Another thought on this subject....


I once saw a scheme for this work and work well. It provided
bi-dirrectional signaling. Even better, it was done over an internal AC
line and also provided 1,500 volts of isolation from both ends by using a
1:1 transformer built with 4 turns each tap on a ferrite bead.

What the circuit did was provide a 10 volt pulse wave (in the megahertz
range I dimly remember... to keep the transformer that small) to provide
power. This was caught by a diode and cap. The signals were all 5 volt
pulses so they could 'slide in" under the rectfier voltage and never
forward conduct it. Signals were interped by sending an ID pulse coded
train (this design had a driver side and 4 listeners), which send either
data or a command to an individual listener. An addressed listener then
could send back a pulse train before the next power pulse occured.



At 10:22 AM 6/3/99 -0700, you wrote:
>I've toyed with this idea and I don't know if it's practical.
>Neither RF nor power transmission is my area of expertise, so
>forgive me if I am asking stupid questions.
>
>Suppose I have a situation where I want to send power and control
>signals to a device. Suppose too that I want to do this all with
>2 wires (power and return). Is it at all feasible to send the data
>on top of the power line, and have it separated at the destination?
>
>Kinda like X10 but this is DC.
>
>Any other ideas along these lines?
>
>Michael
>
>