[time-nuts] Some More questions

James Maynard james.h.maynard at usa.net
Wed Jan 18 09:28:53 EST 2006

Neon John wrote:

> Next question.  I have a 70s vintage HP ovenized 10 mhz frequency
> standard like the one that they sold as an OEM item and as used in
> some counters.  It's in the oversized metal relay can.  I'd have to
> open the case to get the model number but I do recall the -01 option,
> high stability.  I have this unit mounted in a case along with a power
> supply and have used it for 20+ years as my frequency standard.  It's
> been powered more or less all that time and as determined by beating
> it against the 10 mhz WWV signal, has settled down and is more stable
> I have the patience to watch the scope.
> If I divided this down to 1 PPS, would it be a better timing reference
> than the GPS receiver?  Is there a better method of calibrating this
> oscillator than beating it against WWV, assuming the usual compliment
> of electronic lab equipment but no special timekeeping instruments?

Yes, there is a better way than comparing your 10 MHz oscillator against 
WWV. I have a 5 MHz HP 10811A ovenized crystal oscillator that that I 
acquired back in the 1970s.

My first method of calibrating that oscillator was to count it down to 
get a 1 PPS second that I used to trigger an oscilloscope, feeding WWV 
audio to the vertical input of the 'scope.  My little cobbled-together 
circuit would let me vary the phase of the 1 Hz trigger signal. But I 
found that the WWV audio time tick would vary over the course of a day 
by about a millisecond or so. At attributed that variation to the 
varying path length that the signal from WWV travelled to reach me by 
sky wave. (As the ionosphere rose and fell during the day, the path 
length would vary; hence the variation in arrival time of about a 
millisecond over the course of a 24-hour period.)

So, instead of comparing my HP 10811A oscillator against the sky wave 
signal from WWV at 10 MHz or 15 MHz, I compared it against the ground 
wave signal from my local Loran-C chain. I counted down the the 5 MHz 
crystal oscillator output to provide me with an output at the local 
Loran chain's group repetition interval (GRI) -- again, with a provision 
to vary, in 1 microsecond steps, just when during the GRI the pulse 
would occur.  I used the one-pulse-per-GRI to trigger the 'scope, and 
fed the  loran RF directly from an amplified loop antenna (from Palomar 
Engineers of Escondido, California) to the vertical input of the 'scope.

Since the Loran signal at 100 kHz is received on ground wave, the phase 
of the leading edge of the Loran pulse did not vary during the day. And 
since I was watching the 100 kHz carrier rather from Loran Station 
George (that's George, Washington, USA) rather than the 1 kHz time tick 
pulse from WWV (5 cycles of 1 kHz = one WWV "tick"), the frequency 
calibrarion could proceed 100 times faster that when using the WWV ticks.

Try Loran-C -- you'll like it!

James Maynard
Salem, Oregon, USA

More information about the time-nuts mailing list