[time-nuts] On some pitfalls of the dual mixer time differencemethod of horology

Tom Van Baak tvb at leapsecond.com
Mon Oct 2 13:53:29 EDT 2006

Ulrich Bangert wrote:
> I built a DMTD and made measurements with it on the few good oscillators
> that I own. While my experiments have shown that the principle works


I am happy to hear you got the project working. I know
last time we talked you had your doubts...

> First Pitfall of DMTD: Transfer oscillator effects do NOT cancel out
> completely
> the two down mixed signals at the SAME time! This is due to the fact
> that at the same time the transfer oscillator is in the same state
> concerning both channels.

Yes, this is one factor that determines the noise floor
of a DMTD implementation. Do the math to decide if
the level of noise is within your design spec. Note that
the better the transfer oscillator, the less effect this
"pitfall" has on your error budget. If you decide to use
a 1 Hz beat, then it would seem that the 1 s ADEV is
the parameter that dictates which xfer osc to buy.

Note also that, based on my limited experience, most
commercial mixer implementations use a much faster
beat note: 10, 100 Hz, even 1 kHz. A faster beat note
may help your concern #1 above, and #2 below.

> Second Pitfall of DMTD: Decreasing slope to noise ratio counteracts the
> magnifying effect of down mixing
> Had we a noise-free signal available then there were no problem at all
> because the noise-free signal crosses the zero line at a sharp defined
> point in time. However, noise-free signals are an idealization not given
> with real-world signals. There is always a certain amount of noise,
> sometimes more, sometimes less, a fact that documents itself in the well
> known signal to noise figure. 

Don't lose sight that you are building a DMTD -- whose
sole purpose is to measure noise. Thus one would
expect a large amount of AM and PM noise at the

The noise you see is some sort of rms sum of ref osc
noise, transfer osc noise, other internal instrumentation
noise, and of course, DUT noise. So the better your
design, construction, and measured choice ($) of parts
the more the DUT noise comes through compared to
all the other noises in the system.

> Third Pitfall of DMTD: Phase corruption due to mutual crosstalk
> If you buy a good coaxial cable, this may have a shielding effectiveness
> of 80 dB at radio frequencies. If you spend some bucks more you can get
> a shielding effectiveness of 90 dB. 100 dB shielding is top and only
> possible with double shielding and I do not remember to have seen a
> shielding effectiveness been advertised better than 110 dB. So, 100 dB

This sounds like a normal quality-of-engineering issue.
I don't have design experience but I know from taking
apart lots of gear that high-end, low noise systems
seem to use hardline instead of coax. They also place
PCBs inside their own shielded or solid brass boxes.

Perhaps you could do this too if your noise floor isn't
as low as you need.

Someone else on the list can tell you what the typical
isolation numbers are for this type of construction vs.
a cheap single PCB and coax design.

> channel? Perhaps equipment like the TSC 5110 uses a very high isolation
> switch to keep the second signal completely out of the box while it

No switches; the ZCD rate is somewhere between 100
Hz and 1 kHz. Have a look at the TSC data sheets and
design papers.


More information about the time-nuts mailing list