On Sat, 12 Jun 2004 22:22:46 -0700, Jim Weir wrote:

And when the relay(s), an electromechanical device with a relatively low MTBF
goes south and dumps 10 watts into the other receiver, your analysis would
be...?

I guess we're having a critical design review...

Since, for this exercise, I'm doing the design, I get to choose the
relay. I'm going to choose a Teledyne RF300-series relay. Here are
the specs:

http://www.teledynerelays.com/pdf/el...rf300rf303.pdf

So, what's the isolation at around 140 MHz? It's not 20 dB, it's more
like 40 dB. In addition, since there are two internal switches, they
can be wired in series. But, for the sake of argument, let's say that
we're limited to 40 dB of isolation. That's 0.32 volts instead of the
3.2 volts that you quoted. That's at or below the conduction voltage
of the protection diodes, assuming that they're placed at the input of
the receiver.

So, given that information, what are the failure modes of a relay that
we're talking about? It's not fair to say that all relay failures
would result in a fried receiver. The one that would cause damage is
where the relay contact (the one to the input of the receiver hybrid)
closes after it was commanded open. What are the chances of that?
Pretty low, I would say, for a relay that is rated for millions of
cycles. If you're really worried about this failure mode, put two
independent relays in series.

Lastly, I would not select a Wilkinson divider for the receiver. I'd
use a transformer hybrid.

Rob-
------------------------------------------------------------------
Robert Cherney e-mail: rcherney(at)comcast(dot)net