So the *truth* about how the Fine Offset WH24 sensor module (aka Ambient 1xxx clones) detects wind speed, then how the console interprets that data, is not as indicated above, and is actually worse than expected.
I have an Ambient 1200 clone version: the Fine Offset WH24 sensor module, a Fine Offset indoor sensor module, a Meteobridge & Meteostick, and a Fine Offset color TTF Console.
Since I can SSH into the Meteobridge, I can intercept the actual bytes being sent out of the WH24 sensor module. Based on work done by others and posted online for earlier versions of Fine Offset sensors, several months ago I reverse-engineered the meaning of all 16 bytes of data sent out by the WH24. Was not too hard, and now I have a small spreadsheet I can cut-and-paste the true-raw bytes into, and get all the metric and imperial results as would be shown on the Console, or the Meteobridge, or the IPobserver.
So today I took my weather station offline, brought down the WH24 sensor, and did some direct testing of how the anemometer actually is measuring windspeed. When one spins the anemometer by hand, you can hear the click of the reed switch as the magnet spins on top of it. So one can easily hand-spin the anemometer, and count the number of clicks heard.
First I looked at the "
average windspeed" sensor. Looking at the raw bytes, my initial guess was that it is: the quantity of revolutions of the anemometer in the 16 second interval. However, I was somewhat shocked to find that its actually and only:
the quantity of revolutions in the last 8 seconds of the 16 second interval; and all revolutions in the first 8 seconds are ignored. I tested this quite thoroughly, doing the below a number of times using different rates of spin of the anemometer:
- If I hand-spun the anemometer *only* in the first 8 seconds of the 16 second interval, the sensor always reported 0 revolutions aka 0 windspeed no matter how many clicks of the reed switch I heard.
- If I hand-spun the anemometer after first letting 8 second elapse with no spin, the sensor reported as revolutions the exact quantity of clicks I heard from the reed switch.
- If I hand-spun the anemometer briefly in the first 8 seconds while counting the clicks, then again briefly in the second 8 seconds while counting again the clicks, the reported quantity of revolutions from the sensor was ONLY the clicks from the final 8 seconds.
So the "average windspeed" is even less accurate than supposed from the WH24. We are getting wind readings from effectively half the 24 hour period: 8 seconds ignored then 8 seconds measured then 8 seconds ignored, and so on. If you get a brief wind in the first 8 seconds of the 16 second interval, then dead air, the average windspeed sensor will report 0 wind speed for the entire 16-second interval.
And average windspeed is measured very simply: it is just the quantity of revolutions in the final 8 second measurement period. It does not matter if the anemometer spins fast and slows down in the 8 second measurement period, or if the anemometer has a constant spin in the 8 second period; it is just the count of the reed switch clicks in that 8 second period.
Next I looked at the "
gust windspeed" sensor. When decoding the bytes, the conversion factor from revs/sec to windspeed (in MPH or m/s) for wind gusts is exactly 8 times larger than the conversion factor for the average windspeed. So my initial guess was the sensor was measuring the largest number of revolutions out of eight 2-second periods; in 16 seconds it would have 8 measures and keeps the largest value. However, my initial guess was wrong. The gust windspeed sensor is actually measured very simply:
it is just the largest quantity of revolutions (reed switch clicks) of the anemometer seen in any one of the eight final 1-second periods in the larger 16-second interval; the initial eight 1-second periods are ignored.
The gust windspeed on the WH24 ignores the first 8 seconds of the 16 second interval:
- If I hand-spun the anemometer *only* in the first 8 seconds of the 16 second interval, the sensor always reported 0 gust windspeed.
- If I hand-spun the anemometer after first letting 8 second elapse with no spin, the sensor reported a gust windspeed.
- If I hand-spun the anemometer slowly in the first 8 seconds, then very fast in the second 8 seconds, the reported gust windspeed had a high value.
- If I hand-spun the anemometer very fast in the first 8 seconds, then slow in the second 8 seconds, the reported gust windspeed had a low value.
So the gust windspeed suffers from the same "missing half the day" problem as the "average windspeed". If you get a brief gust in the first 8 seconds of the 16 second interval, then dead air, the gust windspeed sensor will report 0 wind speed for the entire 16-second interval.
And the gust windspeed is not an average ... its the largest spin rate-per-second observed, but only in the final 8 one-second intervals of the entire 16 second interval. I did quite a bit of testing for this, doing controlled slow then fast or none then fast, or fast then fast spins of the anemometer in that final 8 seconds. Empirically I observed that the largest quantity of reed switch clicks (revolutions) was being reported by the gust windspeed sensor, not some kind of average.
And given the characteristics of the anemometer (its diameter, cup size, etc), it turns out that one revolution per one second equates to 1.1 meters/second of airspeed (2.46 MPH). This is why you will never see a reported gust windspeed less than 1.1 m/s, and why the reported gust windspeeds are always a multiple of 1.1 m/s. The average windspeed is measured over 8 seconds rather than one second, and so it has 8-times the "resolution", or will increment in multiples of 0.14 m/s (0.3 MPH).
I hope this helps all understand the limitations of what the sensor actually reports to software like the Console, IPobserver, or Meteobridge. That software post-processes those actual sensor measurements, performing generally more averaging of the data. But in keeping the cost low for the sensor module, the tradeoff is the actual accuracy of the measurement technique used for the windspeed sensing. Its just a hobbyist weather station, and gives decent results. But if you are looking for absolute accuracy and high resolution of windspeed and other measures, well the WH24 has its limitations.
I do hope some others will repeat this physical testing of the WH24 sensor module to confirm or refute my results.
