Gas (Co2) sensors on batteries
-
Hello,
Anyone can recommend a gas sensor which I can run on two AA batteries? I have currently MQ-135 + Nano running from a 12v AC/DC adapter via a voltage regulator, but this is not precise and I cannot run it on batteries.
Did anyone have any experience with mh-z14 on batteries?
I know @epierre has done a lot of work in relation to a few gas sensors. Maybe there is a universal low power sensor capable measuring several gases (3-in-1 or something).
Regards
Alex -
I'm very interested in this too!
I have two Senseair K-30 CO2 sensor modules (http://www.co2meter.com/collections/co2-sensors/products/k-30-co2-sensor-module, http://www.senseair.com/products/oem-modules/k30), which use the non-dispersive infrared (NDIR) principle. The same principle is used in MH-Z14 as far as I understand.
The K-30 consumes quite some current and is not ideal for operation from batteries. However, my thougth is to connect it to batteries using a MOSFET, making it possible to switch on/off the sensor using an arduino output. Something like this:
In order to save battery the strategy is to:
- power up the sensor
- warm up for a few seconds
- take a measurement
- power down
- wait for some time (e.g. 10 minutes)
- start over again
The required warm-up time given in the datasheet is about 1 minute. However, a comparison of two K-30s - one powering up/down and the other powered on constantly - showed that a satisfactory measurement is obtained less than 10 seconds after powering up.
Any thougths on this issue will be very much appreciated.
-
Hello,
I did not manage to get the Senseair K-30 CO2 so I cannot tell for it. I would guess that all this is a good approch, given this fits with the sensor's maker datasheet.
I use my mh-z14 inside, so I've not tested on batteries, if I would make it outside I would use the ceech board with a 4000mAH LiPo and 1W solar panel which has proved to be the best configuration so far ! I have a barometric, uv and lux working this way for nearly one year yet and it is surviving the winter so far ! I guess a lower LiPo would do the job too but I feared the loss of light in winter.
-
@epierre Interesting. I see why you'd want to use ceech board outside, but I have the following setup:
http://forum.mysensors.org/topic/1901/my-own-board-50mm-x-30mm/95
I have several of these nodes as well as one sensebender in 3D printed boxes. Some of them are in the rooms and one next to the circuit breakers box in the cupboard . Currently, I have them measuring hum/temp/pressure (SI7021, BMP180 and BME280), but I'd like to add Co2 measurement and uv/lux as well (SI1145 + BH1750).
By the way Senseair K-30 is not good - voltage is 4.5 to 14.0 V DC and I have from 1.6 to 3.1 V. -
@epierre Interesting. I see why you'd want to use ceech board outside, but I have the following setup:
http://forum.mysensors.org/topic/1901/my-own-board-50mm-x-30mm/95
I have several of these nodes as well as one sensebender in 3D printed boxes. Some of them are in the rooms and one next to the circuit breakers box in the cupboard . Currently, I have them measuring hum/temp/pressure (SI7021, BMP180 and BME280), but I'd like to add Co2 measurement and uv/lux as well (SI1145 + BH1750).
By the way Senseair K-30 is not good - voltage is 4.5 to 14.0 V DC and I have from 1.6 to 3.1 V.@alexsh1 said:
@epierre Interesting. I see why you'd want to use ceech board outside, but I have the following setup:
the ceech has built in lipo/li-on and solar panel and all the led removed ;-)
By the way Senseair K-30 is not good - voltage is 4.5 to 14.0 V DC and I have from 1.6 to 3.1 V
yes hard for a step-up ... or you can have an alternate powering, two li-on in parallel e.g. but it would harder to load.
senseair is hard to obtain outside soime countries, currently to France I did find no way to get it.
-
@epierre There is a french company manufacturing a weather station called Netatmo. They manage to implement Co2 sensor battery powered. I wonder what Co2 sensor they have used. Do you know by any chance?
@alexsh1 the Co2 sensors in the Netatmo modules are the main reason I got the Netatmo system in the first place. The price is high but the performance is accordingly.
Anyhow I'm surprised that no one has opened a Netatmo indoor module to see whats inside. The only info I could find is that they use some kind of optical sensor.
-
@korttoma I have Netatmo myself and yes it is a good premium product. The problem comes if you want to add a new module - it is a whopping Euro 60 or around that. I have MQ-135 with mysensors for probably around Euro 10-12 and it is working really well with Domoticz and push notifications. Why paying more? Besides, I like building and programing things myself.
OK, I found an interesting module - http://www.co2meter.com/products/cozir-5-100-co2-sensor
Ultra-low power: 3.3V, 3.5mW
Peak current only 33mAIf I find it on aliexpress, this could be a good sensor for a battery powered node.
PS I just saw prices: COZIR 100% CO2 for $109.00! Crazy! Back to my search...
-
@arraWX I wonder in your setup how long you can run the Co2 sensor on 4 x 1.5V batteries. Assuming everything goes to sleep for 5-10mins and a Co2 measurement takes only 10 seconds.
I have been trying to keep voltage at 3.3V (2 x 1.5V batteries) running booloader @ 8Mhz or below.
-
@epierre There is a french company manufacturing a weather station called Netatmo. They manage to implement Co2 sensor battery powered. I wonder what Co2 sensor they have used. Do you know by any chance?
