Wind gauge accuracy

New tempest owner and so far love it. Only question I have is has anyone compared the Tempest wind gauge to another anemometer? Reason I ask is I have a lacrosse anemometer in close proximity to the tempest and the readings are pretty far off. We had some really strong winds come through with the front 20mph+ tempest didn’t seem to get the same data. Anyone have any thoughts or experience?

Lacrosse anemometer

welcome to the forum WarEagleAU

Both systems do indeed read differently but the gap shouldn’t be that much. Can you share some pictures of both setups ?? The Tempest is pretty sensible to nearby objects that make turbulences.

keep in mind that when zooming out on the tempest data, the wind is averaged over longer periods. In your case it is the average over half an hour. However the gust data is the max over that period. If anything I would expect the gust data to be higher for the tempest, as it doesn’t have mechanics that need to get up to speed.

1 Like

Here is my setup

1 Like

The following is based on a small sample of data BUT…

I am finding the same thing in comparing the Tempest with the Ambient WS-3000. Both are ultrasonic but in the recent wind event… The Tempest highest gust was 25 MPH and the WS-3000 was near 40 MPH.

I did some checking and the WS-3000 is doing rapid sampling to show the wind speed… i.e. they check the wind speed at least every second, perhaps more often, and average the readings out to come up with a wind speed reading. They then use the highest wind speed in that collection to come up with the max gust. I have a support question to Ambient asking them for the sample rate that the WS3000 uses.

I don’t believe the Tempest works the same way in determining maximum gusts. Perhaps someone on this forum can provide the information on how exactly tempest determine max wind gust.

I should add that I live in a pretty gusty environment. The 1 acre lot does not have many trees but we are surrounded by trees in neighboring lots. The results is we get strong wind in gusts but they are short lived. I would expect the different between stations would be less in an open type environment. Ocean, fields, etc…

1 Like

Does tempest support monitor the threads or would we need to reach out for support?

Someone on this forum likely knows how wind gusts are calculated but a support ticket might get an answer quicker.

My assumption is as follows (assuming non-powersaving mode):
It emits several ultrasonic pulses to measure windspeed. It is send once every 3 seconds. The gust calculation is the max of those transmitted data during some period. The shortest period is 1 minute.

Thanks… I received a reply from Ambient saying they sample every second.

I seem to recall a similar discussion regarding the first generation SKY. It seems that it sampled once every 3 seconds, though I don’t know how long the sampling window was. I assume that the Tempest sensor does the same thing.

My main testing was with a Sky which has the same anemometer in a different unit now called a Tempest.
My tests while mounted on my car on several occasions were very accurate apart from the error that is now fixed. This is how I mounted it:
https://community.weatherflow.com/t/wind-speed-seems-to-be-reading-low-solved-in-firmware-v103/1510/22
But the wind gradient has a huge effect and I experimented with wind gradient by installing 3 Skys which are essentially the same anemometer as the Tempest in 3 different locations within an urban environment on top of a hill and described my results here:

The two anemometers you have mounted on different fence posts are much different heights from the ground which would cause both a wind gradient difference and a compression difference. The wind gradient is due to the friction from the ground slowing the wind nearest the ground while the compression difference is due to the wind being compressed as it blows over a hill causes less space for the same amount of air to move so it blows faster. Then there is the turbulence which is a mixture of those and any upwind obstacles.
The 3 second updates will produce different results to different time spans.
And an ultrasonic anemometer has no momentum from the cups so it is more responsive.
Whenever I have wished to test the accuracy of my anemometers I have always mounted them on my car and driven using a GPS or adjusted speedo to get an accurate calibration.
When I tested the Sky I was amazed how it was within a kmh of my actual speed, and adjusted for wind when required by driving in multiple directions and in sheltered valleys.
The Tempest at normal battery voltage does multiple calculations to determine a 3 second speed and direction. At the end of the minute the values are averaged and the strongest 3 second value is reported as the gust. The same happens for longer time periods where a graph might display a daily value, it is the fastest 3 second value that day.
cheers Ian :slight_smile:

3 Likes

To add as extra to Ian’s excellent explanation go read the wind speed section in this page where you have also some illustrations.