44 kHz Doppler radar air speed meterr phase shift in wind tunnel

Phase shift no vane:

formatting link

Phase shift with vane:

formatting link

In the movies I start the fan in the wind tunnel, then switch it of again. The transducers are mounted into the tube and receive the reflected signal from the opposite side.

rx tx -----------------------[ ]---[ ]---\ PC FAN / -> air \ / \ vane ----------------------------------- The signal has to go against the wind...

The presence of the vane causes reflections backward that affect the Lissajous xy display It is clear that using phase once can detect even very slow air movements.

Reply to
Jan Panteltje
Loading thread data ...

On a sunny day (Sat, 30 Nov 2013 13:20:20 GMT) it happened Jan Panteltje wrote in :

So, after the experiment I was wondering how to convert, if at all possible, the measured phase angle to air speed. After eating a pizza it would not let me go, so I measured the sizes in the 'wind tunnel'

|--25mm-| -----------------------[ ]-----[ ]----- PC FAN / -> air \ . / | d1 \ a / d2 |60 mm \ ./ | 330 m/s speed of sound in air

44000 periods per second frequency So 1 meter = 133 periods. 1 period = 1 / 133 m = 7.5 mm Verified, see youtube video.

The distance a = 60 mm, sqrt( (12.5)^2 + 60^2) = 61.288 for d1 and d2 The length the sound has to travel is d1 + d2 = 122.58 mm.

That makes 122.58 / 7.5 = 16.35 periods = 16.35 * 360 degrees = 5883.67 degrees. We measure 30 degrees or so phase shift when the wind is 'on'.

30 / 5884 = 0.005099 = .5% .5% of 330 (m/s) = .5 * 3.3 = 1.15 m/s air speed.

Could this mathematical solution, that I just scribbled on a piece of paper after the pizza and a cookie, possibly be right? It is awfully close to my estimated air speed of 1 m/s, but how did I even know that?

And if not, what is th3 right calculation? The above calculation is very easily done in a PIC. The 2 coil setup (circuit diagram) sort of looks like a Foster-Seeley or ratio detector could do the phase measurement with just one extra turn, and 2 added diodes?

It all to easy, all too beautiful, it is either genius or totally wrong. Rip it apart!

BTW the pizza was mushrooms, and the cookie,,, had marmalade in it, :-)

Reply to
Jan Panteltje

Have you considered why the airplane airspeed meters are using the dynamic pressure?

--

Tauno Voipio (CPL(A), CFII)
Reply to
Tauno Voipio

You've gotta wonder whether the air flow in the tube has uniform velocity across the cross-section. Well, obviously it won't. Of course, you can just empirically calibrate phase shift vs velocity. Turbulence will make phase noise.

I just made another Pancake Descending A Staircase

formatting link

big one, served three people and one parrot with leftovers. Scorched pecans and banana slices and blueberries inside. Parts were cooked, parts were burnt, parts were raw batter. People seemed to like it.

Speaking of leftovers, we USians just had Thanksgiving, which involved a 15 pound turkey. We need to go out and buy some onions and celery and make turkey broth out of the various partly-eaten and un-eaten bits. Homemade turkey broth is even better than homemade chicken broth.

Reply to
John Larkin

On a sunny day (Sat, 30 Nov 2013 10:23:58 -0800) it happened John Larkin wrote in :

Yes

That looks a lot like trubulence!

We has 'Sinterklaas' first here, December 5:

formatting link
with lots of marsepein (Marzipan):
formatting link
Already been stuffed with that,,,

It is pretty much what Christmas is in the UK and US, as far as giving present is concerned. although Christmas is beginning to be accepted for that here too.

Reply to
Jan Panteltje

On a sunny day (Sat, 30 Nov 2013 20:17:23 +0200) it happened Tauno Voipio wrote in :

Ultrasonic air speed meters are also in use afaik on airplanes, google gives lot of links, the first one to this patent:

formatting link

Reply to
Jan Panteltje

You did not respond to the question.

