I just noticed my Rachio was on rain skip this evening and I wondered why. I took a look at the zones and it showed that it rained yesterday (8/16 - .44 inches) when in fact it had NOT. It did rain nearly an inch two days ago (8/15 - .87 inches) and that was correctly recorded in Rachio. Check out the image I uploaded to see for yourself. Has anyone else seen this before or have any insight why this might happen?
Also, just so you know my Rachio is tied to my personal weather station in my backyard (https://www.wunderground.com/personal-weather-station/dashboard?ID=KCOPARKE162), and the precipitation values show no precipitation on 8/16 which is correct. I’ve also never had an issue with the values being incorrectly transferred to Rachio since I’ve been using it.
I noticed that the rainfall on August 15 was late in the afternoon. One hunch is that the daily weather data collection for flex daily operates on UTC, and some of the August 15 precip was incorrectly double counted for the next day.
I just took a second look on PWS, and see that the observation for 7:30 PM on Aug 15 shows daily precip of 0.42 inch. The data points before and after show 0.87. The temp and pressure readings are also inconsistent. Not sure if this bad reading is the cause of your problem.
How do you get your weather data data loaded into PWS?
Unfortunately it does seem that the rogue 0.44" of precipitation has made it’s way into your data on 8/16/17 at 4:54AM and again at 12:20PM. Alas there is a known issue with the data coming from the wunderground via their public API endpoint which was used by wufyi so far.
The good news is that we are testing a working solution, which has proven to be much more accurate, over at beta.wufyi.com and it will soon make it’s way over to the general wufyi.com implementation.
If you wish to make use of a better data interface before it is available to a general public, just edit your cron-job URL from starting with “wufyi.com?..” to start instead with “beta.wufyi.com?..” (where … are your custom parameters) and you should automatically start getting a cleaner data source, otherwise your URL will start fetching from improved data source early next week after the testing is complete.
Gene - since this is only the first time this has happened and you’re so close to updating the backend fetch process, I’ll just leave it as is and look forward to the update next week. Thanks a ton!
