Carbon Monoxide sensors

[manjh]

I just saw an item on TV about low-level, long-lasting CO intoxication. It immediately triggered me: if I could add a simple CO sensor to each and every ESP unit in my house, it would improve my safety enormously.
Before someone asks: yes, I already have a standalone CO sensor, and several smoke detectors.
But if I could add a sensor like the MQ-9 to each unit, at the cost of about 1$ each, it would be a lot better.

Has anyone thought about this? Or am I on uncharted grounds here?

[Martinus]

Hello Hans,

I've just heard the announcement on Tros Radar, but missed the item because i got a phone call.

So are we all soon to die because of this?

[manjh]

Hello Hans,

I've just heard the announcement on Tros Radar, but missed the item because i got a phone call.

So are we all soon to die because of this?

Not really. It was a faulty smoke exhaust pipe, that allowed a small amount of gas into the house. The eye-opener for me was the fact that a very low level of CO in the air (as low as 10 to 50 ppm) can give very serious health issues, without an obvious root cause. Complaints like headache, vague pains, etcetera.
I am aware of the higher level dangers, where faulty boilers treat us with a lot of CO. And we can all buy detectors to warn us.
But these same sensors donot measure as low as needed. We will need to go as low as 50 ppm, or better even down to 10 or 20.

[Shardan]

On top the usual sensors are unreliable.

I'm just running some testing with the usual MQ-7 and MQ-9 sensors.
They seem to work, basically. But they are extremely sensitive to smoke,
far more sensitive then to CO. Using them in a room were one single cigaret
was burning the sensor showed nearly full scale. In a room with a fireplace
you will never know: Is it smoke or is it CO? Some chinese shops even started
to sell them as CO & Smoke sensors......

I want to test other sensors but most sensors are running with a heated
SnO²-resistor element as the MQ-x types do. Sensors based on other
methods like the fibaro types are extremely expensive.

There might be some kind of filters to keep smoke particles from the sensor
but up to now I couldn't find any.

Regards
Shardan

[manjh]

How about the Figaro TGS5042? Quite more expensive than the MQ types, but I believe these are also used in industrial detectors, so I'm guessing they are more reliable?
Found one under 15 Euro at AliExpress:
https://nl.aliexpress.com/item/TGS5042- ... autifyAB=0

[Shardan]

How about the Figaro TGS5042? Quite more expensive than the MQ types, but I believe these are also used in industrial detectors, so I'm guessing they are more reliable?
Found one under 15 Euro at AliExpress:
https://nl.aliexpress.com/item/TGS5042- ... autifyAB=0

I was thinking a lot about the figaro sensors as they are quite nice.
But reading the data sheet shows they are somewhat difficult to handle.
They need a special OpAmp type badly availlable at my dealers with ultra low input off set.
This OpAmp adds to the sensor and it is not cheap.
On top the OpAmp as far as I can see needs a symmetric power supply making things even more
complex and expensive.

I tend to give it a try if
- i get an OpAmp for this somewhere
- i find a way to avoid symmetric power supply
- and there is a way to integrate that "fat" sensor into small cases together with other devices....

Honestly: not very likely

[manjh]

I agree, it becomes complicated and expensive.

I checked Banggood and AliExpress and found really cheap sensors, in the pricerange of 1 Euro each. If only that would work reliaby, I would add a sensor to each and every ESP unit that I run, so about 15 in total....

But I guess it's either a lot of money and a complex setup, or a few Euro's for a non-reliable but simple design.

Since this CO issue is no joke, I would not bet my life on the second option.
For now I think it is best to simply buy an extra standalone unit, adding to the one I already have installed.
No linkage to Domoticz, but at least it will work reliably.

[Shardan]

Definitely.
Anyways, there should be a filter for smoke and such.
I've bought a simple CO sensor for about 6€ (Complete wall mount sensor).
It most likely works with the SnO²-sensor method, an electrochemical sensor is no way for
that price.
That sensor is not affected by smoke!

I'm wondering if i should break the sealed cover and see how they did that.

The OpAmp issue is solveable, I've found an OpAmp that meets the requierements for about 3€.
(Type LTC 2050 by Linear Technologies).
The Issue about symmetric power supply might be solveable by using a precision voltage source
(basically a special type of zener diode), for example a LM4050-2.5. So you can build an artificial ground
half way from 5V supply.
This might need an ADS1115 for ADC as a much wider range and a differential input is needed for the input.

hm.......

Regards
Shardan