[time-nuts] Re: timing lab, remote control

[ed breya]

Well, it wouldn't be turnkey, but it would be almost trivial to do it 
the old-fashioned way with direct signalling. It depends on the ethernet 
connections. If there's only one cable/port/etc that is already 
committed, then not so good, but if you have a Cat-5 or whatever point 
to point hard line available, or you can readily add one, then you've 
got 8 wires per cable available that you can work in countless ways. 
Simple DIY logic at each end can give as many bits as needed, 
bidirectional action, and low noise/interference. Optocouplers and such 
can give isolation and protections as needed too. You could even send 
low speed analog signals - pure, clean, accurate, and simple.

If there is one and only one ethernet line available and committed, then 
you likely still have four unused wires available (unless it's a POE or 
special type setup that needs them), that can be used for low speed 
signalling - not as good or clean, but better than nothing.

The best part of this old-school kind of scheme is that you would have 
absolute full control and knowledge of what's in there and what it's 
doing at all times, and the flexibility to easily modify it as needed.

BTW if this sort of function is desired but no cables are available, 
there are RF transceivers with simple parallel bit-for-bit protocols, 
including bidirectional. I got a bunch of these for planned remote 
control and monitoring projects at the farm. They are 915 MHz ISM, with 
about a mile range outdoors.

Ed