--

-TV
Reply to
Tauno Voipio

On a sunny day (Sat, 30 Nov 2013 21:28:57 +0200) it happened Tauno Voipio wrote in :

What question? ;-)

I am not sure what you mean by 'dynamic' pressure. Speed of sound in air depenss on air pressure, I have air pressure available from an SPC01 module. It also has magnetic compass, barometer, and weather prediction. So plz elaborate on 'dynamic'.

Did you actually read the link? It pretty much answers any 'question' you maye have.

Reply to
Jan Panteltje

Mostly because the effort to get anything that is new in a more revolutionary sense certified. It is almost as difficult, or sometimes even more so, than in mdecial devices. There are alternatives and sound is a very old one. Here is a patent from about 35 years ago:

formatting link

The old pitot tube method is prone to icing, clogging, and other failures and has killed many people. For example, 228 people on Air France flight 447. It would be nice to have at least an alternative as backup.

--
Regards, Joerg 

http://www.analogconsultants.com/
Reply to
Joerg

It is the difference between the face-on pressure and static (sideways) pressure. Both come often from the same Pitot tube.

The question is: Is the aerodynamically significant speed more important to the pilot than the true airspeed?

--

-TV
Reply to
Tauno Voipio

This sounds right.

While the answer is plausible, I don't follow the calculation.

Here are some references:

Google for "13. Doppler Radar and MTI" at site:ohio-state.edu to get a very useful set of lecture notes. (I can't get Google to give a useful, uncontaminated URL.)

Foster-Seeley does not measure phase, it measure frequency, and so may have difficulty on very slow windspeeds. Nor are F-S detectors particularly stable with temperature.

The marmelade conflicts with the required beer. That's your problem...

As for the Chinese Remainder Theorem, for radar applications there is a better algorithm. See "Range and Velocity Ambiguity Resolution", G. Trunk and S. Brokett, IEEE National Radar Conference, 1993, pages

146-149. Proposes a one-dimensional clustering algorithm that avoids the noise-sensitivity of the CRT.

Joe Gwinn

Reply to
Joe Gwinn

On a sunny day (Sat, 30 Nov 2013 23:22:54 +0200) it happened Tauno Voipio wrote in :

It is the true airspeed, the plane relative to the air, that gives lift.

You may fly at 160 km/h ground speed with a 160 km/h tail wind, and you will buy the farm really quick. because relative to the air you have zero speed, and zero lift (approx.)

This setup I am testing measures real airspeed. In one direction though..

Reply to
Jan Panteltje

On a sunny day (Sat, 30 Nov 2013 16:34:23 -0500) it happened Joe Gwinn wrote in :

It is really simple, and that brings me to an other important point with this setup, but step by step: Get the distance the wave must travel. Calculate how many wave wavelength fit in that. Multiply by 360 to get the number of degrees in that, Take the measured phase shift in degrees, and see how much percent that is of the total degrees. Multiply that percentage by the speed of sound.

But now that other important thing, in a phase comparator, as soon as the phase becomes > 360 degrees you start all over again. From the above calculation we can see that happens at 360 degrees, or 360 / 5883 = 0.061193, say 6 % .

6 % of 330 (m/s) = 9.80 meter/second = 3600 * 19.8 = 71280 meter per hour, or 71.2 km/h. If we make the distance between the transducers smaller, than there will be less effect of the air speed on the travel time (shorter in the medium), and the max speed that we can do with Doppler only will increase. From this we can calculate (using the above method) the maximum allowed distance to say be able to still measure up to 200 km/h or whatever the plane normally flies plus a little bit. So Doppler for high speed measurements by comparing phase of the carrier only works over short distances for high speed.

Yep, not much in depth.

Got the pdf, thanks, yes that is much more interesting, will take me a while to read it.

Actually it does measure phase... The frequency shift causes a phase shift in the tuned circuits. There are many ways, 2 tuned circuits, xor, logic circuits, sampling, much has been written about it here, I have used most of those circuits at some time or other.

Mm, possible.

Yes, but here we work over really short distances, have to, see above. I do not expect much 'terrain' to interfere. Also the signal is really really big over these short distances, as you can see in the xy display on the scope noise free too. I am not looking for weak signals from far away targets, so that makes things a lot easier. There will be turbulence and motor noise though, more experiments required :-)

Also I keep thinking, at least in this test setup, if I do not need to multiply by sine (angle of beam).. That would not be an issue in a real setup where the transducers are inline with the air flow, and receive the signal directly, not reflected.

