Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!hplabs!hp-pcd!hplsla!tomb
From: tomb@hplsla.HP.COM (Tom Bruhns)
Newsgroups: sci.electronics
Subject: Re: hot wire anemometers
Message-ID: <5170063@hplsla.HP.COM>
Date: 19 Dec 89 19:25:13 GMT
References: <5653@orca.wv.tek.com>
Organization: HP Lake Stevens, WA
Lines: 23

henry@utzoo.uucp (Henry Spencer) writes:
>In article <5653@orca.wv.tek.com> philb@ptolemy.TEKTRONIX.COM (Phil Biehl;685-2122;60-850;684-2867;orca) writes:
>>	Anybody know anything about these things? Can one fairly accurately
>>	measure wind direction as well as speed? ...

>You could measure direction by measuring X and Y components of velocity,
>if you can devise some sort of ducting system to get only one component
>flowing past each of two wires.  Actually I guess it would have to be a
>bit more complicated than that, since the hot-wire anemometer won't tell
>you *which way* the air is moving through the duct.  It should be practical,
>although again it would have to be calibrated.

One way to do this is to have a heater in a duct with sensors arranged in
a differential manner, with one sensor up-stream and one down-stream
(or, obviously, the other way around if the direction reverses...)
The heater could, itself, be a hotwire anemometer, if you don't get
enough sensitivity from the adjacent sensors.  I have no idea if anyone
actually makes anything like this; just a thought.  (I envision the
whole thing being two of these duct/heater/sensor arrangements set
at pi/2 radians to eachother--or maybe even three of them set orthogonally
if you want vertical component, too.  Two might well be set in a
vertical round cylinder; three in a sphere.  Or even just have one
heater, in the open, with six sensors around it...)