[TowerTalk] To get a truly balanced antenna feed

[Ian White]

Paul N9AC wrote:


>>"That is why, out of the many thousands of amateurs who claim to be
>using "balanced feedline", there are few or none who can actually
*prove*
>it. All the rest are really just expressing a hope or a wish which will
almost
>never be granted."

>Positively depressing, Ian.  If we can achieve reasonable, albeit not
perfect
>line balance, does it really matter?  

"Does it matter?" is a separate question from the ones about measurement
technique that follow later.

One answer is based on treating CM current as a potential interference
problem. Many types of interference have a threshold, and if the
imbalance isn't causing a problem to anyone then the answer is "No, it
doesn't matter at all". This applies particularly to RFI from a
transmitted signal. But common-mode RFI to *received* signals is much a
more subtle problem and many amateurs don't even realise they have it -
much less that it could be cured. In that latter case, even if nobody is
aware that the line imbalance exists, it still matters a lot.


Moving on to "How might you measure CM current or line balance?" I'm
afraid the two-ammeters method is only the roughest of checks. Because
it only measures the overall magnitudes of the currents in the two legs,
it is only capable of detecting gross levels of unbalance. A more
accurate measurement must also take account of the differing phases in
the two legs.

The simplest method would indeed be to pass the line through a large
toroid, and extract a sample of CM signal on a terminated secondary
winding.  

However, it is much more informative to measure the suppression of the
CM signal relative to the wanted differential signal, and this requires
something more elaborate.  I had been thinking of something similar to
the method that Jim Lux described, using two identical current
transformers. 

I was thinking of something superficially resembling a SWR bridge where
the 'vector math' is done by connecting the signals from the current
transformers either in-phase or in opposition. From this one could
measure both the wanted differential current and the unwanted CM
current. However, the same could be achieved as Jim describes, by
measuring the two signals separately using a vector network analyser or
a vector voltmeter, and either adding or subtracting the two results
using vector math. 

Unfortunately, the current transformers and associated networks would
need to be carefully 'floated' independent of ground... hence the
proliferation of transformers, chokes and baluns in methods of this
kind.

But thanks also to Rick N6RK, for pointing out that ground-referenced
measurements from a 4-port VNA could also be used to extract the desired
results, given sufficient hardware and heavy-duty math.

Either way, open-wire line is much harder to measure than it looks.


73 from Ian GM3SEK