Switch from Console to WLL

[iandrews]

Been running my WLL (and Airlink) for a few weeks now, and all seems ok with them. I have set up the Airlink in Cumulus, but now need to do the WLL. Are there any gotchas I need to look out for when switching, or is it just a case of changing the station type, and putting in the WLL details. Thought I saw a post recently about someone's rain total going wrong when they did it.

One of the reasons I got the WLL was because I recently moved my wind and Solar / UV sensors to a pole up on the side of the house. Since I did the move a few months ago, today was one of the first days to have (almost) continuous sunshine from sunrise to sunset, and my trace shows that, though not sure if the dip about 13:30 is some cloud or the sun going behind some branches at the top of a tree a few gardens away.

Sun.JPG

Hopefully I get as good a uninterrupted trace in the summer months, though think I may get some houses the other side of the road blocking the early morning sun until it rises above the rooflines, and in the evening I think a tree to the West is going to cause some shade.

[mcrossley]

You shouldn't get a rain "glitch" if your year to date rain total is the same on both your VP2 and WLL. Cumulus uses that figure to register increases in rain. If your WLL was set up before 1st Jan and you use January as the start of your rain year you should be OK. You can adjust the WLL value either via the app or the web settings.

[iandrews]

Thanks. Both Console, Cumulus and WLL have the rain year starting on January, but as I set the WLL up on about the 8th Jan, the readings are different between the console and WLL.

Console ytd - 62.2mm

Cumulus ytd - 62.2mm

today.ini

[Rain]
High=0
HTime=00:00
HourlyHigh=0
HHourlyTime=00:00
Start=62.200000000000003
Yesterday=0
LastTip=2021-01-24 23:00
ConsecutiveRainDays=0
ConsecutiveDryDays=1
RG11Today=0

year.ini

[Rain]
High=31.200000000000003
HTime=2021-01-20T23:38:10
HourlyHigh=4
HHourlyTime=2021-01-14T01:30:00
DailyHigh=28.800000000000004
HDailyTime=2021-01-14T18:25:15
MonthlyHigh=62.200000000000003
HMonthlyTime=2021-01-24T23:00:30
LongestDryPeriod=3
LongestDryPeriodTime=2021-01-09T00:00:00
LongestWetPeriod=7
LongestWetPeriodTime=2021-01-16T00:00:00


WLL

57.8mm for both Month and Year.

I Edited the "Year to date rainfall" under the ISS configuration and at first put in 4.4 (the difference), but then the yearly rainfall showed 4.4, so did again but with 62.2 and now the WLL web page (and app) shows 57.8 for Month and 62.2 for Year.

Assume that will fix any glitch occurring, and I should just be able to configure Cumulus to use the WLL now.

[mcrossley]

Yes it should, Cumulus just uses the year to date figure.

[iandrews]

Set up a version of Cumulus MX on a spare (virtual) server so that I could test the WLL on Cumulus before moving my main install over. The main install is currently 3.9.4 b3099 while the new setup is 3.9.7 b3107. It's been running overnight, and I have noticed the following differences this morning. Not sure if it's a Console vs WLL or Cumulus version difference.

The first is wind speed:

Original set up, Data coming from Console (via USB logger and Meteobridge).

conwindspeed.JPG

WLL Data

wllwindspeed.JPG

The WLL Wind Speed line is a lot more fluid (less blocky / straight lines), seems like it's reporting in a .1 resolution rather then whole numbers.

The second is wind direction:

Original set up.

conwinddir.JPG

WLL Data

wllwinddir.JPG

On the WLL one there are quire a few blue Bearing dots right at the top of the chart, but aren't there on the console one. I have checked and "Use bearing zero when calm" is off for both installs, not sure if there is a settings on the WLL that reports this when wind speed is zero. With the original setup Cumulus shows a red line (dots) at the bottom of the chart when wind is 0.

[mcrossley]

Point 1 - yes the WLL reports average speeds to 0.1 mph not whole mph as the console does.

Point 2 - What was the wind speed at the time, were lulls to zero likely? And can you switch on debug and data logging, let MX run until you have collected a few samples where you get the dots at the top of the plot. Then you can zip up the MXdiags file and post it please.
(I enable "use zero for calm" and see plots at the bottom of the chart, but they do correspond with lulls)

[iandrews]

Thanks for confirming point 1.

Re point 2, the wind chats posted above show the wind about 2mph at the time. I will start in debug mode when I have time later.

Thanks.

[mcrossley]

OK, so lulls would have been likely then. The WLL data is sampled more often than the console (with the WLL MX gets every 2.5 second update of wind data) which may have a bearing. But lets see what the log file shows.

[iandrews]

Under Program settings I have ticked Debug logging and Data logging and pressed Save. Is that it, or do I have to restart.

Also, got these chats from this morning (from before I turned on debug).

Console:

concalm.JPG

WLL

wllcalm.JPG

The wind speed chart shows the wind was calm between about 10:15 and 10:45. Seems the console may still report a bearing when wind is calm, but the WLL reports N (360?). Seems both report versions report N (0?) for average though.

[jon_iz]

Out of interest, does the WLL average the gust speed over the same period and are the archive intervals the same as the VP2?

[iandrews]

Latest file from the MXdiags folder attached. I was running debug between about 12:15 and 13:30.

Looking at the chart, there are several times the bearing shows at the top (e.g exactly 13:00).

wind.JPG

The wind speed at the time was low, but not totally calm.

windspeed.JPG

[iandrews]

Found the following on a couple of forums:

weather-watch.com forum

I also have found out that WLL sends the wind direction as N when no wind

wxforum.net forum

Learning all the interesting surprises about when the Live Link does and doesn't send data - for example, it stops sending valid wind direction data every time the wind speed goes to 0. And if still, it stops sending the avg wind directions too



So, assume that Cumulus is receiving a direction of N (or nothing) when the wind is zero (even though the wind speed chart isn't showing the 0).

What calculates the avg bearing, is that Cumulus, and if so, does that automatically show N (0 or 360) if the wind speed is recorded as 0mph.

[freddie]

In the meteorological world, if there is no speed then there is no direction. The direction gets set to 0 degrees. This is correct and appropriate behaviour.

[iandrews]

That makes sense (and what I was thinking), but 2 things,

1. The wind speed wasn't being shown as 0mph at the time (though it was a very low speed).
2. My original CumulusMX reading from the console still shows a bearing when 0mph, though the Avg bearing does show N when wind speed is 0mph.

[mcrossley]

OK, if you look at the last data record the WLL sent before the 13:00 log entry was written, you see...

Code: Select all

        "wind_speed_last": 0.00,
        "wind_dir_last": 0,

So there was a lull to zero at that point. In fact it was calm from 12:59:47 to 13:00:20 (13 data packets in a row with zero wind speed).

I think you are just seeing the higher resolution to the data that the WLL provides. The latest wind speed and direction plotted on the graphs is the last packet received before that minute.