Topic: Super high resolution wind direction?

[Bushman]

I have this https://www.novotechnik.com/rfc48/ in another application.  Occurred to me that it could make a great wind direction sensor.

[DaleReid]

Of all the sensors I watch, the wind direction varies the most, from second to second, and even with heavy winds, there are many deflections to my wind vane type of direction sensor.

I guess I would question the why of high resolution direction when over a minute period there are 20 degree deflections as the wind passing by varies, not that this isn't a cool encoder.

[Bushman]

Well better a thousand data points to average down than 16, no?

[Mattk]

Well better a thousand data points to average down than 16, no?

The average of an average is still the same average

[Bushman]

Well better a thousand data points to average down than 16, no?

The average of an average is still the same average

No - right now you get only 16 cardinal points.  Why not have 64?  Or binned decimals?  Everybody is always complaining about accuracy and resolution...

[CW2274]

Well better a thousand data points to average down than 16, no?

The average of an average is still the same average

No - right now you get only 16 cardinal points.  Why not have 64?  Or binned decimals?  Everybody is always complaining about accuracy and resolution...

What could possibly be the advantage of knowing the wind direction at that resolution, let alone the inability of any instrument's accuracy to even come close to it? Even in aviation the wind is only issued at 10 degree intervals.

[Mattk]

Yes and 16 cardinal points is simply a subset of 64 cardinal points averaged from the same data records but your comment "better a thousand data points to average down then 16, no?" and the answer is No 

[DaleReid]

I like the sharing of the info about the high resolution encoder, and would really like to play with one, but I'm not sure for what I'd use it right now.

I have an RM Young 26800 sitting on my bench, being fed by a 05103 wind monitor.  I show degrees but will look to see if I can resolve the encoded voltage with a change in settings, but as I was looking at it, and the wind is only a few MPH right now, it does 'bounce' around at least 5 degrees about the most likely readings.  When the wind is really gusty, it can deviate at least the 10 degrees noted, and the cardinal points of the compass which it also displays will change across two readings with some frequency.

I guess when you live where I do and there is major terrain turbulence within a few hundred yards each way, that will happen.  It is fascinating to watch the actual monitors swing around and my wife accuses me of being strange watching the tower once in awhile.

I think smoothing might help, but then am I fiddling with the raw data, rejecting the swings because they conflict with my ideal of a steady breeze (maybe a reality in some open spaces, hard surfaces, or deserts) being very well directed.

Flags tend to do this a lot, too, but their mass is larger and does dampen some things out.  I know wind socks can change from entering a pattern to land and by the time I get on final.  My instructor/friend/plane-saver refers to this as conditions.

[badhairday]

How do you average degrees of angular rotation from a weather vane?  It is something I thought about a while back.  I figured I would have to convert from polar to rectangular and then use statistics to get meaningful data.  My old mind started hurting and I dropped the idea.  I am still curious about it though.

[graculus]

How do you average degrees of angular rotation from a weather vane?  It is something I thought about a while back.  I figured I would have to convert from polar to rectangular and then use statistics to get meaningful data.  My old mind started hurting and I dropped the idea.  I am still curious about it though.

Here you go: Vector vs scalar averaging of wind data.