H*ll writing all this is more work than building it... But it adds to my documentation. :-)

Reply to
Jan Panteltje

On a sunny day (Sun, 01 Dec 2013 10:12:29 GMT) it happened Jan Panteltje wrote in :

While thinking about it, maybe short distance does not help (less phase error, less periods??????). The only thing that would help is a lower frequency? So for 142 km/h 22 kHz, and 280 km/h 11 kHz ?????

Sorry, it is Sunday morning. :-)

Reply to
Jan Panteltje

across the cross-section.

Not if you've ever heard of the boundary layer. The Reynolds number for Jan 's set-up has to be a lot over 4000 so the flow is turbulent across the bul k of the cross-section of the tube.

formatting link

There is a smooth transition from zero flow at the surface of the tube to t he bulk flow velocity at the top of the boundary layer, but the boundary la yer is relatively thin in things like wind-tunnels, and what you see is "pl ug" flow, rather than the parabolic flow profile you see at Reynolds number s low enough for laminar flow across the diameter of the tube.

hase shift vs velocity.

True, for a rather unusual use of the word "noise". I worked on a "vortex-s hedding" flow meter at Kent Instruments in Luton, back in 1973. You get vor tex shedding from Reynolds numbers above 90 on up to about 100,000 where to see perturbations in the flow path tend to shed vortices rotating in alter nate directions at a rate that is linearly proportional to the bulk flow ve locity - they are called von Karman vortex streets

formatting link

The kind of "phase noise" that that would produce would look pretty regular .If Jan's flow rate is being estimated over a long enough path it will incl ude both sides of the vortex and may average the rotary component to zero, or close to it.

--
Bill Sloman, Sydney
Reply to
Bill Sloman

On a sunny day (Sun, 1 Dec 2013 04:21:45 -0800 (PST)) it happened Bill Sloman wrote in :

If you look at the noise in the scope xy youtube video, then you hardly see any. From a straight line (no wind) to some 'O' (with max) wind. Most 'noise' is granularity from the conversion youtube does to the video.

I have the original slightly higher quality movie AVI available if you want to see it. It is only 700 kB or so.

Reply to
Jan Panteltje

True airspeed does *not* measure the aerodynamically important speed as the air temperature, pressure and density change with altitude.

The airspeed measured by the dynamic pressure (1/2 * rho * v squared) is what counts with the wings and control surfaces. It is tens of precents off the true airspeed at altitude.

--

-TV
Reply to
Tauno Voipio

On a sunny day (Sun, 01 Dec 2013 16:03:23 +0200) it happened Tauno Voipio wrote in :

As I already pointed out I have these variables from a SPC01 module, Again: air pressure, altitude, magnetic compass, and weather prediction. Those are just variables used in the airplane electronics calculations. In case you also want to troll about that: I have attitude and acceleration too from a 6 axis accelerometer module. and most of that software, including an on screen attitude demo was already posted here as C code.

Planes do not fly on their side for example, you forgot to mention that,

Pitot tubes suck (pun not intended) and better solutions are required, and in use already. But 4 sure keep using them if it makes you feel comfortable.

Reply to
Jan Panteltje

That's true - I have experience of pitot icing, despite of heating.

What you're missing in your list is temperature. To get density, which counts in dynamic pressure, you need pressure or altitude and temperature.

--
TV 

PS. Been in the air for some time, there are my pupils flying as 
 Click to see the full signature
Reply to
Tauno Voipio

The small signal speed of sound in air varies as the square root of absolute temperature, it is entirely unaffected by changes in pressure except indirectly as a result of changes in temperature caused by changes in pressure.

While it is true that lift is a function of airspeed and air density (rho), the simple (partial) incompressible fluid stagnation (pitot) pressure equation you presented is significantly in error at typical aircraft speeds.

Compare the incompressible flow and compressible flow equations for stagnation pressure:

formatting link

Note that the two equations are identical at low airspeed and diverge with increasing airspeed.

Reply to
Glen Walpert

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.