couldn’t these steel pipes resonate with high winds going by ?
I don’t think so. Even if they do, it’s very shallow and 40 km/h (that’s when false rain starts) gusts shall not be causing this.
The origin of my report is the actual erroneous wind reading as reported by the OP and shown in my graph as well.
Hi @petercek,
The direction change can be a faulty Sky and can also be something like water in the gap. I do have a faulty Sky that virtually every day goes incorrect for a short period. Weatherflow replaced it due to determining it is a hardware issue. I continued testing it to see if it would be of any use to me but it just keeps doing it. Here is its chart and it is currently inside away from any wind: https://smartweather.weatherflow.com/station/4555/graph/44661/wind/3 .
It took me a while to prove the problem up on the pole but bringing it inside makes it clear to see.
If you sat it inside for a while if it does the wild wind thing then you have evidence that it is faulty and contact support.
As for the vibrations false rain issue I did several tests to determine that movement is not a problem. The Sky can be moved and waved around while holding the pole as much as you can move it and it will not record false rain. BUT while perfectly still inside if I tap the pole quickly repeatedly with a plastic ball point pen it records heavy rain. This test taught me that the vibrations of concern are like sound travelling along the pipe. So my vibration reducing mounts were designed to allow movement but stop sounds travelling through the mount.
I see on that fence a loosely tied cable that could move in the wind and be swinging tapping the rail like a rope on a flag pole or a rope down a yacht mast.
I suggest placing rubber under the pole and around all the mounting points to reduce sound vibrations. And if the Sky has any movement within the mount or solar accessory tape it solid. And if you can see anything that might move against the railing try to make it secure.
best of luck
cheers Ian 
2 Likes
Thank you, @iladyman , very useful explanation. One of the issues is I have it mounted in a location which I can not access every day or anytime I want. I decided to leave it there as is.
What makes me upset is this is the second Sky I own that does the wild wind thing. I would guess they’re both from initial production run even if this Sky in question is “new”. I’m from EU and maybe WF produced our versions just once.
Anyway, I decided not to open another support ticket.
Off-topic but referring to your other explanation: moving to my other setup, I do have lots of false rain also there. A month ago I started thinking that sound “vibration” could be triggering false rain. I planned to put rubber or other material around the pole but am not sure how will rubber behave on longterm. Do you have a solution or a photo to share? Maybe not in this topic not to mislead the thread.
Thanks
Hi @petercek,
This is where I posted my experiments and findings regarding the false rain due to vibrations. This particular post shows the rubber mount broke down under our severe Australian sun.
https://community.tempest.earth/t/rain-sensor-and-vibrations/1345/170
There are other ideas on the thread.
cheers Ian 
After one year in operation my Sky started showing this issue again. Conditions are very reproducible; after a longer period of stronger wind (few days about 50-70 km/h), the Sky resets and swaps the direction and reports unreal wind speed. I have a DVP2 anemometer just beside the Sky to prove there are no “micro bursts”, etc.
One plot shows average windspeed, the other shows wind gusts.
This is my third Sky I think showing exactly the same behaviour. I conclude this as a design and quality issue.
We haven’t had rain for days but just blue skies. Due to strong wind I also reject the theory of droplets or other subjects in the hole.
This is the station link:
Reported rain is false due to pole vibration.
Not seeking for technical support anymore but just want to share my experience with others.
Seems like someone found something similar early on when they were testing one mounted on their car. I wonder if the issue with yours is similar. Maybe sustained wind over a certain speed overflows something in the firmware or circuitry.
That said, however, your SKY is showing a wind sensor failure.
Maybe @WFsupport will send you a replacement. I recommend you open a support ticket so at least they are aware of the problem.
This actually is a replacement already ( I think the 2nd). The whole logistics procedure is difficult (pay the customs, etc). They are/were aware of this but somehow ignored or try to avoid confirming the real cause.
Yes, the sensor status turns to “Failed” occasionally.
I know @WFstaff is willing to help but it seems the hardware is not capable to handle a bit more “serious” weather on a longer term.
1 Like
The wind has calmed down and readings are back to normal.
I would also suspect if power supply may be causing this. Batteries are OK (one or two months old), working in power-save mode. Maybe strong wind readings consume more power which causes the Sky to reboot. After boot up it maybe starts some “initial calibration” and goes crazy since it reads strong gusts?
Other thing to think about is that strong wind gusts move the batteries a little and this causes intermittent power loss; after boot up, the “initial calibration” doesn’t really expect such strong wind and goes a little crazy.
Some diagnostics and advise from @WFstaff would be welcome. I imagine this scenario shall be relatively easy to reproduce in the wind tunnel and remote power supply (wires to the external battery through a switch).
I’d really appreciate a response from @WFstaff about my assumption.
Edit: Nov.20, 2020
Another event with exactly the same procedure. Strong wind, then Sky starts rebooting and switching direction an increasing wind speed.
I think it’s a design issue. I have had 3 Skys with exactly the same behaviour.
As the thread originator, sorry for not keeping up with events.
Long story short, original Sky unit eventually failed WIND_FAILED UV_FAILED. Installed replacement on deer fence strainer post about 9 feet above ground. Got erroneous rain and wind readings at certain combos of strong wind and heavy rain (possible vibrations giving odd readings?) Moved unit to new spot on house gable but eventually failed with WIND_FAILED UV_FAILED. If left unpowered for a while, the unit will start working again but will fail after a few hours.
Support have offered a new Tempest at a discount but obviously I have my doubts about the longevity of these units. They don’t seem to like strong winds and heavy rain!
Have two faulty units now. Can they be opened (non-destructively) in order to ascertain the mode of failure?
Cheers!
if there is no warranty on, sure you can open them. Perhaps some water got inside. you never know.
The Tempest units are sealed against the elements much better than the SKY units. My field test units are still working fine. I assume that you’ll have a new warranty on the Tempest so there is that as well.
2 Likes
From the 3 Sky units, 2 are still installed and show erroneous wind direction and speed when the winds are strong and last a longer time. I don’t even check the rain since it is far from real. It is interesting, however, that I never had problems with “rain on plate”. Never got wrong wind readings even under persistent heavy rain. The issue only appeared during persistent strong winds (70+ km/h).
I liked the original idea and design but the device is not reliable so I’m back to classics.
I opened the third unit but saw no obvious damage or a possible point of failure. I’ve removed the central part (battery holder) and middle part where the wind sensors are mounted. With the top cover, I got a nice waterproof chassis which I used for a Boltek’s antenna
I got a discounted offer for the Tempest as well but I just don’t trust the device. For mild weather measurements, I could use a cheap weather station. I want to have a hardware capable to work 100% under severe conditions.
1 Like
Yup, really liked the idea of no moving parts but the device has not proved to be reliable enough for me. Gone back to the old Davis VP2 so we will see how long that lasts before the wind takes parts of it away. Will have a go at opening the Sky devices to investigate further.