Power and data transfer over RF coaxial?

I have coaxial line through which is transferred 200W (50V) sinus on 10MHz, length between power to load 3m.
I need after this coaxial cable to transmit data and power the processor.
Power approximately 100mW for MCU
Data transfer, bidirectional, speed 2400 bps
Something like Power LNB and transfer control commands for motor and relay in satelit reciever.
Seeking a model solution to a similar problem

Do you really expect one wire to carry all that? Sounds like a lot to ask.

X10 is a commercial home remote control system. It works by sending digital signals briefly at times when the mains AC waveform is near 0V. You won't be able to use that since you state the AC sinus is 10MHz.

Your 200W power implies a very low impedance circuit (14 ohm). OTOH a microcontroller is relatively low power, high impedance. This reduces your chance of success at combining them on one wire.

I could be wrong.

If its only 3m distance, why does it all have to go down one coax ?

Why not use a mixed multicore cable ?

Multicore coaxial unfortunately does not apply to power supply must be only 1pc N connector.
Yes low w impedance is one of the potential problems.
Unfortunately, there is no simple solution around.

Astrid said:

Something like Power LNB and transfer control commands for motor and relay in satelit reciever.
Seeking a model solution to a similar problem

Click to expand...

You can try a "bias tee" solution, i.e. capacitor for the high frequency path and inductor for the low frequency /DC path.

volker@muehlhaus said:

You can try a "bias tee" solution, i.e. capacitor for the high frequency path and inductor for the low frequency /DC path.

Click to expand...

The task is fully possible. Satellite TV receivers are connected to LNBs in antenna by one IF coaxilal line. It feeds +15...20 V DC to the LNB, 22 kHz oscillator-switching signal, and From LNB down to the receiver, IF line sends 0.95 to 2.15 GHz at a very low level.

You can transmit anything from DC to several GHz over the coaxial cable. The DC or low-frequency signal requires to use the DC Bias Tee on each end. This device passes DC separately from the RF components using low-pass filters. For other signals you must design filters that pass a wanted spectrum (a digital signal) while rejecting unwanted spectrum (RF and DC).
Filters must be designed as band-pass and band-reject types, so that only one part of spectrum is passed to the desired input or output while unwanted part is rejected (while passed to another input/output).
You will need to find specifications of the digital transmission system as to what out-of-band interference is tolerable, then design the bypass filters as well as the DC bias tee.

- - - Updated - - -

As your cable is only 3m long, I think separate cables for each signal can be easier than the set of filters.

Thank you for your answer, especially Jirkovi.
Separate GHz on mW of power is certainly less of a problem than separate MHz hundreds of Watts.
The second problem is the my load has a small DC coupling, too. But we do not resolve it now.
Do you know of example?
One thing is the design of the filter second thing is the overall concept.