Not enough observations

[weswitt]

Not enough observations

[jonathankoren]

The most common reason someone sees "not enough observations" is there are not enough observations to correctly calculate the AQI for the pollutant for the configured standard. For instance, a PM 2.5 US EPA NowCast requires 12 hours of continuous observations, with each hour having at least 75% of expected observations.

In the log you you pasted, it looks like 16 data points were found in the DB, but you need 540 in order to calculate an AQI.

If you have been recording less than 12 hours observations, then this is expected behavior. However, if you have been recording data for more than 12 hours, then something may be wrong.

My first two questions are:

1. How long should you have been recording data? In other words: How long has it been since weewx has been running with what you think is correct configuration to write PM 2.5 measurements into a database?

2. Are the observations actually being written to the DB? It looks like 16 were, so maybe that's not the issue? I don't know. It depends on what the answer to question 1 is. If the measurements are in a DB, then weewx-aqi may be misconfigured. If they're not there, then maybe your sensor is misconfigured or broken.

My advice is simply to wait, but since it's been 13 hours since you posted this, I guess you can check after your read this. :)

[weswitt]

thanks for the response.

i changed the data source that i'm providing to weewx-aqi. my data source emulates what purple-air provides -- a sql database with sensor reading values. once i did this everything works perfectly. weewx-aqi thinks it's talking to a purple-air database and all is good.

my next task it to add mqtt support to weewx-aqi so that the aqi values are reported out over mqtt and can be used by devices outside of weewx.

[jonathankoren]

Awesome. Great to hear this.