voltage

[abattya]

Hi, Your code is working for me, it shows around 8.79 PPM, I dont know why, after a time it starts increasing :
21:07:15.528 -> PPM = 8.76
21:07:20.551 -> PPM = 8.88
21:07:24.535 -> PPM = 8.97
21:07:26.546 -> PPM = 9.02
21:07:30.559 -> PPM = 9.11
21:07:37.558 -> PPM = 9.22
21:07:40.545 -> PPM = 9.25
21:07:41.525 -> PPM = 9.27
ETC.
I'm not really in the electronics, but I read that this sensor neeeds 1.4V to measure and 5V to heat, and I've been watching really complicated projects with voltage regulators and other devices to manage the voltage, because if you dont use it right, the sensor will be damaged etc.. Im wondering how this very simple connection with just 3 wires(not even the digital output of the sensor) would work fine. Thanks for your answer! (and sorry for my mistakes:D)

[fjebaker]

Hello! I am not entirely familiar with voltage regulation for this sensor -- the one I had when programming this library had a little board that it was attached to, that seemed to take care of regulation for me.

The reason for it increasing is more likely due to the calibration. In the technical document is described how the sensor should be left on to calibrate for 48 hours, to permit sufficient heat-up and adjustment to the ambient environment, though in practice I found 5 minutes to be more than sufficient. You can either get your script to wait some time during setup, else do it once and hardcode the R0 value as described in the README of this repository, with the assumption that the operating conditions of your sensor wont change much over the lifetime of its use (a hand-wavy assumption at best).

Note that calibration, whether during setup or when hardcoded, needs to be in clean air (~10ppm).

[abattya]

Thank you! Another question, the data sheet says, that it needed to heat up the sensor first, for 60 seconds, and just after this can do the measure for 90 seconds on low voltage, as you can see in the datasheet(pic). If you do the measure every second its not showing uncorrect results. Im doing this project for my university thesis, thats why im so curious :D

[fjebaker]

I'm afraid this is a little beyond my understanding of this component. The model I had, with the circuit board, seemed to operate fine and consistently over long periods even when measuring rapidly. I don't know if the circuit kept the device constantly at high or low heating voltage.

It would seem the device should be kept on a heating cycle:

My naive understanding of electronics and trends suggest keeping the device at low voltage heating saturates the signal to its "true" value, and unfortunately the datasheet does not expliticly clarify the need for the high voltage.

If you do come up with an answer to this, please drop it in this thread, both so I can learn, and incase others have the same question.

Thanks!

[stonemull]

The standard breakout pcbs for these modules are effectively garbage and need to be modified for proper use. They simply provide almost 5v to the heater continuously, no low voltage at all.
The sensor is "cleaned" by the high voltage so sensitivity to gas is comparatively low, during the lower voltage stage the sensor absorbs the gas and the highest changes in resistance occur. Look closely at the plot and you can see the resistance is higher (Blue line lower) after the heating stage in clean air but decreases rapidly when the heater turns off in the presence of carbon monoxide.
For a simple mod i cut the track between the 5r1 resistor and the heater and soldered a small n channel mosfet across the cut, cut the track to the digital out line and soldered the gate to that pin. PWM an output to drive the heater at full power for 30 seconds then at 1.4v for 30 seconds.
You could just drive vcc with a p channel and then ensure you measure the resistance when voltage is on, if using a 3.3v micro though and a 5v heater you need an extra stage to drive the gate.