k or ~100k ohm for the sensor itself) the circuit seems to work - nevertheless I am not 100% sure that it is functioning as intended, therefore I would like to create a testing environment in which I should have prior knowledge that sensor should be sensing something if it was working correctly. $\endgroup$
    – user40790
    Feb 3 '17 at 13:07
  • 1
    $\begingroup$ How to do it, and how to do it cheap are two very different problems. $\endgroup$
    – MaxW
    Feb 3 '17 at 16:36
  • $\begingroup$ If the 40% alcohol solution came to vapor-liquid equilibrium at room temperature , then the vapor concentration of ethanol should have been on the order of 2%. $\endgroup$
    – airhuff
    Feb 3 '17 at 18:41

Waving a vodka soaked cotton ball near the sensor is sufficient to generate the required fumes. It was the circuit that was problematic - I did a subsequent testing with analog reader (voltmeter to be specific).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy