In a message dated 10/27/2006 9:07:02 P.M. Eastern Daylight Time,
didier@cox.net writes:

It is  back to the previous method of sending a clean CW signal.

I did not  participate previously, so I have no experience with the test.

If I  understand correctly, they will transmit on all 3 frequencies
simultaneously, but we are supposed to measure only the specific band
they announce at a time?

Since I only have a decent antenna for  80m, can I take advantage of all
3 transmit windows on 80m?

Didier  KO4BB

W4wj@aol.com wrote:

Hi Didier...

The "official" readings are taken for the band that is  announced.

Even though the exciters at W1AW are left on for an extended  period of time
before the test, there is some drift during the
test period.

I would suggest that you listen on 160 and 40 meters during  the callup to
determine wether you have a shot on those bands or  not.

If you are going to submit in the tenths of Hz, I would  suggest that you
only measure during the announced time for that  band.

Good Luck!!

73, Don, W4WJ

W4wj@aol.com wrote:

When measuring a single CW carrier, I would use the 3586A in direct
mode. Its counter has 0.1 Hz resolution, and it has an Ovenair OCXO that
has specs equivalent to the HP 10811 (2e-7/year). Unless I get my GPS
disciplined oscillator working by then, I'll be lucky if I am able to
calibrate it within 0.2 or 0.3 Hz against WWV at 10 or 15 MHz, that will
get me within 0.1Hz or so at 3.5 MHz. That should be a no-brainer.
Greater accuracy is probably not achievable because of atmospheric
variations, the rest will be luck.
I think I will run the 3586 set to WWV with my GPIB logging software for
a few days to see what kind of variations I get during the day and try
and find the best time to calibrate.
Also, we have the cal lab people (company called TMI) coming next week
as they do twice a year to calibrate a bunch of instruments on site
where I work. I am trying to find what kind of frequency standard they
have, but I may bring the 5370 and ask them to check the time base. We
have a bunch of microwave gear that needs to be calibrated to factory
specs, traceable to NIST per MIL-STD, so I would guess they have to have
at least a good rubidium or a GPS locked reference.
If I get my GPS working, I would try John's method. He has done a good
job documenting it. Unfortunately, while I have a bunch of machines
running Linux, they all run in the command mode, so I can't use the
graphical spectrum analysis software he is using :-(

Didier KO4BB

W4wj@aol.com said the following on 10/27/2006 10:06 PM:

Actually, the transmissions seem stable enough over time that I haven't
noticed any real difference in frequency over the 15 minute span.  I've
basically worked with the best chunk of signal I can find during the period.

To be honest, the greatest weak point in the test has been ARRL's
measurement system.  Many of us believe that last year they mis-measured
the 160M signal by about 0.4 Hz, due probably to interference from the
other transmissions as they were measuring the signal off the air into a
counter, and not from a direct tap into the antenna line.

John