Sunday, 25 September 2016

SWR Anomaly! (revised)

Yesterday, I reported on the problems of corroded coax, and a successful fix.

Well, sort of!  Whilst connected to a standard SWR meter, or when using the rig's own internal SWR monitor, all was well in the world of the antenna match.

But, on connecting my trusty SARK-110 analyser to the antenna, I just couldn't get a stable reading.  One minute it was showing a nice curve with low SWR, the next, it was showing a steady SWR of 5 or more.  There was no such instability at the old-fashioned SWR meter or rig.  The key to the problem was that the curve would often shown the hints of a good curve, but with horrendous noise spikes all over it.

This took me a while to sort out, because the strange MCX connector on the SARK never fills me (or some others) with confidence; I do wish it was something more robust-looking, even though it does seem quite well secured to the circuit board.

The problem got more and more unclear as days went by.  One moment it would report a nice curve in the middle of the band and the next, some senseless curve that took off towards 26MHz.

Mightily cheesed-off this lunchtime, I disconnected the delta and connected an alternative feed line to it.  This resulted in the SARK reporting an open circuit!  I was certain both feed lines were perfect.  This had to mean something to do with the SARK was at fault.

Eventually, with a multimeter, I checked the continuity of all the connectors and cable hitched-up to the analyser, to discover the real source of the problem: an intermittent break in the SMA patch lead supplied with the SARK-110.  In case I had really lost track of what I was doing by now, I hooked the whole analyser system up to a dummy load and you can see the result below:

The impedance (red) going somewhere odd, whilst the SWR is up in the sky!


A new patch lead is on its way from RS Components...






No comments: