[NLRS] Fwd: Fw: [MNDX] WWV Reference

[Gerald]

On Sun, 2006-01-29 at 22:51 -0600, Ford Peterson wrote:
> > Consensus has been for decades that if you check WWV on HF the same time
> > every quiet propagation day for a month you might get a result of a part
> > in 10^7, 1 Hz per 10 MHz. And that may be better than your counter
> > crystal. My house standards are a couple orders of magnitude more stable
> > so I use WWVB on 60 KHz to check them and then I think I see propagation
> > variations in a short time that I use for checking.
> > 
> > These days the handiest way to precision frequency is using GPS. The 1
> > pps outputs can be 4 or 5 orders of magnitude more precise than WWV as
> > received at 10 MHz. There is a commercial box or two, HP-2801A seems to
> > be popular though touchy. There was a circuit in the most recent Eastern
> > VHF Conference Proceedings using the 10 KHz output from a Rockwell-
> > Collins GPS board and not a whole lot of circuitry to stabilize at 10
> > MHz crystal oscillator and Luis Cuipedo (sp?) has boards for doing that
> > with the 1 Hz output of most any GPS.
> > 
> > -- 
> > 73, Jerry, K0CQ
> > Dr. Gerald N. Johnson, electrical engineer
> > All content copyright, Dr. Gerald N. Johnson
> 
> Jerry,
> 
> Thanks for the input.  The HP2810A you are referring to used to sell surplus for about $250 in QST.  Now they are up over $500 and occupy a good chunk of bench top space.  The GPS systems you are making reference to are quite the ticket, and well beyond the tolerance of the equipment being calibrated--to say nothing about expensive.  This little 10MHz Xtal filter and amp costs about $20 (less with a good junque box), can be made in a single weekend with ease, is just a wee bit bigger than a pack of smokes, and is powered from a generic 12v wall wart @100ma.  Best of all, it is completely impervious to temperature changes.  It's as accurate as WWV with a few phase inversions thrown in.
> 
> I talked with one of the lab techs at the TIS at the ARRL about this issue.  We talked about the phase inversions and what errors were thought to be introduced.  The consensus of opinion was that accuracy to a few hertz should not be a problem, even at 10MHz.  Obviously, if you have equipment that needs better accuracy, then spending the money and time to make that happen is well worth the investment.  On the other hand, I just need to calibrate my frequency counter, which at 0.1s updates is only accurate to 10 Hz.  When I set the counter to 1 Hz timebase, it flips within a few hertz of 10,000,000 once per second.  But heck, that beats the 200Hz error I had when I first powered it up to check it out.
> 
> Hip shot accuracy--no problems.  Precision instrument--not a chance.
> 
> Ford-N0FP
> ford at cmgate.com
> 
> 
Quite true that the counter oscillator won't justify a better reference
unless the counter has provisions for an external oscillator. My counter
does and uses my external reference that drifts on the order of a couple
parts in 10^9 per week. I've tried to adjust other counter oscillators
and found their drift too much for my WWVB test technique.

Another point is that the number of digits displayed on the counter
doesn't necessarily have anything to do with the accuracy of the value
displayed. As W0PFP used to characterize it, "Measure with micrometer,
mark with chalk, cut with ax."

Getting a part in 10^7 precision from WWV at 10 MHz is supposed to
require averaging the measurement daily with quiet propagation
(rejecting those days with solar storms or odd propagation, like long
path) for about 30 days. I doubt the oscillator in your counter has the
stability for that 30 days of averaging.

If you have a GPS with serial output, Luis's boards are only two or
three times the price of your receiver and less subject to modulation
sidebands or signals sneaking in from the adjacent broadcast bands.


-- 
73, Jerry, K0CQ
Dr. Gerald N. Johnson, electrical engineer
All content copyright, Dr. Gerald N. Johnson