Particulate pollution Block

Post Reply
chmp
Newbie
Newbie
Posts: 6
Joined: Sat Apr 07, 2018 1:04 pm
Location: N 42 ° 42 ' 37 '', E 23 ° 18 ' 52 ''
Station model: Ambient Weather WS-1001-W
Software: Meteobridge
Contact:

Particulate pollution Block

Post by chmp » Sat Apr 06, 2019 7:21 am

Hello Jachym,
I have my own sensor for Particulate pollution.
I can use the API on the site and report the data, but I did not find a ready block to show them, except in the API Viewer. I have not found it or is it not made?

User avatar
Jachym
Site Admin
Site Admin
Posts: 1686
Joined: Fri Aug 18, 2017 10:12 pm
Location: Brno, Czech Republic
Station model: WH1080
Software: Meteobridge
Contact:

Re: Particulate pollution Block

Post by Jachym » Sat Apr 06, 2019 8:35 am

Hi,
this one is hard for me to answer, but I will be honest. Being the head of air quality deparment I have seen dozens of these sensors. At our institute we made very extensive analysis of these and the results are just ridiculous. It does not reflect even trends, not to mention absolute values. All these small sensors are hugely dependent on air temperature and humidity, and we call them "random number generators". Not even when you place two of these next to each other will they report at least similar numbers at all conditions.

If there was a possibility to use sensors worth 10-1000 USD we would use them, trust me. There is, however, a reason we use analyzers worth 10000+ USD, which are accurate, must be regularly calibrated with reference gas from reference laboratory (again something you cannot do to a sensor and which is absolutely crucial).

We try to make sure these do not spread much because it is tempting to buy one and yes it will show "some numbers" to make people happy, but...

I know you didnt want to hear this, but for this personal reason I do not want to develop any blocks supporting this other than blocks using professional data (and there are such). This of course doesnt mean you cannot use or make one yourself.

chmp
Newbie
Newbie
Posts: 6
Joined: Sat Apr 07, 2018 1:04 pm
Location: N 42 ° 42 ' 37 '', E 23 ° 18 ' 52 ''
Station model: Ambient Weather WS-1001-W
Software: Meteobridge
Contact:

Re: Particulate pollution Block

Post by chmp » Sat Apr 06, 2019 9:49 am

Thanks for the fast answer.

Unfortunately, you are absolutely right about the reliability of the sensors of $ 20-30.

There is a site in Bulgaria that is some kind of joint development with Germans. This site is used by all media and is distributed as absolutely trustworthy. And the whole of Bulgaria goes hysterical, with the site showing high values.

See the site description how do you make a station yourself and assemble it :D :D https://airbg.info/en/how-to/

Here are all the sensors http://sofia.maps.luftdaten.info/#5/43.978/27.353

Now I saw you have a few sensors in the Czech Republic :D

andyk1
Professional
Professional
Posts: 431
Joined: Tue Feb 27, 2018 10:56 pm
Location: OKLAHOMA, USA
Station model: Ecowitt WS90
Software: Meteobridge/GW2000B
Contact:

Re: Particulate pollution Block

Post by andyk1 » Sat Apr 06, 2019 10:20 am

I agree with Jachym 100%. I have spent nearly 2 years developing a Sensor array package written in python and am at a point where it is a waste of time and money. I worked on a system which included temp, hum, pressure, wind, direction, rain etc which also included all types of gases using sensors from amazon...MQ-2 MQ-3 MQ-4 MQ-5 MQ-6 MQ-7 MQ-8 MQ-9 MQ-135 etc. No 2 sensors read the same out of the box and required allot of calibration which was written into my python program to set as accurate a reading as I could on all sensors. The problem with the cheap sensors is that over time... some as low as a few weeks to months would require allot of re-calibration to the point it was useless. Almost all the gas and AQI require a 5 min warm up but over time the heater element would give out and the readings become useless. Even the Amazon 100-500 USD AQI's do not last a year before they stop reading correctly. As Jachym said. If you want accurate aqi readings it will cost more then the average person wants to invest. I always go back to the air quality readings from the local government run sensors as they are mostly reliably and not cheap sensors. I've pretty much given up on the aqi and gas sensors as they don't last long and just something to play around with but not worth the effort I've put into this project.
Image

Julius
Advisor
Advisor
Posts: 89
Joined: Fri Sep 24, 2021 10:41 pm
Station model: Barani MeteoHelix
Software: meteobridge

Re: Particulate pollution Block

Post by Julius » Sun Sep 26, 2021 3:29 pm

Ah, that's interesting. But I have to say, even the government controlled sensors over here are not very accurate or reliable. I have an SDS011 next to a SHT31 sensor inside an old wooden Stevenson screen, have been running it for quite a while now, and it's really remarkable (and fascinating I must say) to at least see the differences when winds are flowing from a road nearby and when they are not. You are right to say they can't be used as reference values, but still, they do show short term changes under otherwise (T and RH) stable conditions, and those can often be of value.
Either way, the SHT31 alone is worth the sensor-kit I once bought for this.

I was a bit surprised not to find an AQI block for MeteoTemplate.
https://sensor.community/ and https://opensensemap.org/ are still very interesting, even without the PM sensors.
Image

davidefa
Expert
Expert
Posts: 862
Joined: Tue Jan 12, 2021 8:03 am
Location: Italy
Station model: WH2650
Software: WH2650 (direct upload)
Contact:

Re: Particulate pollution Block

Post by davidefa » Sun Sep 26, 2021 8:40 pm

You can find an air quality block here ( in addition to the 'official' ones )
It calculates the 'instantaneous' aqi value ( based only on the last pm value ).
Image

andyk1
Professional
Professional
Posts: 431
Joined: Tue Feb 27, 2018 10:56 pm
Location: OKLAHOMA, USA
Station model: Ecowitt WS90
Software: Meteobridge/GW2000B
Contact:

Re: Particulate pollution Block

Post by andyk1 » Sun Jan 02, 2022 1:01 am

Given I wrote my post almost 3 years ago... Sensor tech has come a long way and even cheap $50 USD sensors are better than they where and for the most part are a reference but are still unusable in rainy and foggy conditions.
Image

Post Reply