View Full Version : Doppler Frequency shift speed measurement?
guilartec
Dec 09, 2004, 03:24 PM
Has Anyone tried to build a doppler's sound frequency shift speed detector?
I read in this forum some threads where people estimate model speed using a ".wav" file extracted form a video, and a program to visualize frecuency shift; then they use doppler's formula to make the calculations. I was wondering if we can take this idea to make an electronic circuit to make this calculations and bring speed readings. What do you think about this idea guys?
jeffs555
Dec 09, 2004, 04:29 PM
It should be possible, but I don't know how practical it would be. I haven't seen the threads you are talking about, but to calculate speed from the sound of the plane, you would either need to know the exact frequency of the sound source, or else you would have to take samples with the plane going toward you and samples with it going away from you, and assume that the speed and source frequency didn't change. Because the frequency shift is proportional to the speed of sound, a 1% change in frequency is equal to around 6mph.
Zlatko
Dec 09, 2004, 05:07 PM
Hi,
I have seen a PC program that does that for you.
You connect a mike to the sound input of your notebook and then fly the model past you at "constant" speed.
As jeffs555 said it takes frequency readings coming towards you and then away and uses them to calculate the speed.
The other advantage of this "as advertised" is that you can use your notetaker to record the sound and then play it back at home to your PC to work out the speed.
Unfortunately I can't remember what the program was called ....
Just Google for it I suppose.
Doing one in PIC would be nice but it wouldn't be sensitive enough and wouldn't work too well if you are at the local club :)
If you need something to measure speed why dont you look at "Bushnell Speedster" , its available on ebay for $130 and gives you a reading of +/- 1mph . Its the handheld type radar gun, similar to the ones used by the friendly police world wide ;) .
Cheers
hul
Dec 09, 2004, 05:31 PM
it's called Waveosscope and can be downloaded form here:
http://www.sprut.de/electronic/soft/scope.htm
the software is in German but it shouldn't be too difficult to use even if you don't know that language. A screen shot is attached.
You can use a laptop or a video camera to record the sound.
The stand alone solution would have to keep track of one of those traces during a fly by. It would have to make sure it doesn't use one trace coming towards you and another going away.
Hans
hul
Dec 09, 2004, 05:38 PM
sorry, wrong picture...
PilotSmith
Jan 25, 2005, 04:12 PM
Very cool info Hul. I'm going to take my laptop next time I get a chance to fly to record the plane and see what speeds I get. The program was not too hard to figure out in German. I've been curious what speed my planes fly at. It seems like the more I fly the slower they seem to get!
it's called Waveosscope and can be downloaded form here:
http://www.sprut.de/electronic/soft/scope.htm
the software is in German but it shouldn't be too difficult to use even if you don't know that language. A screen shot is attached.
You can use a laptop or a video camera to record the sound.
The stand alone solution would have to keep track of one of those traces during a fly by. It would have to make sure it doesn't use one trace coming towards you and another going away.
Hans
HELModels
Jan 25, 2005, 05:06 PM
What if you had a sound source in a fixed spot on the ground and a small sensor in the plane to record the shift? Fly past the sound source, device senses the sound and computes the shift to display speeds after landing?
The sensor and source could be matched for optimum and to filter motor sound.
you know...snap your fingers turn on a light.
hwhall
Jan 25, 2005, 05:36 PM
Remember too, that to measure the airplane speed most accurately, the plane ought to pass fairly close to the sensor. Since Doppler indicates the relative speed along a line between source & sensor, if the plane isn't flying straight toward & away from the sensor the maximum shift won't be observed.
--Wayne
HELModels
Jan 25, 2005, 07:18 PM
O.K. Then 2 different sound sources and sensor that can pick up both sources and filter out motor sound. kinda like a finish line - you just have to cross it and doesnt matter exactly where.
mmormota
Jan 25, 2005, 07:43 PM
A Java version of the program could be practical for field measurements. Running on a Palm or Java mobilphone, it is rather comfortable. These devices are already equipped with the sound recorder, most people already have them etc.
Put it to the grass, stay far from it to be safe, fly above the phone, then get the result immediately. ;)
vintage1
Jan 25, 2005, 08:28 PM
..and remember to do a run in eac direction, to cancel out windspeed effects, and don't cheat by diving into the level run :D
mmormota
Jan 25, 2005, 08:37 PM
..and remember to do a run in eac direction, to cancel out windspeed effects, and don't cheat by diving into the level run :D
I did exactly the opposite to get good results from my scratch-built foam edf... :D It is a almost slowflyer, but with good wind and diving as hell I got 85 km/h. :D :D :D
vintage1
Jan 25, 2005, 08:58 PM
Obviously destined for a career in politics....:D
vBulletin® Copyright ©2000-2009, Jelsoft Enterprises Ltd.