MQ-4 and MQ-7 as Gas Sensors

Hi

Does anyone have experience with MQ-4 and/or MQ-7 gas sensores? Are their readings accurate?

I'm asking because I'm using them, and at the start, they are a lot unstable. I know that they have to heat at first.

If anyone has some tips for them (like how to make the values more stable or how to make their intern resistor heat faster) or other gas sensors, I would be so grateful.

I'm using this ones:

Thanks in advance

Pedro

Are their readings accurate?

The readings don't mean anything until you have calibrated the sensor, using a known mixture of clean dry air and the gas of interest.

Each of the MQ-series sensors respond to several different gases or vapors. If there is a change in the sensor output, without additional information it is impossible to know which of those gases is the cause.

jremington:
The readings don't mean anything until you have calibrated the sensor, using a known mixture of clean dry air and the gas of interest.

Each of the MQ-series sensors respond to several different gases or vapors. If there is a change in the sensor output, without additional information it is impossible to know which of those gases is the cause.

Hello, Jremington

Thanks for your reply

I've solved the problem. It was an electronic failure, the breadboard was instable, with drops on the bottom line, which causes the sensors to be unstable as well.

But in any case, I would love to ear some experiences with those sensors, do you think they are accurate in a home environment? Are they good for making a device for selling (again, in a home environment)?

Hi

I'm developing a device that measures the gases in the air, temperature and humidity. For that, I have made a circuit with ATmega328P-PU and ESP8266 (other the other sensors, obviously) so, when the sensors detect gas, a message (via module SIM800L) is sent for the user and, at the same time, he can see the results in real-time, via a mobile APP and a server (developed on NodeJS).

But, as I'm thinking, I will face a huge problem, and that huge problem is accessing the DATA on the ESP8266-01 from the exterior, what IP should the server look for, if I want to access my ESP8266-01 data from the exterior?

In my house, I'm using the private IP to get to the ESP and the public IP (on the port 3333) access the server, on the mobile App.

Thanks in advance.

If something looks confusing, tell me, please.

Pedro

batista742k2:
Hello, Jremington

Thanks for your reply

I've solved the problem. It was an electronic failure, the breadboard was instable, with drops on the bottom line, which causes the sensors to be unstable as well.

But in any case, I would love to ear some experiences with those sensors, do you think they are accurate in a home environment? Are they good for making a device for selling (again, in a home environment)?

Any gas sensor must be tested before you sell it or you will discover why you need insurance and a competent law firm. In addition, gas sensors should be calibrated before being used. Each time, if you want reliable readings.

Have you actually priced the cans of gas to do your calibration? Likely not or you would not be thinking of selling your device.

The vendor of a nitrogen generation system tried to calibrate the oxygen sensor which told the system the purity of the nitrogen. Really low value means high quality nitrogen. Calibration required an hour and the 4 cans of nitrogen were not enough, so he gave up because the company would not supply any more cans. He would have to start over.

Find another project or don't attempt to sell your device.

Paul

Hi

I’m developing a device that measures the gases in the air, temperature and humidity that, when detected a gas, it alerts the user via SMS and WiFi, with the support of a mobile App.

I started learning how to do a schematic and a PCB on EasyEDA software, as it looks easier than others, but I think I’m doing something wrong, I don’t know why, but I feel that I should add a Ground and Vcc symbol on the schematic.

The schematic will be attached, and the PCB files, as well, if anyone wants to check.

If it looks all ok, please tell me, it’s kinda urgent.

In the power section (“painted.png”), the “DC” is the power input, the “LM7805”, “LM1117” and “LM2596” are power regulators.

Note: The “original.png” and “painted.png” are the same, the second one just have the red rectangle indicating the power section.

Best regards,
Pedro

Gerber_SMARTAIR_PCB_FINAL_2020-06-27_21-49-39.zip (83.3 KB)

Paul_KD7HB:
Any gas sensor must be tested before you sell it or you will discover why you need insurance and a competent law firm. In addition, gas sensors should be calibrated before being used. Each time, if you want reliable readings.

Have you actually priced the cans of gas to do your calibration? Likely not or you would not be thinking of selling your device.

The vendor of a nitrogen generation system tried to calibrate the oxygen sensor which told the system the purity of the nitrogen. Really low value means high quality nitrogen. Calibration required an hour and the 4 cans of nitrogen were not enough, so he gave up because the company would not supply any more cans. He would have to start over.

Find another project or don't attempt to sell your device.

Paul

Chill, I'm now trying to sell my project, it's a school project :smiley:

I ask that because, if they are good to do it, that means they are great.

I'm too young to start selling products, not to talk that many tests have to be done ahahah

Again, do you have experience with those? What do you think of theme?

Paul_KD7HB:
Any gas sensor must be tested before you sell it or you will discover why you need insurance and a competent law firm. In addition, gas sensors should be calibrated before being used. Each time, if you want reliable readings.

