Topband: QSB on 160

[Ken Brown]

Hi Dennis,

I was probably the operator who QSOed you this weekend at KH6LC. I 
operated roughly 9 hours of the contest, and Lloyd KH6LC filled in for 
me when I had to take breaks. I remember a couple of runs of JAs, one of 
them followed by a W1, so that was probably our QSO.

> What struck me about the propagation this night was the QSB.  Peaks 
> were 2-8
> minutes apart, depending on the station; the peak cycle also didn't 
> seem to
> matter how near or far the station was from me.  It was also strange that
> the peak signal of a station that I was trying to work did not 
> coincide with
> what I gathered was my signal peak back to him; KH6LC was as strong as 
> S6, but
> he was S1 when he finally copied my grid.  

I would like to propose another explanation for the number of repeats 
necessary for copying a grid square. I do not doubt what you say about 
the period of the QSB being two to eight minutes. I think there may be 
more than one QSB cycle occurring though.

Suppose that it takes you five seconds to send "FN42". And suppose it 
takes me five seconds to send "AGN" or "FN??". Now suppose there is a 
QSB up/down cycle that has a period of eleven seconds. If at the 
beginning of our attempts to exchange grid squares, you are transmitting 
during the "null" of the QSB, and I am transmitting during the peaks, 
then if our exchange period is ten seconds and the QSB cycle period is 
eleven seconds, it will take three to five of our exchange cycles before 
you are sending during the peaks. In this case it will seem like there 
is "one way" propagation, you hearing me Q5 while I keep asking for 
repeats of your grid square, for about the first twenty seconds of our 
attempt to pass the information. Then as our exchange cycle changes 
phase with relation to the QSB cycle, it will start to seem like there 
is one way propagation in the other direction.

Now suppose that our exchange cycle is still ten seconds, and the QSB 
cycle is also ten seconds. In this case we would stay in sync with the 
QSB, we would never get an easy copy in both directions, and you might 
think I am running incredible power and have a poor receive capability.

Of course this is an oversimplification, and it is unlikely that our 
transmission exchange cycle could remain constant for a long time. 
Although with canned messages from the function keys of a PC running a 
logging program, who knows? This does point out that one possible way to 
get the exchange done is to alter the exchange cycle period, to get out 
of sync with the QSB cycle. You could send "FN42 FN42 FN42" instead of 
just "FN42", to get us out of sync with the QSB cycle. (You probably did 
something like that)

I have always been skeptical of claims of "one way skip". There are 
several reasons that it may "seem" that there is one way skip, when 
there is not. Most times when one way skip is claimed, it is probably 
due to higher man made noise levels at one end of the path. Differences 
in transmitter power and antenna efficiencies are also factors. This 
notion of QSB cycle and exchange cycle sync is another.

Having said all of the above, I have worked on many VHF, UHF and 
microwave systems that employ circulators or isolators, which 
intentionally DO have more transmission loss in one direction than in 
the other. Perhaps that can happen in the ionosphere at MF and HF too.

Ken N6KB