[time-nuts] Racal Dana 9478 frequency distribution system with option 04B high stability oven (5x10^-10/day aging rate).

Bob kb8tq kb8tq at n1k.org
Sun Nov 4 08:20:26 EST 2018


One would *guess* that the PLL bandwidth in the 9578 is wider than what you see in a typical 
GPSDO. As long as they don’t both line up at the same point, noise and phase peaking should 
not be a big deal. Depending on how clean your GPSDO is, the 9578 might clean things up a bit. 

For timing / high precision, putting another PLL into things is going to add some delay and possibly
some temperature dependence. Since you are running normal test gear, neither of those issues 
should be significant enough to impact anything.  In this environment 75 ohms vs 50 ohms is also 
not a big deal, but it will have slightly more impact, fix it if you can, don’t sweat it if you can’t. 


> On Nov 4, 2018, at 5:44 AM, Dr. David Kirkby <drkirkby at kirkbymicrowave.co.uk> wrote:
> I want a distribution amplifier to distribute 10 MHz from an HP GPS
> receiver to various bits of test kit - spectrum analyzer, VNA, signal
> generators, HP 5370B time interval counters, signal generators etc.
> Currently I am using 75 ohm video amplifiers, but I suspect that is sub
> optimal. I also don't like the form factor of this amplifier - I would
> prefer a 19" rack mount unit.
> Does anyone know how well the Racial 9478 performs? If my understanding is
> correct  this is not simply an amplifier, but has its own internal high
> stability reference oscillator, which then feeds 9 independent outputs, all
> of which have their own amplifier. The internal OCXO will phase lock that
> to an external reference.
> Fundamentally, it does not seem a great idea to me to lock one OCXO to GPS
> then lock a second OCXO to the first  OCXO.  But maybe I am mistaken.
> I have the chance to buy one for £150 (GBP) but are a bit worried the
> performance.
> Any comments?
> Dave.
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.

More information about the time-nuts mailing list