How to calibrate gas sensor?

Thread Starter

presish

Joined Nov 18, 2014
7
Does anyone know how to convert ADC output values to ppm readings for a gas sensor without exposing the sensor to a known concentration?

such as, if ADC says 54 through a common conversion method we know it is 20ppm... is there such a method that can directly convert the ADC output to ppm?
 

mcgyvr

Joined Oct 15, 2009
5,394
Does anyone know how to convert ADC output values to ppm readings for a gas sensor without exposing the sensor to a known concentration?

such as, if ADC says 54 through a common conversion method we know it is 20ppm... is there such a method that can directly convert the ADC output to ppm?
Nope.. You always need to calibrate from a known concentration.
And if you are making more than one device they each need to be calibrated independently as variations will be present.
 

MrChips

Joined Oct 2, 2009
30,706
@persish

1) please do not hijack someone else's thread
2) please do not post your new question to an old thread
3) please do not make multiple posts

Now you have your own thread.
 

jpanhalt

Joined Jan 18, 2008
11,087
What other details do you want in the graph? It is a ratiometric sensor in which the resistance of the sensor varies with the concentration of the gas being measured. Its output cannot be fed directly to an ADC. From the datasheet:
upload_2014-11-19_5-11-21.png

Please post the circuit you intend to use. Your design of that circuit will determine the voltage range you will see and the "ADC counts." Be sure to read the last section (Sensitivity Adjustment) of the datasheet.

John
 

Thread Starter

presish

Joined Nov 18, 2014
7
What other details do you want in the graph? It is a ratiometric sensor in which the resistance of the sensor varies with the concentration of the gas being measured. Its output cannot be fed directly to an ADC. From the datasheet:
View attachment 75922

Please post the circuit you intend to use. Your design of that circuit will determine the voltage range you will see and the "ADC counts." Be sure to read the last section (Sensitivity Adjustment) of the datasheet.

John
The circuit I am using is the same circuit listed in the datasheet. I am supplying 5VDC.... I read the entire datasheet multiple times but I do not have a fixed concentration of gas to calibrate.. thus my question on any alternative methods to calibrate these sensors.
 

jpanhalt

Joined Jan 18, 2008
11,087
The circuit I am using is the same circuit listed in the datasheet. I am supplying 5VDC.... I read the entire datasheet multiple times but I do not have a fixed concentration of gas to calibrate.. thus my question on any alternative methods to calibrate these sensors.
That is not the same question you asked in post#8. I will assume you know how to read the resistance of the sensor using the ADC and your test circuit.

Is it that you don't have hydrogen to make standard concentrations, or you do have hydrogen and don't know how or don't have the equipment to make the mixtures? What level of accuracy do you need?

John
 

Thread Starter

presish

Joined Nov 18, 2014
7
That is not the same question you asked in post#8. I will assume you know how to read the resistance of the sensor using the ADC and your test circuit.

Is it that you don't have hydrogen to make standard concentrations, or you do have hydrogen and don't know how or don't have the equipment to make the mixtures? What level of accuracy do you need?

John
I do not have the means to make known concentrations that is why I wanted to find a general conversion method for my digital output
 

jpanhalt

Joined Jan 18, 2008
11,087
If you can't calibrate it with known concentrations of hydrogen, then all you can do is get in the ball park, as shown in the datasheet.Whether that is 5% off or 30% or more off, I don't know.

John
 

Thread Starter

presish

Joined Nov 18, 2014
7
I guess then the remaining question on calibration via data and ballpark measurements would be... how can I convert my data either using Linear Regression or some other method to best fit the graph in the datasheet? I am using a Sharp Dust sensor and I found an equation that places my ADC readings directly on the datasheet graph.. is there a method to do this with the MQ sensor?
 

jpanhalt

Joined Jan 18, 2008
11,087
That is a log-log graph and looks pretty linear. (We used to joke that any data looks linear on a log-log graph.) I haven't looked at the dust sensor graph.

John
 

KL7AJ

Joined Nov 4, 2008
2,229
Nope.. You always need to calibrate from a known concentration.
And if you are making more than one device they each need to be calibrated independently as variations will be present.
This is correct. We had to do this calibration with mass flow controllers and reference gases. Kind of a fun process, but you can't be sloppy!
 

MrChips

Joined Oct 2, 2009
30,706
I have designed many industrial gas sensors. You cannot calibrate your instrument properly if you do not have a calibrated standard gas or an instrument that is already calibrated. Yes, we also used volumetric gas mixtures to characterize the sensors from 0% to 100%.

Note also that your sensor is also affected by environmental conditions such as temperature, pressure, humidity and presence of all the other gases.
On top of that, most sensors tend to be non-linear.
 
Top