driving a signal onto powerlines

[kripacharya]

I need to drive a signal of around 50v p-pk and 200KHz onto the power line (220Vac/50Hz).

The generator is a variable duty cycle TTL level, where two non-overlapping pulse trains are available.

One topology I have used successfully has complementary MOSFETs which drive a series LC and couple to the line using a 1:1 isolation trafo. This works fine for upto around 20v.

Beyond this I find that the MOSFETs I use (SMD's) heat up too much.

Any suggestions right off the bat ? Circuits etc I will post if someone shows interest.

 

[chuckey]

I would think that the input impedance of a power line is very capacitive, This would need tuning out with an inductor leaving a low resistance load for your signalling kit. So the problem is can your amplifier drive 50V into a resistance of say 5 ohms?
Does your isolation transformer work OK at 200KHZ? I think the first thing to do is to get the typical impedance of the mains supply, else you are in the position of your amplifier at 17 V RMS delivering the power at the same rate as 230V, i.e 1 KW load = amp delivers 17/230 X 1 KW = 140W !!
Frank

 

[jiripolivka]

I do not know the AC line impedance, but you apparently use the AC line as a long antenna radiation your 200 kHz signal over India everywhere your AC lines go.
Is it legal in India to use 200 kHz to interfere other services? I think also harmonics will be radiated. Really a bad idea.
Sine 1990, there were "digital power line communicators" sold and installed around the world. Such devices utilize frequencies between 5 and 30 MHz, and have caused a lot of problems, also to radio amateurs.
Many ideas are technically possible but these are immoral and only realised to generate money through very poor engineering.

 

[kripacharya]

I would think that the input impedance of a power line is very capacitive, This would need tuning out with an inductor leaving a low resistance load for your signalling kit. So the problem is can your amplifier drive 50V into a resistance of say 5 ohms?
Does your isolation transformer work OK at 200KHZ? I think the first thing to do is to get the typical impedance of the mains supply, else you are in the position of your amplifier at 17 V RMS delivering the power at the same rate as 230V, i.e 1 KW load = amp delivers 17/230 X 1 KW = 140W !!
Frank

Didn't quite understand your 140w calculation, but yes - the trafo works fine at 200Khz. At least at the 20Vp-p levels.
And yes, my testing is usually for a variable line impedance between 1 - 20 ohms.

The amp should work - as long as the switching takes place fast enough for negligible power dissipation in the MOSFETs. That is my chief area of concern.

- - - Updated - - -

I do not know the AC line impedance, but you apparently use the AC line as a long antenna radiation your 200 kHz signal over India everywhere your AC lines go.
Is it legal in India to use 200 kHz to interfere other services? I think also harmonics will be radiated. Really a bad idea.
Sine 1990, there were "digital power line communicators" sold and installed around the world. Such devices utilize frequencies between 5 and 30 MHz, and have caused a lot of problems, also to radio amateurs.
Many ideas are technically possible but these are immoral and only realised to generate money through very poor engineering.

Thanks for the reminder on legality & morality issues.

Having set those aside, any ideas on actual circuit topologies which don't result in releasing the magic smoke ?

 

[kripacharya]

here is the circuit i currently use. For higher voltage the MOSFET pair needs to be changed, and of course the schottkys.