@alexsh1 said:
@epierre There is a french company manufacturing a weather station called Netatmo. They manage to implement Co2 sensor battery powered. I wonder what Co2 sensor they have used. Do you know by any chance?
haem... http://www.aragonvalley.com/en/assessment-netatmo-weather-station/
calibration has a cost, but too few found it useful to check if this was worth it... Many kickstarter propose readings of gases without calibration, they provide their own indexes... I've proved many times that you could be alarmed of too much gas just by restarting a sensor, or break a perfume bottle nearby and why would your CO2 raise so much ? only because there are many organic particles in the air, full stop.
-
@epierre - I understand, but paying $109 for a sensor is just ridiculous. I would spend on a node project about $10-20 per node and paying 10 times more for the sensor is just not right in my opinion.
Anyway back to google searching for a reasonably priced alternative.
How do you find mh-z14? Accurate? -
@epierre - I understand, but paying $109 for a sensor is just ridiculous. I would spend on a node project about $10-20 per node and paying 10 times more for the sensor is just not right in my opinion.
Anyway back to google searching for a reasonably priced alternative.
How do you find mh-z14? Accurate?@alexsh1 said:
@epierre - I understand, but paying $109 for a sensor is just ridiculous. I would spend on a node project about $10-20 per node and paying 10 times more for the sensor is just not right in my opinion.
Anyway back to google searching for a reasonably priced alternative.
How do you find mh-z14? Accurate?sure cost can be an issue, but you get what you pay for in the end... people say netatmo expensive but they have chosen their quality level, as many kickstarters projects...
MH-Z14 is very fine, you can see it on Davide Gironi posts where he compared both technology.
cooking hacks has calibrated CO or O3 gas sensor $200 ...too much for me too...
-
just a glimpse of where CookingHacks / libellium is now going: https://www.cooking-hacks.com/shop/sensors/gas
-
The circuit diagram that I posted above did not show communication lines between the K-30 and the sensebender. These are included here:
However, I now realise that the K-30 and the sensebender cannot be connected this way because the sensebender (ATmega328P) digital inputs do not accept voltages higher than Vcc + 0.5V.
Does anyone have a suggestion how this problem can be solved?
@alexsh1 As you can see my experiments with powering the K-30 from batteries are not going that well...
-
The circuit diagram that I posted above did not show communication lines between the K-30 and the sensebender. These are included here:
However, I now realise that the K-30 and the sensebender cannot be connected this way because the sensebender (ATmega328P) digital inputs do not accept voltages higher than Vcc + 0.5V.
Does anyone have a suggestion how this problem can be solved?
@alexsh1 As you can see my experiments with powering the K-30 from batteries are not going that well...
@arraWX OK, found an excellent sensor. The only problem is price - one has to pay for a precise sensor with ultra low power consumption.
Linearity:Nonlinear (% F.S.)
Sluggish:None (% F.S.)
Repeatability:Standard (% F.S.)
Sensitivity:High sensitivity
Drift:No
Resolution:High Resolution
MH-Z19-W01-GSS-COZIRCarbon dioxide sensor
Ultra-low-power carbon dioxide sensor
MH-Z19-W01-GSS-COZIR carbon dioxide sensor (hereinafter referred to as COZIR) COZIR is an ultra-low power consumption (3.5mW), high-precision carbon dioxide sensor, ideal for battery-powered portable instruments and equipment. Based on IR LED, detector technology and patented optical detection cavity, so COZIR become the lowest power NDIR sensor, optional temperature and humidity testing. GSS is carbon dioxide detection infrared LED leader in the field, COZIR GSS company is the third generation of products.
COZIR Ambient measurable range 0 ~ 2000PPM, 0 ~ 5000ppm and 0 to 1 percent, and is suitable for building control such as gardening and other occasions.
Ultra-low-power 3.5MW
Measuring range: 0 to 1%
Low noise measurement (<10PPM)
3.3V power supply
peak current is only 33 mA
Temperature and humidity output optional -
@arraWX OK, found an excellent sensor. The only problem is price - one has to pay for a precise sensor with ultra low power consumption.
Linearity:Nonlinear (% F.S.)
Sluggish:None (% F.S.)
Repeatability:Standard (% F.S.)
Sensitivity:High sensitivity
Drift:No
Resolution:High Resolution
MH-Z19-W01-GSS-COZIRCarbon dioxide sensor
Ultra-low-power carbon dioxide sensor
MH-Z19-W01-GSS-COZIR carbon dioxide sensor (hereinafter referred to as COZIR) COZIR is an ultra-low power consumption (3.5mW), high-precision carbon dioxide sensor, ideal for battery-powered portable instruments and equipment. Based on IR LED, detector technology and patented optical detection cavity, so COZIR become the lowest power NDIR sensor, optional temperature and humidity testing. GSS is carbon dioxide detection infrared LED leader in the field, COZIR GSS company is the third generation of products.
COZIR Ambient measurable range 0 ~ 2000PPM, 0 ~ 5000ppm and 0 to 1 percent, and is suitable for building control such as gardening and other occasions.
Ultra-low-power 3.5MW
Measuring range: 0 to 1%
Low noise measurement (<10PPM)
3.3V power supply
peak current is only 33 mA
Temperature and humidity output optional -
@alexsh1 at that prize you could buy a few Netatmo additional indoor modules. Might be that the sensor you suggested would be allot more accurate though.
@korttoma Yes, I know what you are saying, but Netatmo is not precise. I have MH-Z14 showing 750ppm and Netatmo 1100ppm, for example. I think Netatmo is using index.
All I can say that a calibrated low power CO2 sensor is very expensive. Probably for a home use you do not need to know the exact number. Personally, I have MH-Z14 - it is reasonably expensive and very accurate.
