I was checking my brightness today thinking about using some values for some home automation. I noticed I had a large spike this morning, when it would have been dark. It was off the chart, had to be a bad reading as there is no way some environmental could have caused a value that high unless I am mistaken.
Definitely seen this as well. Also seen this with my Ambient Weather WS-2092A where the UV would spike between midnight and 5am. The spike shows up on Wunderground and AmbientWeather.net. It’s not just a WeatherFlow issue.
Wow that is weird. I thought maybe lightning but there were no storms. I expect the sensor works from solar radiation correct? So wouldn’t it actually take a high amount of uv to actually cause it if it were real?
Also is brightness straight from the sensor because the UV and solar radiation values were not impacted. I wondered if they were all derived from a single value. If so maybe it’s a calculation issue somewhere.
It’s not a big deal for me but I was more curious than anything. My mind kept going through all the crazy ways such a high value was recorded, everything including lightning and even aliens, lol!
You’re not the only one. I’ve got a replacement Sky (July 2020) and a replacement Tempest (Sept 2020) that have had time to settle for UV. Given the big UV differences between the two, not sure if anything is accurate anymore. Sky was showing about 2.5 while Tempest was way up in the 8 values… way too high for this time of year in my latitude.
(Wind values on Tempest are still off during complete wind stops, but that’s another category on here.)
I’ve submitted another ticket for this one, but I’m not sure how reliable the units are given the frequency of problems. Station 25741.
Value is indeed bogus and technicians are pretty sure the sensor is ‘wacko’. Can you please open a ticket with customer support ? They’ll handle it from this point on.
Sure thanks Eric. One more question, did you all see this in the remote data? I assume yes since the app showed it.
I’m curious if abnormalities like this could be detected using some “ai” logic where you all (or the station owner) could be made aware automatically. Or would that be more of a headache? Granted you usually don’t see too much consumer tech hardware doing that type of self diagnosis other than newer vehicles (but typically that’s because ultimately that likely leads to revenue for the dealership’s garage). Anyway, just a thought. I know that sort of discussion would typically stay internal.
Thanks again. I’ll open the ticket with support this evening and see what they say. You all very responsive on these forums. Honestly the tempest is one of the best products I’ve purchased where the community forums are actively monitored by staff.
Matt, these weird spikes to SR and Lux are caused by anomalously high individual samples that can occur when switching the light sensor’s configuration from reading UV to reading Lux. We’ve got a fix in place for this that will be rolling out in the next firmware update (soon).
Very glad to hear the problem was identified, and a fix is on the way. Wunderground needs the super-sharp eyes of WeatherFlow technicians to find and fix the same issue on their system.
