The timing of your solar spikes, is similar  but not exact to hifi_hound's data for today  (he is not sending). Although , he has 0 solar values.

https://www.wunderground.com/dashboard/pws/KCAEMERA6  (in the table view, compare your solar spike times, and his entries)

almost like there is pattern

Also, please post links to your WU sites.

INOARD3 is my WU id GW2000 Ecowitt
IFRIESLA82 is my WU id with meteobridge Davis

I will turn off upload to WU and see if any data comes up

Rover1822:  Yes KCAEMERA18 is my new ID.  I have never had an Ambient WS2000, but it is pretty much the the equivalent to my Ecowitt branded system, and WU doesn't list Ecowitt systems in their weather station choice's.  The only alternative is to list the station as "other".

I also noticed the location difference which explains why I would get UV spikes + temperature spikes in the middle of the night.   If you go back a few days or more, through the graph data for previous days, you will see what I mean.  It also explains why my low for the day was often lower than the actual data that would show on my Ecowitt site. This other station is sending daytime readings at random times during my evenings and early mornings, and nightime data during my days. I don't understand the reporting intervals though.  They are very random intervals.

The idea of someone typing in the wrong key by a digit seems strange though as I would think they would have to put in the station ID as well (KCAEMERA6), which would seem unlikely.

Interesting your data for KCAEMERA6 shows 0 for solar values today, also if you compare the reporting times for the last 2 days, they are roughly (very roughly) the same -- very weird

It is when you have CumulusMX running against the GW1100 but (my mistake) using the GW1000 archive on Ecowitt for catchup! You end up with negative day rain amount!   

Stuart

Why don't you have both your GW1000 and GW1100 post into the Ecowitt cloud and use the same MAC for backfill inside CMX which you use to retrieve the data into CMX (MAC related to that IP) ?

In the normal operation I don't see differences in CMX -  having 5 different CMX installations with 5 different GWyx00 devices -  not when the different consoles use the same rain gauge. Of course my three different rain gauges (WH65, WH40 and WS90) produce different counts during a rainfall event, but two of my GW1x00 are connected to the same rain gauge each.

By the way - I have two WS2320E consoles (same outdoor array) and the rain count of the daily rain dfferes sometimes by 0.1 mm - but as for the monthly and yearly totals they are the same again ...

My 2550 has been operating for just on a month, no issues at all.

In the last few days my console has started flashing a warning for my WH32, seems to be saying the battery is low.  I was surprised at this, expecting better battery life than that, but as I had some spare fresh AAs I threw them in.  No effect.  The warning continues to flash.  The data is still being collected by the console and uploaded, and the WH32 shows no distress at all, cycling between its two screens and showing no warnings at all.  On the surface it looks like I can ignore the warning on the console (except the cycling of that part of the console screen is rather annoying), but I wonder if there is a problem somewhere that I should attend to?

The WH32 has been sitting within a few cms of the console, so I don't think it is a RF transmission issue.

Does anyone have any thoughts, anyone had anything similar?