increase counter resolution with two oscillator

[bug13]

Hi guys

I am reading this datasheet here (DS00VQ100), it's a laser range finder chip (FPGA). From page 12 from page 22, it (sort of) explains how it measures time-of-flight of a laser pulse. What I am very interested in is: how does it get down to picosecond while using only two 12.8MHz oscillators?

By my calculation, the minimum resolution is 156.25uS, but datasheet says it can get down to ~60pS.

Thanks guys!!

PS: sorry can't attache the datasheet, see link here:
http://forum.allaboutcircuits.com/attachments/ds00vq100-laser-range-finder-controller-data-sheet_rev_01-pdf.85019/

PS2: I meant to say page 22 above

[Rerouter]

page 21, it wants 2 voltage controlled oscillators, and says it mixes to downconvert the signal, i'm not that far into RF, but its common with spectrum analysers to use something like an IF to measure frequencies higher than your equiptment is capable of,

I can only imagine it works off the beat frequency, however the fact they have 2 independant voltage controlled oscillators, atleast in my head sounds like they could be using phase skewing of the 2 clocks to get picosecond delays (one clock starts, the other stops, and you phase shift the second to give you a very narrow delay)

[bug13]

OK, I think your level is a lot higher than me, so there are a few terms I need to google before I can understand more. To start, what is IF in your replay

"to use something like an IF to measure frequencies higher than your equiptment is capable of"

[Rerouter]

http://www.analog.com/library/analogDialogue/archives/43-09/EDCh%204%20rf%20if.pdf

[T3sl4co1l]

One example is doing it analog: feed the pulse delay into an integrator; the final voltage, between when the pulse was sent until it arrives, will be proportional (i.e., not quantized) to the duration.  It will be noisy (sensitive to risetime, amplitude and threshold), which makes this not the greatest for a ranged (wide dynamic range, think radio) application, but there are ways around that (lock-in, filtering).

Another is doing it by vernier, where you start counting one clock at the pulse start, and start counting the other at the pulse arrival.  Then continue counting until both clocks coincide, and calculate the quotient of them.

Yet another is to do it by phase, so you have an oscillator that times pulses going out, and a 90 degree complement of it.  Use mixers to determine the I/Q components, phase-detector-ing the return signal.  If everything is square pulses, phase is linear with whichever quadrant it's closest to, so some ADC action (and filtering to smooth out the noise) can unwravel the answer.

If you have sinusoidal waves, you get sinusoidal phase response, so you need to do some trig.

Tim

[awallin]

I am reading this datasheet here (DS00VQ100), it's a laser range finder chip (FPGA). From page 12 from page 22, it (sort of) explains how it measures time-of-flight of a laser pulse. What I am very interested in is: how does it get down to picosecond while using only two 12.8MHz oscillators?

By my calculation, the minimum resolution is 156.25uS, but datasheet says it can get down to ~60pS.

From a quick read of the datasheet I would guess it uses a two-stage approach: first a direct time-of-flight measurement on the incoming laser-pulse using a simple time-interval-counter running at 12.8 MHz. A +/-1 count error is 156 nanoseconds (or half of that - anyway not microseconds like above!). For this you would ideally want very short laser pulses and a wide-bandwidth receiver so the received pulse is also a narrow spike. (contrast to phase measurement where you filter the received signal to a sine-wave).

This time-stamp is then further enhanced using a phase measurement. This requires that the rangefinder continuously sends out laser-pulses so that the received signal is also continuous and can be filtered to a sine-wave. It would be hard to directly measure a phase-shift in the range of 0...78 ns (corresponding to 0...2pi at 12.8MHz), so the signal is instead mixed down.
When mixing the phase-information in units of radians(angle) is preserved, but the frequency is lowered. So if we have two oscillators at 12.8+0.5df MHz (e.g. pulse-train sent/received) and 12.8-0.5df MHz ('local oscillator', used for mixing) then the beat-note at df (the difference frequency) will have the same phase infromation 0...2pi as the original received signal, but stretched out in time by a factor 12.8MHz/df. So that same phase infromation 0...2pi at 12.8MHz which was hard to measure is now at df (say 1 kHz) which maps the 0...2pi angle into a much larger time-interval of 0...1 ms - which can be easily measured using the direct-counting (at 12.8MHz for example)

Combining the coarse time-stamp with the phase-measurement into a precise time-stamp could be a bit tricky when the phase is close to 0 or 2pi - some other systems I've looked at use a quadrature approach with two local oscillators, sine-waves 90-degrees shifted to each other, and we choose smartly between the two so that we never go close to 0 or 2pi phase.

HTH,
AW

[bug13]

Thanks guys

After reading all the replies, and a lot of googling, I am now have a basic understanding of how it works. Thanks a lot!!

[krenzo]

You might find this page helpful: http://www.ko4bb.com/~bruce/TDC.html

[bug13]

This is actually a very good summary, thanks a lot!!