16-08-2017, 09:12
Hello,
I think that to make a comparison with a station which is +/- 35 km is rather hazardous.
Yesterday, with 35km W of Cercal the Lourinha station had 34AQI, with 60km E of Cercal the station of Chamusca had 42AQI.
If one compiles data of AQI.ORG and PurpleAir: for example in Chandigarh (India) the two networks have a station with +/- 10km.
http://aqicn.org/city/india/panchkula/se...chkula/fr/
https://www.purpleair.com/map?&zoom=8&la...4287109375
The differences are large.
The temperatures, the direction of the winds, the reactivity of the sensor, the ventilation of the sensor can distort calculations. It should not be forgotten that standards AQI it is a measurement doubts the hours (table 1) and an average during the last hour (table 2).
https://thingspeak.com/channels/182605/c...&type=line
https://thingspeak.com/channels/182605/c...&type=line
When I measure a AQI with my detector, it is impossible (without instruments of laboratory) to be certain at 100% of measurement.
![[Image: 26-01-10.jpg]](https://i37.servimg.com/u/f37/18/78/82/62/26-01-10.jpg)
On the graph how to be certain that the curve punt is not a bad ventilation or a saturation.
It is as the probe of t° NETATMO which is a true filth (overheating) but when one places probe NETATMO in a shelter DAVIS 7714, measurements are very close to a VP2.
Thus I think that it is necessary to make statements on a single place/environment.
If one assembles comparative on the same place with my detector + WeatherDuino, everyone would be winner.
My remark on the PurpleAir detector is simple: according to my Internet searches produces it is well conceived and was testè in a scientific environment.
We thus have a point of reference (+/- correct) and as it is a Arduino base…
But with the Portuguese customs, I do not want to pass by the case.
The finality of the my WeatherDuino future is the comparison of the various solutions.
This evening I will post photo information/my detector.
PS: I have emmerdeur (in french
) … to obsess by the precision… that irritates my wife for 35 years 
I think that to make a comparison with a station which is +/- 35 km is rather hazardous.
Yesterday, with 35km W of Cercal the Lourinha station had 34AQI, with 60km E of Cercal the station of Chamusca had 42AQI.
If one compiles data of AQI.ORG and PurpleAir: for example in Chandigarh (India) the two networks have a station with +/- 10km.
http://aqicn.org/city/india/panchkula/se...chkula/fr/
https://www.purpleair.com/map?&zoom=8&la...4287109375
The differences are large.
The temperatures, the direction of the winds, the reactivity of the sensor, the ventilation of the sensor can distort calculations. It should not be forgotten that standards AQI it is a measurement doubts the hours (table 1) and an average during the last hour (table 2).
https://thingspeak.com/channels/182605/c...&type=line
https://thingspeak.com/channels/182605/c...&type=line
When I measure a AQI with my detector, it is impossible (without instruments of laboratory) to be certain at 100% of measurement.
On the graph how to be certain that the curve punt is not a bad ventilation or a saturation.
It is as the probe of t° NETATMO which is a true filth (overheating) but when one places probe NETATMO in a shelter DAVIS 7714, measurements are very close to a VP2.
Thus I think that it is necessary to make statements on a single place/environment.
If one assembles comparative on the same place with my detector + WeatherDuino, everyone would be winner.
My remark on the PurpleAir detector is simple: according to my Internet searches produces it is well conceived and was testè in a scientific environment.
We thus have a point of reference (+/- correct) and as it is a Arduino base…
But with the Portuguese customs, I do not want to pass by the case.
The finality of the my WeatherDuino future is the comparison of the various solutions.
This evening I will post photo information/my detector.
PS: I have emmerdeur (in french
) … to obsess by the precision… that irritates my wife for 35 years