[time-nuts] Digital TV signals

David I. Emery die at dieconsulting.com
Wed Apr 16 00:18:37 EDT 2008

On Tue, Apr 15, 2008 at 04:50:29PM -0700, Stanley Reynolds wrote:
> From: david brown <davidarbrown at bigpond.com>
> >From a complete amateur, is there any useful timing info to be gained 
> from the newer format of digital tv transmission(Australia) to replace 
> that available from current analogue transmissions.? My recently 
> repaired TV derived frequency standard looks to  be becoming obsolete! 
> David

	> GPS based standard would be the most common now.

	FWIW my understanding is that in the US few local broadcast
facilities make any very serious attempt to lock their signal timing to
high accuracy standards or GPSDOs.   Much digital TV broadcast equipment
is designed so this COULD be done, but there is no current reason to
spend the money and engineering effort to actually do it -  more likely
than not signal timing on the output signal is determined by at most a
medium grade OXCO calibrated every once in while and possibly just a
TXCO not much better than barely meeting the FCC spec.

	And the same often applies to carrier frequency accuracy, though
the old analog NTSC broadcast standard does require some degree of
accuracy here to help reduce the impact of co-channel interference.

	As many list members probably know, many many many years ago in
the 70s in the USA NIST did do some work with time and frequency
distribution via TV networks when they were shipped from place to place
on very stable terrestrial AT&T analog FM microwave systems.   Timing
and chroma subcarrier frequency was derived from early rubidium
standards at the network operations centers and when a broadcast station
was running a network show it was very likely the chroma subcarrier was
accurate to rubidium level accuracy as was the video timing related to

	But time marched on and the networks started distributing video
via geo comsats rather than terrestrial microwave and because geosats
move around significantly in their box in the sky in the course of a day
doppler shifts from satellite motion destroyed the utility of the stable
chroma frequency - shifts in the order of several tenths of a hertz in
the chroma frequency could easily happen - and they'd be different at
different satellite receiver locations so no two retransmitted satellite
signals would be exactly the same.

	And to add to this, advances in RAM technology made it more and
more usual for local broadcast plants to use  multiple stages of
digitizing the analog signal, reading  it into a memory and maybe
processing it there in the digital domain in some way and then reading
it out of the memory later on with output timing based on a local clock.
Often by default the local clock was not very good or based on house
sync so any time or frequency accuracy in the input signal was
completely wiped out (or reduced to the accuracy of the timing from the
local house sync generator, not the network rubidium standard).

	And since the transition to almost entirely digital technology
in broadcast plants in the 90s and onwards to today there is no longer
any analog network signal with chroma subcarriers as a time reference
anywhere anymore anyway and more likely than not the signal timing
coming out of the IRD that decodes the network digital program feed is
locked to local house sync rather than the satellite signal.   And the
timing of the ATSC modulator at the transmitter may be locked to its own
little oscillator, not even house sync.

	So it is not clear that TV signals are good time or frequency
references any more - though there is little doubt that if there was
some purpose to doing so both time and frequency could be rather closely
locked to a GPSDO at the transmitter - it is just with the entire system
designed to deal with small time, frequency and rate errors at many of
its interfaces most stations haven't seen fit to implement high accuracy
frequency or time control on their actual transmitted signal.   And of
course that completely begs the question about the majority of TV
viewers who get their signal from cable or satellite sources with all
sorts of timing issues of their own and no particularly tight timing
relationship to the original broadcast signal.

  Dave Emery N1PRE/AE, die at dieconsulting.com  DIE Consulting, Weston, Mass 02493
"An empty zombie mind with a forlorn barely readable weatherbeaten
'For Rent' sign still vainly flapping outside on the weed encrusted pole - in 
celebration of what could have been, but wasn't and is not to be now either."

More information about the time-nuts mailing list