WD wind gusts are low compared to WeatherLink

I did a comparison of wind gusts for WD, WeatherLink (WL), and Wunderground (WU) from my VP2 for 2 March 2018 during the recent Nor’easter on the US East coast (I live in Northern Virginia). I’m running WD 10.37S B66, which uploads to WU. Both WD and WL run on my PC and get the data from my console data logger via WeatherLink IP. The maximum gust recorded on this day by WL was 35 mph, WD was 26 mph, and WU was 23 mph. I created a couple of Excel charts (attached) to demonstrate the difference in gusts. I used WD and WL 1-minute gust and WU 5-minute gust (downloaded from WU webpage). Looking at the chart for entire day (0000-2400 ET) the WL-recorded gusts (green) are consistently higher than the WD (blue) and WU (yellow). The average gust for the 24 hour period are: WL: 14.9 mph, WD: 9.0 mph, and WU: 9.3 mph. The 1100-1400 ET chart shows the maximum gust for the day of 35 mph recorded by WL but WD recorded 12 mph and WU recorded 14 mph at the same time.

Is this inherent in WD in that it does not record the same gusts as WL or might I have a setting wrong in WD? I would think that 1-minute WD gusts should be pretty close to WL. Anyone have suggestions on how to get WD to [closely] match WL or can this not be fixed?


I often get winds over 50mph
and WD has the same gust speed as the console
as WD requests the last 10 minute gust from the console, every 10 minutes, and uses that value, if its higher than what WD had recorded from the real time data
but it might be that is not happening with an IP data logger,if you have that set to disconnect every minute,in WD, to allow an update to weatherlink.com

I don’t think I have a disconnect set every minute in WD to allow to allow an update to weatherlink.com. Where would I find that setting?

in the control panel,com port setup, IP setup

I did have the IP data logger set to disconnect every minute in WD, to allow an update to weatherlink.com. I unticked it but as you can see in the attached Excel graph for today from 0000-0800 ET, WD gusts are still lower than WeatherLink. For the 8-hour period, the WL average gust was 5.5 mph and WD was 3.1 mph (40% lower) while the max WL gust was 14 mph and WD was 11 mph (25% lower).

Any other suggestions for something I might have set incorrectly in WD?


are you sure you have the units correct?
how/where are you getting the data from WD for the graph?
i,e it might be you are comparing kts with mph

How are you running both WD and WL?

C:\wdisplay\databackup\databackup32018.zip\32018lg.txt. The values from this file match the values plotted from 0000-0800 (within red box) on the attached file, curr24hourgraph.gif, which is plotted in mph. The Weatherlink data is also in mph.


niko: I’m not sure what you’re asking. Both WD and WL are running on my Windows 10 PC. Both get their data from the WeatherLink IP data logger. See attached screen captures of the comm port setup from WD and WL. Does that answer your question?


Understood, thanks. I wondered if you were using something to manage the two programs competing for the logger’s attention. I’m surprised it’s happy running in parallel like that :?

Yeah, I had wondered if they’d run in parallel too - I do have a 7th generation Intel Core i7 PC with 12 GB or RAM. For a while I was only running WD and then decided to try WL since I paid for the data logger and software when I purchased my new VP2 I thought I might as well try WL software. It’s not nearly as capable as WD and it’s kind of clunky but I believe it’s recording the gusts correctly. We don’t often have sustained wind storms in Virginia, but the Nor’easter the other day made world news due to the deaths, down trees, and 1.2 million homes without power. I live about 5 miles from Dulles (IAD) and the ASOS recorded wind gusts near 70 mph with over 36 hours of gusts in the 50’s and 60’s. Even though my VP2 ISS is more sheltered than the airport, I find it hard to believe that the maximum gust was 26 mph. That’s why I think 35 mph is more likely. I probably have some setting wrong in WD!

I just checked two stations on Wunderground, both less than a mile from my house, and one station is a AcuRite Pro Weather Center, which recorded two gusts at 30 mph and the other has a Davis Vantage Vue (Wireless) running weatherlink.com 1.10 and they recorded two gusts at 37 mph and one at 35 mph which more closely matches my 35 mph gust from weatherlink, not the 26 mph from WD.

the davis weatherlink IP data logger is not designed to have 2 connections like that
WD is likely missing out on data

Okay, thanks. I shutdown weatherlink this morning and tomorrow is supposed to be a relatively windy day (gusts to 30-35 mph). I’ll see how my wind gusts compare to the two other weather stations in my neighborhood.

you can see the daily max on the console easy enough

The console max gust for today was 24 and WD was 20. Is that just an artifact of the way the console receives the data versus how often WD grabs it?

I don’t know how WD works (but I’ll proffer a view anyway! :lol:), but WD will not see every ‘live’ update as it cannot receive loop packets continuously, it has to do other stuff stuff as well which requires different queries. Also I’m not convinced the loop packets come in fast enough to send every 2.5 second windspeed reading?

The way Cumulus handles this is to read the highest gust for today from the console every minute, and use that if is higher than the highest value obtained from the “live” data. It does mean that some lower speed gusts may still be missed, but ensures that the console and your recorded data agree about the value and time (approx) of the days highest gust figure.

Maybe something that Brian could add to WD? - Assuming it doesn’t already do something similar.

WD has a similar gust comparison:

Just checked and WD’s max windspeed (gust) for the day matches the console.

Maybe I have an option set incorrectly in WD?

Maybe it’s an issue with the data logger - it’s new as of January 2018 but my console is the original I purchased new in 2005. I still have my old serial-to-USB data logger, which I could try.