[time-nuts] Lucent CDMA Reference Question
David I. Emery
die at dieconsulting.com
Thu Feb 1 22:42:20 EST 2007
On Thu, Feb 01, 2007 at 08:57:51PM -0600, Bill Hawkins wrote:
> Have we settled the question of whether or not the RB can
> be dominated (no, that's not it) disciplined by an external
> 1 PPS or 10 MHz signal?
Settled is vastly too strong a word in my case, but I am
currently pretty positive that the Rb is disciplined (naughty Rb) by the
hardware provided which supplies EFC voltage to its C field pin. I
believe this comes from a 16 bit DAC on the mezzanine board.
AFAIK the jury is still out as to whether there is a way of
making the existing Lucent firmware perform this operation without the
cooperation of firmware running on an XO. Experiments so far suggest
the RB firmware pays attention to the binary messages from the Oncore
GPS on the XO and may not be willing to accept that there IS a GPS
supplying 1 PPS to lock to unless that stuff is there and sending out
the right messages.
It does seem at least possible the firmware supports other
reference inputs and it certainly seems rather likely it could be
patched or hacked to accept GPS input from a bare Oncore GPS board -
maybe even one mounted on its unstuffed header.
So probably the best answer is that surely the hardware can
without modification support disciplining of the Rb, and maybe the
existing software can be made to work without an actual XO present and
maybe not. And maybe a patched version of the software would do this
even if the existing version won't.
> If it can, I'd load up on Rb units and discipline them with
> a modern GPS receiver - or maybe replace the OXCO in Z3801s.
In theory a Z3801 should be able to control a LPRO-101 instead
of its 10811 with very minor hardware tweaks, but one supposes there are
ALL kinds of tweaks in the Z3801 firmware that are 10811 specific (loop
parameters, thermal compensation, oven monitoring etc) that would be
moderately to highly wrong for a LPRO-101 as 10 MHz source. And of
course a LPRO-101 generates a lot more heat and would require other
thermal treatment and its related mechanical issues.
I imagine that most of the actual disciplining PLL constants and
other Rb specific stuff in the Lucent RB firmware is relatively isolated
from the higher level part that pays attention to the system
configuration and worries about talking to the companion XO and GPS
about status and configuration. Perhaps someone with time and patience
COULD patch the RB firmware to ignore whatever it tests to validate the
XO state or GPS configuration allowing the 1 PPS tracking PLL to still
work but not need good GPS status or XO messages. And maybe there is
some configuration process to make this happen anyway as a test mode or
for alternative system architectures with 1 PPS from some other source.
And for those who like such approaches, it might also to be
possible to program a PIC or other cheapo one chip micro with suitable
serial capability to pretend to be the GPS and XO and fool completely
unmodified RB firmware into thinking they were there and fine. It
stands to reason that the real synchronizing input is the 1 PPS and that
all the rest of the dialog is related to determining whether it is good
or not and whether the RB should turn on its outputs or not based on
higher level system state. And the PIC code involved might not be
overwhelming - given the option of monitoring a working system and
determining what is looked for and when, it might merely involving
canning the dialog with a good XO and GPS as a script rather than any
intricate decomposition of the actual messages and reformatting them.
Dave Emery N1PRE, 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