Have you actually priced the cans of gas to do your calibration? Likely not or you would not be thinking of selling your device.

The vendor of a nitrogen generation system tried to calibrate the oxygen sensor which told the system the purity of the nitrogen. Really low value means high quality nitrogen. Calibration required an hour and the 4 cans of nitrogen were not enough, so he gave up because the company would not supply any more cans. He would have to start over.

Find another project or don't attempt to sell your device.

Paul

And by the way Paul, sorry for the ignorance, but why would I need insurance and a competent law firm? I mean, that products for home don't need to be 100% accurate, my project is the same as a "Xiaomi Mi home Honeywell".

batista742k2:
And by the way Paul, sorry for the ignorance, but why would I need insurance and a competent law firm? I mean, that products for home don't need to be 100% accurate, my project is the same as a "Xiaomi Mi home Honeywell".

Oh, that is an easy question to answer. If a person relies on the you instrument and it gives an answer that there is no dangerous level of gas, and there is and the person gets sick or dies, your insurance and the lawyers you have will attempt to protect you in court.

Honeywell is an old company and has loads of liability insurance and a big staff of lawyers to defend them.

Paul

https://www.electronicshub.org/control-esp8266-over-internet/ scroll down to the section How to Control ESP8266 Over Internet?

Just a few observations (based on painted.jpg)

  1. Consider replacing the 555 timer with an Arduino pin and, maybe, a transistor. (Depending on buzzer specification)
  2. Check the data sheets of the voltage regulators to see if mandatory capacitors are specified.
  3. Check the crystal load capacitors
  4. Reverse a (freewheeling) diode across the buzzer
  5. Clarify connections on push_button
  6. Add a 100nF (ceramic) decoupling capacitor to (at least) the ATmega328p
  7. Check mixing of 3.3volt and 5volt tx/rx connections.
  8. Consider adding a ftdi header to the ATmega328p to program and debug it.

edit 1

  1. The networks LM2596_gnd and mainGnd are probably joined anyway at the LM2596 (looks like a buck converter module)

Hi,
Can you post an exported jpg image of your PCB please?

That way we can all see your PCB layout.

Thanks.. Tom... :slight_smile:

original.PNG

painted.png

ieee488:
https://www.electronicshub.org/control-esp8266-over-internet/ scroll down to the section How to Control ESP8266 Over Internet?

Thank you, ieee488

I have checked it, solved my problem, I think

Have a great day

TomGeorge:
Hi,
Can you post an exported jpg image of your PCB please?

That way we can all see your PCB layout.

Thanks… Tom… :slight_smile:

Thanks for your reply :smiley:

Well, I have attached it, but I can attach another one here :slight_smile:

Hi,
YOUR PCB LAYOUT not the schematic.

Tom.... :slight_smile:

6v6gt:
Just a few observations (based on painted.jpg)

  1. Consider replacing the 555 timer with an Arduino pin and, maybe, a transistor. (Depending on buzzer specification)
  2. Check the data sheets of the voltage regulators to see if mandatory capacitors are specified.
  3. Check the crystal load capacitors
  4. Reverse a (freewheeling) diode across the buzzer
  5. Clarify connections on push_button
  6. Add a 100nF (ceramic) decoupling capacitor to (at least) the ATmega328p
  7. Check mixing of 3.3volt and 5volt tx/rx connections.
  8. Consider adding a ftdi header to the ATmega328p to program and debug it.

edit 1

  1. The networks LM2596_gnd and mainGnd are probably joined anyway at the LM2596 (looks like a buck converter module)

Hi, 6v6gt

Thanks for your reply :smiley:

I may have to say that this project is already running, on a protoboard, but as it is hugly as hell, and the protoboard has some power limitations, I'm managing to create a PCB :D, so this power problems stop happening.

I think that if it is running on a protoboard, it will work on a PCB too :D, I just think that something is missing, because it was way easy to do it, and I'm just a newbie on PCB things ahahah

I will check some of the points you have stated

Thanks for your reply, again :smiley:

TomGeorge:
Hi,
YOUR PCB LAYOUT not the schematic.

Tom… :slight_smile:

Sure thing Tom, sorry, forgot completely.

It is attached right here :smiley:

Thanks again

"pcb_files.zip" - PCB files
"pcb.zip" - EasyEDA file (it contains pcb.json, because the arduino forum don’t support .JSON files)

Gerber_SMARTAIR_PCB_FINAL_2020-06-27_21-49-39.zip (83.3 KB)

pcb.zip (47.8 KB)

Hi,
You can in EDA Export as png, that will work.

Tom... :slight_smile:

TomGeorge:
You can in EDA Export as png, that will work.

Here it is,

pcb_bottomLayer.png” → Bottom Layer of PCB
pcb_topLayer.png” → Top Layer of PCB
pcb_topBottomLayers.png” → Both Layers of PCB