[time-nuts] WWVB Response

[David I. Emery]

On Thu, Sep 27, 2012 at 11:16:45AM -0700, J. Forster wrote:
> The bandwidth is secondary. The time-domain response of the filter is the
> issue.

	Seems clear that you need a reasonably wide preselection filter
to avoid this problem, maybe as someone suggests several hundred or more
hz to really kill serious phase trajectory problems.

	BUT this does not imply anything but that the dynamic range and
noise properties of the signal processing that deals with that bandwidth
must be able to handle all the energy there.   And yes, with a wider
bandwidth there WILL be more noise and crud at the output of the RF
preselection.   But not so much as to make a quadrature multipliers
(mixers) for a Costas loop unimplementable.

	The actual tracking or detection bandwidth can - however- be
made arbitrarily small using a standard - known for 70 years - Costas
loop - and is a function of the LPF characteristics in the Q arm...
and of course the drift and random noise (eg Adev) characteristics
of the VCO (and signal and propagation).

	Since both the local loop VCO and the WWVB carrier are assumed
to be from reasonably high quality stable sources and propagation is not
expected to be wildly variable this bandwidth in locked mode can be made
almost arbitrarily small... hundredths of a  Hz or less...

	And THAT bandwidth is the true noise bandwidth.

	The only real issue in such an approach is locking the loop in
the first place... it will be necessary to start with the VCO close to
the carrier or widen the loop bandwidth initially in search mode.   In
the digital or data domain of course FFTs help here a lot.


> 
> -John
> 
> =============


-- 
  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."