That might be true for lossless, linear networks (I honestly don't
know), but it's a simple manner to create a lossy three-port network
that presents 50 ohms at all three ports. For example, a 16.67 ohm
resistor from each port to a common point, or simply a 50 ohm resistor
at each port with no internal connections between ports. A mixer is
nonlinear besides having loss, so there's no good reason to believe that
it conforms to rules that apply to lossless, linear networks.

Roy Lewallen, W7EL

OK1SIP wrote:
"Alfred Lorona" wrote in message ...

Most mixers I've seen are not matched to 50+j0. They are simply terminated
in 50+j0. Is this the best or optimum termination or does it matter?


Hi Alfred,
a mixer is a three-port device. If I remember the theory, a three-port
device can never be matched to the same characteristic impedance on
all three ports. If two ports are matched, the third one is isolated,
so that the network degenerates to a two-port device. Therefore a
mixer with 50+0j ohm RF input cannot have 50+0j ohms neither at the LO
input nor at the output. Some mismatch losses seem to be necessary
with mixers, but they can be easily overcome by increasing the LO
power and more amplification at the output.

BR from Ivan