A tale of six ODINs

On Tuesday I told you about the tests of our PACMAN units before they were being deployed in our volunteers' homes. Now is the turn of ODIN (Outdoor Dust Information Node).

A little bit of history ... After the first PACMAN deployments we realised that we needed to have information from the outside of the home as well as the inside so we developed a stripped down version of PACMAN focusing solely on the dust measurements.

Here you can see a presentation we gave about the ODIN after the very first tests.

 

Now, we have six ODINs to be deployed in Rangiora but before we go ahead and distribute them across the town we need to know that they are working fine so we set all of them up at ECan's air quality station in Rangiora.

For this test we're looking for indications that the units are recording all the data, that the temperature looks plausible and that the dust sensors respond to something other than temperature.

In simple terms, we're looking for plots that look like this:

That's the plot for ODIN #06 which is promising ... unfortunately not all of the units were so well behaved ... in fact one of them only recorded 2 datapoints!

In all, 2 units were OK, 2 were OK-ish and 2 were dead wrong (see here the full test result ... including the ugly ones!)

After this, we replaced the dust sensor on one of the worst ones, fixed the clock on another one and replaced the temperature sensor on a third one. The one that only logged 2 datapoints required more extensive reconstruction so we have 5 to go with at the moment and the sixth one will be up and running very soon.

Let's hope that they behave for the next few weeks!