Weatherflow Humidity Spikes

Any idea what is causing these spikes? They do not show on my WF page, only in WD (running S98)…

25 7 2019 17 44 70.2 77 62.6 29.988 6.7 13.2 345 0.000 0.000 1.175 22.356 71.9
25 7 2019 17 45 70.2 78 63.0 29.987 5.5 11.8 344 0.000 0.000 1.175 22.356 71.6
25 7 2019 17 46 70.1 78 62.9 29.985 5.4 11.8 344 0.000 0.000 1.175 22.356 71.6
25 7 2019 17 47 70.0 78 62.8 29.984 5.0 11.8 344 0.000 0.000 1.175 22.356 71.5
25 7 2019 17 48 70.0 78 62.9 29.983 6.9 13.6 342 0.000 0.000 1.175 22.356 71.6
25 7 2019 17 49 70.0 77 62.4 29.986 7.5 13.6 343 0.000 0.000 1.175 22.356 71.9
25 7 2019 17 50 69.6 44 46.7 29.987 8.2 21.5 341 0.000 0.000 1.175 22.356 77.1
25 7 2019 17 51 69.6 70 59.4 29.985 7.2 21.5 341 0.000 0.000 1.175 22.356 73.8
25 7 2019 17 52 69.6 78 62.5 29.984 5.5 9.6 347 0.000 0.000 1.175 22.356 71.4
25 7 2019 17 53 69.6 79 62.8 29.984 6.0 17.4 340 0.000 0.000 1.175 22.356 71.0
25 7 2019 17 54 69.6 79 62.8 29.986 6.8 17.4 342 0.000 0.000 1.175 22.356 71.0
25 7 2019 17 55 69.6 79 62.8 29.983 6.2 13.4 345 0.000 0.000 1.175 22.356 71.0


Seems WD stopped collecting all data around 1900 hrs. Initial thought is a problem with cronweatherflow as I see no gaps in smartweather data. Still trying to figure out why WD appears to be having trouble gathering data…


Here’s the graph after I re-started WD this morning looking clean so far. I don’t know what caused the humidity spikes but I confirmed the negative spikes were recorded in the log file. cronweatherflow was not running before I re-started WD. I checked my smartweather and the humidity spikes were not recorded in that data.

Has anybody else reported something similar?


The negative humidity spikes continue…now running S99…

See attached…cronweatherflow screen is updating once a minute (I thought before it updated anytime new data was received) and displays error 143.

Smartweather data grabbed once per minute does not show any dips in humidity.


is the station ID 6695 actually your station ID?

Yes it is

Initial troubleshooting points to a problem with my WD computer rather than WD or WF. I have setup WD on another computer and can monitor both at the same time since WF uses WiFi for connectivity. I will continue to monitor both setups and attempt to find the culprit, but for now I will remove my post from bug reports since I don’t believe WD or cronweatherflow is the issue…

Scraping barrel bottom, but a suggestion. Where do you get the 5 V for the HUB? I had all kinds of anomalies with the HUB powered from a computer USB socket. Changed to dedicated 5 V PSU and problems solved. Also, FYI, some USB splitter hubs can cause anomalies. I tried 2 passive 4-way hubs: problems (nothing to do with WF/WD). 1 active 4-way: no problem. 1 active 7-way: big problem.

Power for WFHub is from a USB hub…but I am now running WD on 2 computers both “grabbing” data from WF via WiFi…only 1 computer shows humidity spikes. So the choices boil down to either the old computer, that has been running WD and different versions for a long time, has something in the settings messed up, or, possible WiFi issue with the older computer. Once I get all running on the new computer I will wipe out all WD on the old computer and start from scratch and see if I can narrow down what is causing the spikes.

Thought I had found the issue, but turns out the spikes continue. I had noticed cronweatherflow was not updating as fast as it did when I first switched to the WF. Normally, cronweatherflow updates at something close to every second…mostly wind data, but updates happen almost constantly. Watching mine, it appeared to be updating, at most, every 30 seconds or longer. So the first thing I did was download the most recent full WD install. That has returned cronweatherflow to updating almost constantly again. I had hoped this would stop the spikes but it hasn’t.

So I decided to try Devil’s suggestion and moved my WFHub to get power from my new computer running WD. I figured it wouldn’t hurt anything to try it and at least prove that my old WD computer was not the problem. But this morning I see spikes on my console graph.

I will now continue to move all WD functions and history over to the new computer so I can wipe WD off my old computer and install from scratch. I suspect that something being performed in WD on my old computer that is not setup on the new computer is what causes the spikes. So I will be adding the missing functions one at a time until I discover the culprit…hoping it dupes…

Sorry! Perhaps I didn’t make myself clear. I found that I could have clean HUB results only if it is powered from an independent 5 V supply and never from a USB socket on a computer or a hub. I hypothesise that USB sockets on 'puters or on hubs perhaps use time-sharing to feed multiple outlets. My problems disappeared when I used an independent 230 V-powered 5 V source from a wall-socket. FWIW!

The hub is sensitive to having a good clean power supply. Some have even had problems with the stock WF supplied PS and had to switch to a higher quality unit.

I don’t think the WFhub is the problem…if the hub was the problem I would expect both instances of WD would exhibit the same issue. Since only 1 copy of WD is showing the spikes I have to conclude there is a setting or process that for some reason causes the humidity data to record much lower than actual. I have no idea what could cause this to happen.

Have you actually tried a clean 5 V source for the HUB, independent of any computer, USB source or whatever?

I’m sure as an experienced user you have probably checked this already, but just in case…

What I recall from other instances of “oscillating data” is that there has often been a second data source, like a metar, selected in WD that’s conflicting with the station data.

Niko was correct…a few months ago my Davis Vue station humidity sensor quit working (again) so I set WD to use humidity from metar. I hand edited the wdisplayftp.reg file to set “Use humidity from metar = no” and the spikes are gone. I assume the reason the spikes seemed intermittent was because of the difference (or lack of difference) in humidity readings.