Phase comparison

Thanks! Now, I'm having some trouble understanding how that works =]

In the phase comparison case I imagine there's a standing wave of sorts, probably that's not physically accurate, but I kind of see why phase difference tells you where you are within a 100m wave if f = 3 mhz.

What you are proposing on the other hand is doppler radar ?

Yes, pretty similar (it isn't doppler effect from a moving target, but it's the same kind of frequency-shift detection). LFM (linear frequency modulation) and a 'chirp' cycle are the keywords.

This online book might be useful...

Thanks for the link! Downloading it now.

I apologize if I'm getting this wrong...

So, if I were to measure a distance of 1 meter, frequency shift would be

Here is a little basic information on laser distance measurement.

Bob

Thanks Bob!

ght...

The factor of 2* 20 kHz/sec makes the frequency shift about 10**-4 Hz; if you make a 1 second chirp, this means you only get a full cycle (suitable for curve-fitting) at distances of 10 km. There's some room for improvement, like using a higher base frequency and 20 MHz/sec modulation, and a Fourier analysis of the lowest sampled frequencies with 8 bits signal/noise can get you to 1 meter that way. A PC sound card input can do the digitization.

A 1 second chirp also has an upper limit, something between synchronous orbit and Luna. Apollo 11 left a nice retroreflector out there if you want to try for distance...

distance...

Hehe =] Well, it seems (to me) that the system you describe is more suited to distances of the order of kilometers and resolutions in meters. On the other hand, I'm guessing that laser measuring tapes use phase comparison implemented by knowledgeable people (not me!!) can achieve milimeter resolution when measuring 100m ranges.

typo : laser measuring tapes use phase comparison implemented by knowledgeable people (not me!!) and* can achieve milimeter resolution when measuring 100m ranges.

stance...

In retrospect, I'd think there's another level of mixing involved, to generate two swept frequencies, like

F(t) transmit frequency, 10.0006 MHz + t * 10 MHz/sec F2(t) mix frequency, 10.0000 MHz + t * 10 MHz/sec so you mix the receive frequency down by the F2, and that beat frequency is near 600 Hz at zero distance (this makes it possible to chirp long enough to get many cycles of the beat). You can use a delay line to derive the F2(t) from F(t); F2(t) =3D F(t - .00006 seconds), so it'd be a good match for those acoustic delay lines used for TV receivers; I think the PAL delay lines are 64 micro seconds.

Then mix down F2(t) against received R(t), sample/digitize/ calculate some Fourier components, and curve-fit the frequency peak. Dozens of few-millisecond chirps can be sent, and the received signals summed, until signal/noise is good.