How can I multiplex a digital signal on a DC line?

[geoffd123]

How can I do the following, or does anyone know of a similar circuit?

I have a remote Solar Panel with a Tracking controller 300 feet away. It has a 24 volt power line running to the house. I can talk to the Tracker with RS232, but it is akward to run a second cable.

I want to multiplex the RS232 data on the 24 volt line, perhaps using DTMF tones or similar. Alternatively I can redesign the comms to use something other than RS232, but it will still need to be multiplexed on the 24 volt line.

The data is bi-directional. Commands from the house, Sun, voltage and watts data from the Tracker.

Any thoughts?

Cheers
Geoff

 

[ScottyUK]

Have a look at the HART protocol. It is used in industrial control systems to superimpose digital data onto a 4-20mA current loop. It is a relatively 'open' protocol, but where you would buy a chipset in low quantities I have no idea. There is a lot of information on the net about the HART protocol. Next stop Google!



----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[OperaHouse]

What is so bad about running one extra wire? You can do bi-directional modified RS-232 on a single dedicated conductor. There were a couple companies offering computer communications on your house 120V outlets. These might be easily modified for 24V. Think Thompson and SEVU.com (if they are still in business) had them.

 

[melone]

Can you use wireless transmission?

 

[automatic2]

Your power line may not have the best characteristics for data transmission but it certainly can be done. You simply add your signal voltage on top of your power voltage. Slew rates, capacitance, and all that stuff will ultimately limit your ability to perform such a task.

 

[Skysoldier]

Just had a similar design requirement last month, ableit
with a requirement of much greater distance. If you want
to do it quickly, and cost is not sensitive check out
Aerocomm.com, The make small, 900Mhz wireless modules that
require no further liscensing, and can be implemented in
less than a few hours. You can buy everything you need for
under 150 bucks, and they work fantastic. Check them
out on Mouser.com, and notice that developement kits have
just been reduced from 900 to only 199 bucks! If you want
to borrow a development kit for a few days, I could loan
you one if you pay the shipping.

 

[geoffd123]

Thanks, I will look into a few of these suggestions.

The actual panels are not in line of sight of the house (behind a couple of large rocks) so I am not sure if wireless would work. It also took a lot of effort tree and rock climbing to get the original wire in.

Cheers
Geoff

 

[dhwilliams]

Some baby alarms/intercoms use the mains to connect the 2 ends of an audio link. I think these modulate a high frequency with the audio signal and the high frequency is injected onto (and taken off) the mains wiring via capacitors. I am sure I have seen a similar principle used over low voltage d.c. but I can't find any details. Try searching for "power line communications"

 

[BrianG]

HI Geoff, I think your own suggestion of DTMF tone signalling may be the most useful. Trying to maintain a minimum + and - 3v signal swing required by RS232 devices will likely be difficult.

The big problem with superimposing signalling on top of power is the relatively low impedance of the 24V power supply and the load. If you can fit a fairly large series inductor at each end of the cable this will isolate these from the cable yet still provide a d.c. path without too much voltage drop. Connect the signalling system at each end after the inductor, i.e. where it joins the cable. It should then be relatively simple to transmit and receive using a few hundred milivolts of DTMF tone from standard chips.

 

[OperaHouse]

Have you tried asking on the solar/wind board of

http://www.fieldlines.com

Maybe someone there has tried it.

 

[maged541946]

Hi,Please before spend some money, check the following...
RS232 has several signlas, are they all used, the least is sent path, and receive path. no hand-shaking (4 more lines)
next will be speed, how many bauds(bit per second) you transmit.
Now the neck of the bottle is the data is not at a steady state rate as it is Asynchronous, that imposes a modulation.

A frequency for modulation as rule of thumb is at least 10 times the maximum rate, so if a standered 9600 b/s, you need at least 100khz carrier, now DTMF is out of question. Scale that 100Khz as you scale the 9600 (4800 needs at least 50Khz and so on).I guess this application does not need high rate and hand shaking...

A carrier of 100-200Khz can be handeled with Tone decoder chip like LM567C with ease to generate the RS232 back again and may be LM555 to generate, but higher frequencies, will need different ways.

Now to superimpose on 24V, those lower frequencies are better handeled with any transmisiion line as the problem begins when the physical wire length gets comparable to 1/10 of wave length transmitted, all you need is 2 coils (Thick wire and ferrite core) at the ends to path DC to load and stop the carrier from seeing the low load impedence, and 4 cacitors to couple the 2 transmitters and 2 receivers, also remember to use different frequency for each path

 

[VEBill]

Something else to think about...

Make sure that the average power consumption of the entire solar power control system, including the proposed added communications system, does not consume a significant chunk of the average power output from the solar cells.

In other words, don't select a communications system that sits there at the remote end drawing significant current continuously.

 

[pebe]

You could use RS232 in a half duplex arrangement. Receivers and transmitters at each end all share the line. The line sits at logic ‘1’.

To prevent collisions, the house could send a ‘0’ polling pulse to the tracker once every second. If the tracker had data to send it would be sent its ‘0’ start pulse, say, 5mS after the end of the polling pulse.

When the house wants to send data it first checks that the line has been at ‘1’ for at least 10mS after the end of the polling pulse, to prevent data collision. If so, it sends its data.

 

[benta]

What you're looking for is actually "Powerline communications".
Whether this happens at 100, 230 or 24 VDC is unimportant.
I suggest you look to Echelon, who has a monolithic solution here with protocol and everything implemented.
See:

http://www.echelon.com/products/oem/transceivers/smartpl/plsmarttransFAQ.htm

Regards.