[time-nuts] Any reason not to use one power amplifier and splitter for distribution amplifier?

[Dr. David Kirkby (Kirkby Microwave Ltd)]

On 7 Jan 2015 01:24, "Charles Steinmetz" <csteinmetz at yandex.com> wrote:
>
> Dave wrote:
>
>> At 50 MHz, the loss from the common port is 12.8 dB, and the isolation
>> between two ports sets of ports is either 38 or 48 dB
>
>
> To get the worst-case output-to-output isolation, you need to test two
output ports that are electrically adjacent (i.e., that share the same last
2:1 splitter, assuming that the 1:16 is a hierarchy of 1:2 splitters --
which is the case with the multi-output splitters I'm familiar with).  You
may already have found an electrically-adjacent pair (ports 7 and 8), but
to be absolutely sure, you would need to repeat the test from one output to
each of the 15 others (or find a full internal connection diagram, which
does not seem to be on the datasheet).

IIRC the odd port numbers are on one row, and the evens on the other.  I
found the highest isolation was between a port at the bottom and one at the
top,  which I assume is two splitters.

> I wouldn't bother retesting with out-of-band signals, but when you test
at 10MHz it is something to think about.

Yes, but I was aware of this, and that's why I got two different isolation
figures.

Another issue I thought of later, is that the reflection coefficient of all
the open ports will be the same, as there's virtually no phase difference
between the output terminals.  If it was used as a distribution amplifier,
that would no longer be the case.  Potentially that could produce issues.

Anyway,  as I said it was just a quick test with suboptimal equipment used.
But later I will perform a more thorough test with more suitable test
equipment.

Dave