MQ Suggestions

What would be the most reliable way to deploy MQ sensors in my home and garage without having to manually calibrate each one, etc... I've been using analog with a ppm trigger of 400, etc... But, the ppm varies from day to day. I haven't found it to be reliable yet.

What is the best way to approach this? Analog or Digital...

And if analog, would it make sense to wait 5 minutes for it to heat up, then save the current ppm into a variable, then let the trigger be say 200ppm (or whatever it takes) above the initial variable snapshot?? That way it would have a self calibration so to speak of the environment on bootup and maybe even periodically reset again every few hours.

Also, is MQ2 the best choice for smoke only?

The MQ sensors are very cheap, but they are not accurate. They draw power and they get hot. They should not be used outdoors with low and high temperatures. They need time to heat up, a few minutes should work.
They can not be used for serious ppm measurements.
Despite all of the disadvantages, they are very sensitive. So it is fun to see how they react.

Good sensors are very expensive. For a long time there was nothing in between.

Since a few years there are small chip-sensors. They might work in the same way as MQ sensors, but on a much smaller scale.
For example:

Yes, I see, they are much higher. Power consumption is not an issue as none will be on batteries in the house.

I just want something that is reliable and will last a long time. One will be outdoors under a patio area.

What do real smoke detectors use? Not the photo ones, just the gas detection ones.

Real smoke detectors ? They are often for indoor use only (stable temperature, moderate humidity). I think the good ones use photo/light/particle detection. It is for a reason, they have to work reliable for 10 years.

The Fire Angel ST-622-BNLT has good test results in the Netherlands (advised and sold by burn wound prevention foundation). So what does it use ? Optical detection of course. They added a heat sensor to detect hot fumes near the ceiling.

Even CO2 detectors can be made with IR light.
Such as this one: CO₂ Humidity and Temperature Sensor - SCD30 - SEN-15112 - SparkFun Electronics.
It works from 0°C to 50°C for 15 years.

Have you seen test results of cheap gas alarms ? Some are bad quality, and calibrating them was probably too expensive.

The chip sensors are a cheap way for small devices to detect gasses or smoke fumes. You have to check the temperature and humidity range if you want to use them outdoors.

The MQ sensor are to play with. You get what you pay for.

Thanks I'll get a SGP30 to play with. Although that's Co2.

Any idea on a reliable commercial grade photo to play with? Photos must be fairly inexpensive if they can sell the Fire Angel for $30 US.

Even photo smoke detectors are miniaturized: SparkFun Photodetector Breakout - MAX30101 (Qwiic) - SEN-16474 - SparkFun Electronics. The MAX30101 could cost 5 dollars, the rest is for the module and Sparkfun.

Hmmm, that's weird, the MAX30101 is a pulse oximetry and heart-rate monitor module.
The MAX30105 is made for smoke detectors.
The explanation is here.
I think it means that for normal persons the sensor can not be used to detect smoke.

A smoke detector of 10 dollars has a chamber with a IR led and a photo led: Connected Smoke Detector | Hackaday.io.
Modules with such chambers can be found at AliExpress. I don't know if they can be used with Arduino.

Have any of you all detected smoke with a MAX30105? I've seen where a couple of people online have tried but it looks like a very challenging project. Anyone had luck with any photo sensor and arduino?

maxim integrated wants users to no longer use the soon EOL MAX30105 and use the MAX30101 instead ("Not recommended for future designs. Please refer to MAX30101."). The latter is marketed primarily for wearable health (where the wonga is these days, lol), but also is functional for smoke detection and many other applications where you want to accurately measure something where reflected light comes into play.

It depends, of course, what in your use case "smoke" really is, because it can mean smaller or larger particles and VOCs. To tune the code that you get from the Sparkfun library it is probably best to produce the kinds of smoke you want to detect.

This topic was automatically closed 120 days after the last reply. New replies are no longer allowed.