[time-nuts] New topics (was Re: He is a Time-Nut Troublemaker....)

Lux, James P james.p.lux at jpl.nasa.gov
Tue Dec 23 20:57:27 UTC 2008


> [mailto:time-nuts-bounces at febo.com] On Behalf Of John Ackermann N8UR
> Sent: Tuesday, December 23, 2008 12:29 PM
> To: Discussion of precise time and frequency measurement
> Subject: Re: [time-nuts] New topics (was Re: He is a Time-Nut
> Troublemaker....)
>
> Hal Murray said the following on 12/23/2008 01:58 PM:
> >> 2.  Several measurement techniques require a given phase
> relationship
> >> (e.g., quadrature) between DUT and reference.  For HF frequencies
> >> (ie,
> >> 5  or 10 MHz) is there a *practical* phase shifter design covering
> >> 180+  degrees that doesn't involve switching various
> lengths of coax
> >> in and  out of the line?
> >
> > PLL up by a factor of N, use that to drive a DDS, then
> filter.  Maybe
> > a pair of DDSes will get better tracking.
> >
> > For each possible phase offset, you need N slots in the
> table.  (N/4
> > with more work)
>
> If doing phase noise or short term stability measurements,
> wouldn't the noise of the DDS impact the results?
>

I would assume that one could have enough bits in the DDS phase accumulator AND in the DAC (and potentially use various error feedback/feedforward schemes) to get the noise low enough.  Vankka's papers (and book, for that matter) talk about "how low you can go" with this sort of thing.

Maybe not, though.. There's always going to be some noise added to that of the reference you're feeding in. Might be small, but it's not non-zero.  For instance, a 16 bit DAC adds quantization noise at roughly -96dBFS. (there's actualy a correction of a dB or two, because quantization noise is uniformly distributed over an interval, not gaussian). It's spread evenly, so the power spectral density is low.  If the clock's running at 200MHz, the noise is spread over 100 MHz, so -96-80 dBFS/Hz or -176 dBFS/Hz (how's that for an odd unit?)

That's pretty good, in any case.

(for more real numbers.. AD9954 is a 14bit DAC, 32 bit phase accumulator, etc.  And says -120dB/Hz phase noise on the datasheet cover.. -132dBc/hz at 1kHz away from a 40MHz outpt with 400MHz clkin. The 9910 gives -125dBc/Hz at 1kHz away from a 400MHz carrier w/1GHz in. Divide that down to 10 MHz and you'll pick up, what, about 32 dB more.. -157dBc/Hz

The datasheet says residual phase noise at 20MHz out, 1kHz away, is -152 dBc/Hz, presumably with 1GHz in.  Looking at Figure 15, it looks like it's about -135 at 10 Hz away, -145 at 100Hz, down to a floor around -165dBc/Hz around 50 kHz away.



More information about the time-nuts mailing list