With new, inexpensive VNAs, I think it would be interesting to start
looking at redesigns of the typical 4 square control box.
At a simple level, keep the same phase shift network (Collins hybrid),
but add relays to allow measuring S11 of each element and Sx1 mutual
coupling. Sure, a problem that causes mismatch to dump power into the
dummy load will catch some faults and if you have remote power metering
on the load (a temp sensor might work). But a full 4 port measurement
may be more useful.
I can see two ways to do this.. one is to keep the VNA in the shack, and
run a second coax line back for the other port. The other is to remote
the VNA (which typically has a serial over USB interface).
There are some tricky aspects - you probably can't use a SPDT on each
element to switch the element to the hybrid or the VNA, because the
isolation across a single relay isn't good enough (cheap VNAs have a max
of maybe 0dBm - I'd want to see 70-80 dB isolation to avoid destroying
the VNA). And I've just now started to think about how to arrange the
relays (it would be nice to be able to detect failures in the dump load
and the hybrid as well.
You might be able to just provide the ability to short/open each element
and make all your measurements through the hybrid. There's already a
bunch of relays to switch the directions.
TowerTalk mailing list