Serial comms on VCC line

[MegaADY]

hm ... can you please be more specific ?

[Kamikaze47]

I dont think you can even use the modulation technique in this instance because its the slave device he wants sending the data (i.e. the device that doesnt have a power supply).

You can use the mondulation technique if you were transmitting data from the master to the slave, but not slave to master.

If you already have 2 wires between the 2 devices, whats so hard about running a 3rd wire for data?

[MegaADY]

You got the point ! That's why I thought it might not work ! Although, someone did it, I don't know exactly how... Forget the 3rd wire...I just can not use more than 2 wires

[Melanie]

No, it's irrelevant which end is supplying the power and which end is transmitting the data. Quickly, for example as I'm under pressure at the moment...

Slave is a 16F628 (building block of the modern world!). HPWM an output at say 40kHz... hardware cross-connect that back into one input of one of the on-board comparators. Take Hardware Serial out of USART or Software Serial out of any other pin. Hardware cross-connect that into your other Comparator Input. Take Hardware comparator output pin via Capacitor onto your Supply Line (yes, the supply will need a blocking choke). This will then transmit your modulated data signal back to the Master end. At the Master end, take via a Capaitor (PSU will need a blocking choke as well) into a demodulator (which if you chose the values correctly could be as simple as a Diode and a Capacitor), take it's output into USART or Software Serial pin. In it's raw basic form it's not exactly going to pass EMC specs but who cares anyway! It's not noise immune either, but repeat the data five times with a check-digit. Anyone with a problem with this concept, go read basic AM radio theory.

[sougata]

Hi there,

Can you please specify the following :

1. Does the slave consumes a fairly constant current ?

2. What is the baud rate required ?

3. Repetition rate of communication ?

4. Multi-Master / Multi slave needed ?

5. Exact scenario or implementation ?

6. Are you seriously involved or just a casual thread ?

Regards
Sougata

[Luciano]

Hi,

See these links.

DC-BUS transceivers

http://www.yamar.com/index.html
http://www.yamar.com/products.html


Reduced Cable Smart Motors Communicating Over the DC Power Bus-Line

http://darbelofflab.mit.edu/Progress...Report%208.pdf

Best regards,

Luciano

[NavMicroSystems]

Luciano,

thanks for the yamar links!

I have requested the full datasheets.
Samples will not be available befor end of Feb.

What I can tell so far is:
It is not going to be a cheap solution and will require some PCB space as there is quite some external components required.
(It's not the "single-chip" solution I was hoping it was)

So for MegaADY this is certainly some sort of overkill if he has to transfer just a "few" bytes at "low" speed over a "short" distance.

BTW.
How do you collect all those links?
Are you searching the web all day?
;-)

[keithdoxey]

I dont think you can even use the modulation technique in this instance because its the slave device he wants sending the data (i.e. the device that doesnt have a power supply).

You can use the mondulation technique if you were transmitting data from the master to the slave, but not slave to master.

If you already have 2 wires between the 2 devices, whats so hard about running a 3rd wire for data?

I beg to differ.

The data would be used to turn on and off a carrier which would become an AC signal. This can be superimposed on top of a DC signal quite easily.

It doesnt matter which direction power or data are heading. They simply use the same piece of wire as their connection.

A phantom powered electret microphone effectively does what is requested here. Power is fed from the mixer (MASTER) whilst the audio (or Data) is fed along the exact same wires by the Microphone (SLAVE).

See attached diagram for how data and power would be connected. I havent draw the modulator and demodulator as I dont have a suitable design to hand but generating the signal could be as easy as running the PWM on the SLAVE PIC ad using a transistor driven the the TX data pin to modulate it. At the MASTER end you would need a circuit to detect the carrier and provide a logic level to drive the RX data pin on the Master Pic.

The inductors allow DC power to pass but block the AC carrier so that PSU compoments dont attenuate the carrier.

The capacitors allow the AC carrier to pass but block the DC supply voltage.