Reliability of AQE

[Dario Salvi]

Hi,

I am interested into buying an AQE for a community project in Oxford about air quality and environmental monitoring. The reliability of the data is important for us as it could drive decisions about investments in the community.
I have tried to make up an idea reading the wiki and the messages in this group but I haven't found a conclusive statement.

Are data generated by the AQE reliable? To what extent?
If I buy a device, would I need to calibrate it?

Thanks in advance for your support.


Dario

[con...@cleanairuk.org]

Hi Dario,

Short answer: you would need to calibrate it. Also, experience shows
that it is unrealible for montiroing nitrogen dioxide air pollution.
Try something that works!

See our webpage under Air Quality Egg title for our experience:
http://www.cleanairuk.org/citizen-science-2013.html

Other people have also found that the sensor used on the AQE (which is
the same as the citi-sense) one is unrealible.

Incidentally, I also live in Oxford too.

Best wishes,

Andrew

Andrew Wood, Network for Clean Air.
www.cleanairuk.org

> --
> You received this message because you are subscribed to the Google
> Groups "#AirQualityEgg" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to airqualityeg...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[NeilH]

Hi, You might look at this
http://publiclab.org/notes/kensanfran/11-04-2014/smart-sensor-board-for-electrochemical-gas-sensors-basic-info

I posted at the beginning of the project 2012 May 19th, an open source hardware project, on the need for calibration and trace-ability for QUALITY metrics.
https://groups.google.com/forum/#!topic/airqualityegg/LC9S9ic_KaY
There was an active discussion about methods in places including here
https://groups.google.com/forum/#!msg/airqualityegg/9a0GoVzRjLo/fO2lgMYFwOkJ

After two+ years, and watching these threads, calibration is a value added part of a measurement device - I don't buy devices unless it includes a specification on their calibrated range, and it matches the project quality specification.

[Victor Aprea]

Dario, Andrew, et. al,

It's important to remember that the Air Quality Egg project in many ways pioneered this field, and a number of projects that used the same sensor technologies (right down to the manufacturer's part number) in our wake. I think Andrew's claim is correct in spirit, though not entirely accurate technically. 

Metal oxide sensors have a high degree of performance variability, they have a high degree of co-variance with temperature and humidity, and we have never had the capacity to calibrate them and account for all those factors in software (and have never claimed to), but with our current processes we do our best to at least normalize their responses. With the right facilities at your disposal, one could certainly put the effort in to characterize her personal sensors' responses and modify software to use those results.

I want to take this opportunity to look to the future. We haven't been talking much about it, but we haven't been resting on our laurels. Quite the contrary, we've been actively working on a next generation Air Quality Egg (v2) that uses fundamentally different sensor technology (that is factory calibrated) in hopes of putting some focus on better data quality. We are hoping to share more information in the coming weeks. Hopefully the community is as excited about this prospect as we are.

Kind Regards,

Vic

Victor Aprea // Wicked Device

victor...@wickeddevice.com

T: @vicatcu

[con...@cleanairuk.org]

Hi NeilH,

Thanks for the link but it looks substantially similar to what's
already available - whether it be AQE or Citi-sense.

Thanks,

Andrew

[Dario Salvi]

Hi Andrew,

Thank you for the info.

What a fortunate coincidence that we live close.
Our objective is to monitor the effects on pollution of the works going on at the Westgate mall. If we manage to have reliable measurements, we could ask the City Council to plant more trees and the like.

Would you be interested in the initiative?

I see that an involvement of Clean Air UK would be perfect!

Dario

[César García Sáez]

Hi Darío,

What a coincidence you're working there measuring air quality. If we knew if before hand you could have taken some of the AQE we have at Makespace Madrid :-)

Regarding quality, we did some studies with UPM ambientql sciences department here in Madrid and there was correlation with official stations but dataset was never released because of bureaucratic problems.  If there is any station near you could try to calculate deviation from the calibrated measurement, but I'm not sure it will be very useful. These sensors have a valid lifespan of one or two years working and they tend to drift over time.

My suggestions: check the epa air quality sensor evaluation and guidelines, published in 2014 that include all kind of technical details about these kinds of sensors if you are interested on the technical aspects.

Smart citizen are working on Barcelona to include new calibrated sensors but if will be much more expensive (500 vs 100 eur right now) and they are not ready for deployment.

My best guess right now would be to use passive tubes to measure calibration at the spot you're monitoring and track pollutants over time. They offer a reliable measurement and they're not very expensive, like 15 eur a piece including lab analysis. They are several projects using those in uk so you'll be able to experienced people for sure.

If you need any pointer or additional link, feel free to ask :-)

Best,

Cesar

[Dario Salvi]

Hola Cesar !

I could't wait too much to get involved in some makers' project :)


So, as far as I've understood, the current, cheap sensor need to be calibrated and they tend to drift over time.
I guess these sensor are thought to be used for sudden changes in gas concentrations for indoor alarms and this is why they are unsuitable for long lasting and reliable measurements.
To answer you:
1) I have access to air quality measurements daily at http://www.oxford-airwatch.aeat.co.uk/ I could maybe use these data to calibrate my sensors, but I will have to locate the sensor close to the measuring station for some days.
2) Using tubes is possible, but people were complaining about their precision, and they are not in my interest anyway.

I was wondering if ALL sensors are unreliable or if you have experience with some more reliable ones.

What about dust sensors? Are they also unreliable?


Cheers,


Dario

[César García Sáez]

Dust sensors tend to be more reliable, at least from the tests on the list. Check older threats on the group.

You can find better sensors from Alphasense for example, but they are 20x expensive. Smart Citizen are working on those.

Check this paper on sensors if you need more specific information: http://www.epa.gov/airscience/docs/sensor-evaluation-report.pdf

Best,

César

[Dario Salvi]

Thank you Cesar, that's a very good source of information !!!

[Dario Salvi]

Hi,

I've just found another interesting report:

http://www.everyaware.eu/resources/deliverables/D1_1.pdf

similar, negative, conclusions.


Cheers.