Transformer Communication

Thread Starter

cgha20@yahoo.com

Joined Oct 21, 2009
82
So I have two 1:1 transformers with 30 turns at primary and secondary. I would like to couple an ac signal to Transformer 1 (by switching a 5V line) and have the secondary to add this signal to a 20VDC battery line. Then I want the second transformer to connect to this dc battery down the line to receive this signal and remove the dc and just accept the ac so I can demodulate and use as data. How do I do this? I need help
 

wayneh

Joined Sep 9, 2010
17,496
At the receiving end, it could be as simple as using a capacitor to couple the AC out of the signal and block all the DC. Or use a transformer that draws little DC current but still extracts the AC signal. It's hard to say much more without more details.
 

crutschow

Joined Mar 14, 2008
34,285
You could AC couple the signal from the transformer into and out the line with DC blocking caps at the outputs, as wayneh noted.
For that to work you would need AC blocking inductors (chokes) at the output of the supply and the input to the load, so the AC is not absorbed by the low impedance of the battery and load filter caps.

You could also put the transformer outputs in series with the line, but then the transformers would have to handle the 2A DC load current without saturating, which would generally require a large (perhaps gapped) transformer.
 

nsaspook

Joined Aug 27, 2009
13,086
How long is the line? It would seem to be simpler just to add a separate circuit for the AC signal if it's physically and electrically possible.
 

crutschow

Joined Mar 14, 2008
34,285
Here's the concept using inductors and capacitors for the coupling.
The value of the inductors and capacitors depend upon the signal frequency.
Note that, with this scheme you don't really need the transformers.

upload_2016-5-9_8-5-0.png
 

crutschow

Joined Mar 14, 2008
34,285
You are seeing the resonant ringing of the capacitance with the inductance at around 5MHz
You need a much larger inductance to move it to a lower frequency so it isn't a problem.
What is the signal frequency range you need?
Is the frequency or signal modulated?
 
Last edited:

Thread Starter

cgha20@yahoo.com

Joined Oct 21, 2009
82
2 frequencies, a zero represented by 100khz and 1 by 400khz. I plan to use transistor as switch as input to T1. More inductance on inductors on 24v line or transformer?
 

crutschow

Joined Mar 14, 2008
34,285
2 frequencies, a zero represented by 100khz and 1 by 400khz. I plan to use transistor as switch as input to T1. More inductance on inductors on 24v line or transformer?
Inductor's inductance.

Here's an LTspice simulation of some values that see to work fairly well.
I didn't include the transformer since I didn't know it's properties and if it's properly specified it should not have a significant effect the circuit operation.
The diode on the output is for DC restore of the signal through the capacitors so that the negative side of the pulse is near 0V.
I added a filter capacitance on the output since most loads have such a capacitor.

upload_2016-5-10_1-38-38.png
 

Attachments

MrAl

Joined Jun 17, 2014
11,396
Hello there,

Usually when this kind of thing is done the injection signal used is small so it doesnt bother the normal operation anyway. At the receiving end it's almost the same as a radio receiver with input amplifier and demodulator.
Although AM is easier, for best signal reproduction use FM as normal load transients can generate signals that look like AM too. For binary systems the bit time has to be long enough also, and error correction methods are desirable (Hamming, etc). Redundancy when possible. Depending on the importance of the signal, a duplex system might be needed with acknowledgement of the received signal.
 

crutschow

Joined Mar 14, 2008
34,285
As long as the signal is isolated from the supply and load with the inductors, then I don't see that signal amplitude will have any effect on the circuit (unless the powered load is sensitive to the signal frequencies).
You can make the signal as large as you need to get good signal to noise ratio.
I don't see 30-40 ft as a particular problem as long as the return (common or ground) line is part of the same cable
 

MrAl

Joined Jun 17, 2014
11,396
Thanks alot! I went ahead and added the two transformers and looks good. Running to store over lunch to buy parts and test it out! How much smaller than DC should the ac be? should i step down further at T1 and step up at t2? Line length can be max of 30-40 ft. Would this design still apply?
That's cute :)

National Semi back in the 1980's published a schematic for a power line transmission circuit, which sends signals over the 50 or 60Hz power line. It's still around on the web somewhere i think.
 

Thread Starter

cgha20@yahoo.com

Joined Oct 21, 2009
82
so it looks like 100mH inductor is hard to find at required current and size limitations. Any ideas. I reran simulation with 100uH and signal looks way different. More like sign wave.
 

crutschow

Joined Mar 14, 2008
34,285
so it looks like 100mH inductor is hard to find at required current and size limitations. Any ideas. I reran simulation with 100uH and signal looks way different. More like sign wave.
Of course it looks different with an inductor that's a thousand times smaller. :rolleyes:
So how large an inductor can you find that meets your current and size requirements ?
 
Last edited:
Top