[CQ-Contest] sub-harmonics again!

[K1TTT]

I went through one round of this witch hunt a couple years ago and figured
it was a rare problem.  But now its back again and even uglier than before.
And I just had a random chat with someone on 2m up here who has gone through
a similar problem, so I'm starting to think it may be more common than I
first thought.

What I am seeing is when I transmit on 20m something generates a subharmonic
at 1/2 and sometimes now 1/4 and occasionally 1/8 of the 20m frequency.
These are usually very clean signals, though sometimes at high power or very
low power they do become intermittent and raspy. There is one other odd
thing this does, transmitting on 20m not only generates the 40m subharmonic,
but it then mixes with it and makes a 15m 3/2's harmonic... that is actually
where this chase started for me both times since that one is broader and
happens during peak run time on 20m!

When I went through this before I spent weeks trying to chase it out of the
shack.  I swapped amps, radios, filters, tested into dummy loads, just about
everything before I was finally convinced it had to be outside and was not
just a radio or amp problem.  Then I had to play around with coax and diodes
until I finally figured out how such a thing could be done without active
components.  The key is a diode connected to something near a 1/2 wave
length on the lower frequency band.  Basically when the diode conducts on
one peak a pulse goes down the 1/2 wave line, reflects back, and arrives
just in time to prevent the diode from conducting  on the next cycle.  An
example with 2 diodes is at:
http://www.k1ttt.net/subharmonics/subharmonic_files/subharmonic_frames.htm
I also did this with just a single diode... note the frequency used was
determined by the random length of coax I grabbed.  The diode action gives
it a distinctive characteristic which is a sharp turnon threshold.  As the
tx power is increased there won't be any subharmonic until a critical
voltage is reached when the diode turns on, then it quickly generates a
strong subharmonic.


What I finally found the last time was 2 pieces of jacketed hardline that
were not directly grounded to the tower at the top.  Adding a hose clamp to
pull the connector shell against the tower leg killed that one, but not
until almost 2 years after I started hunting it.  

That one usually needed well over 100w, usually near 1kw to get started,
this one I can do reliably with 25-50w on the top 20m antenna when pointing
it south.  So far I have cleaned up and grounded more stuff in the last
couple weeks than I have done in years here.  You can read all about it if
you want in the last couple weeks on my maintenance blog.
http://wiki.k1ttt.net/2008%20Maintenance%20and%20Upgrade%20Blog.ashx

I would be interested in hearing from others who have noticed this type of
interference to get some idea of how common it really is.  I have a feeling
its more common that I once thought, just that you don't notice it unless
you are running 2 radios on the right bands.  

I would also REALLY be interested in anyone who has some systematic way to
track it down.  I have been up and down towers more these last couple weeks
than I have in years and still have no idea where it really may be.  With
cables and antennas all around it seems to be everywhere, just about any
cable seems to have some hint of it. 



David Robbins K1TTT
e-mail: mailto:k1ttt at arrl.net
web: http://www.k1ttt.net
AR-Cluster node: 145.69MHz or telnet://dxc.k1ttt.net