Combining power and data wires, Part 1 - Embedded.com

# Combining power and data wires, Part 1

Recently I saw a blog entitled “Power & Signal Over A Single Wire. Do More With Less” by Paul Pickering. Now, I am always on the lookout for original topics for blogs and I was irritated that I didn’t think of it first. However having read it, I feel that there is still some room for me to contribute.

While metamorphosing from student to engineer in the late ‘70s I used to look at the design ideas in the dominant magazine of the time aptly named “Electronics”. The idea of combining signal and power on a single wire was apparently in vogue since I seem to remember several design ideas on the subject. They all seemed to hinge around the same idea that you see in Figure 1.

Figure 1. Two module setup featuring a host and a remote device in half duplex mode. The remote unit was derived from an old Motorola transceiving UART the MC14469. (And yes I know there should be a base resistor connected to each of the transistors.)

It was best if the electronics drew very little current since the circuitry was run from the charge accumulated on the local capacitor. The supply to the circuitry was decoupled from the signal/supply via a diode and a capacitor was used as a supply when the data/signal line was switched to zero. There had to be a current limiter (here the 1 K resistor) so that the power/data line could be pulled to zero. Let me explain the obvious: The unit works in half-duplex mode. When a “1” is transmitted, it turns on the transistor and sinks the current through the resistor. The data/power line is shorted to ground and the decoupling diodes are reverse biased. Power to the electronics is then provided by the storage capacitors, so they need to be sized for the required current and voltage drop. The pulsed signal on the receiver is the serial digital signal. In fact this is the principle of operation of the Maxim 1-wire system as you can see in the block diagram in that link.

Anytime an engineer faces a challenge there is likely to be a fresh approach. In his design idea “Two Wires Carry Power And Data” Robert M. Hanrahan came up with the idea to allow the remote station to sink current when transmitting and then sensing the current changes at the host. The problem is that this does not allow for two-way communication and Mr. Hanrahan proposed a hybrid approach using a variation of the initial approach I described above, where the voltage is modulated and not switched to ground.

Another approach for transmitting data is to treat the data as an AC waveform and rectify it to provide power. This is described in a Maxim application note “Transmit Digital Signals and Power Over Same Wires”, but you can only transmit limited power this way, and you will still have to figure how to get a return signal.

A sophisticated progression of these approaches is to introduce an AC waveform onto the supply line. The major problem is that a power supply presents a low impedance output and the signal would be swamped by this. The solution is to add a low pass filter between the power supply and the signal source so that the signal source sees high impedance. This approach is embodied in the Foundation Fieldbus approach where data operating at a baud rate of 31.25 Kbps is injected onto the power lines. The signal is specified to be 1 Vpp into an AC load of 50 O. Theoretically all the filter needs to be is an inductor of about 5 mH, but an inductor has some problems. Firstly the inductance together with the capacitance in the network can approach resonance and can “ring”. This can be cured by a series resistance of 50 O, however the supply current can be hundreds of milliamps and the power dissipation of the resistor and the dc current in the inductor can cause problems, to say nothing of the fact that there needs to be additional current limiting.

You knew I was leading up to an electronic solution, right? When I was at university we were taught that a gyrator was a very light current method of using a capacitor to look like a coil built using op-amps. Well, this isn’t my professor’s gyrator. I got it from a design idea from P. Strict in Wireless World (yet another fine publication that has bitten the dust). The concept can be scaled up to much higher currents and to bipolar filtering as you can see in figure 2.

Figure 2. Positive and negative gyrator capable of passing several hundred mA DC. Note the AC termination (R29 & C10)

Having presented the gyrator, I need to backtrack and say that magnetics are still used in some circumstances. The TI article “Combine power feed and data link via cable for remote peripherals” describes inductors for high frequency signals.

Power Over Ethernet (PoE) also allows for the combination of power and data on the same lines, called “Alternative A”. Since the data is transformer coupled at either end of the connection, power can be carried on the same lines. My knowledge of PoE is anecdotal and limited, so I just mention it in passing.

I only became aware of the Distributed Systems Interface (DSI) through Mr. Pickering’s article. From the DSI Consortium’s home page: “The Distributed Systems Interface (DSI) is a flexible and powerful bus protocol designed to interconnect multiple remote sensor and actuator devices to a central control module.” There are some ICs that support this standard, all from Freescale (NXP, I guess, now) and warrant some investigation along with the specification itself.

Now some of you may have concluded that this was the end of the topic, but Mr Pickering did get me thinking about analog signaling as well and so there is a Part 2 (cue ominous music!).

Join over 2,000 technical professionals and embedded systems hardware, software, and firmware developers at ESC Boston May 6-7, 2015, and learn about the latest techniques and tips for reducing time, cost, and complexity in the development process.

Passes for the ESC Boston 2015 Technical Conference are available at the conference's official site, with discounted advance pricing until May 1, 2015. Make sure to follow updates about ESC Boston's other talks, programs, and announcements via the Destination ESC blog on Embedded.com and social media accounts Twitter, Facebook, LinkedIn, and Google+.

The Embedded Systems Conference, EE Times, and Embedded.com are owned by UBM Canon.

## 10 thoughts on “Combining power and data wires, Part 1”

1. antedeluvian says:

“DavidnnThanks for the kind words.nn”You can remote a temp sensor from your electronics and still get 0.5 degree accuracy without worrying about the length of your wires.”nnI thought I should point out a design idea from my colleague Girish Choudan

2. Crusty says:

“Aubery, You have done it again good article and it takes me right back to the time I started working at London Transport Executive Laboratories at Chiswick London. nI had the good fortune to be mentored by my then boss, John Ewins a well know contributor

3. antedeluvian says:

“CrustynnThanks for the kind words.nnI am interested to see that the ” two wire voltage modulation from the master and current shunt return from the slaves technique” had such wide appeal. I was only reminded of it when I did the research for the art

4. Crusty says:

“I wanted to know more about the addressable detection systems I was installing for LU, so I was prepared to work with the detector and panel manufacturers non disclosure requirements. The transmit and receive elements are detectable by any one with test g

5. antedeluvian says:

“DavidnnAppearances can be deceiving.”

6. antedeluvian says:

“Crustynn”the slave simply switches different values of resistance onto the power line to indicate fault, alarm and other states.”nnWere the states mutually exclusive, or was the detector capable of decoding the more complex current of two faults com

7. antedeluvian says:

“Iwill us this as a placeholder for more information on this topic:nnThere is a design idea “Simple Circuit Communicates Over Low-Voltage Power Lines” nhttp://electronicdesign.com/communications/simple-circuit-communicates-over-low-voltage-power-l

8. David Ashton says:

“Aubrey…fine article. I have used the Maxim DS18B20 temperature sensors while playing with PICAXEs, but alas the Picaxe will not let you use it in true one-wire mode. But it's certainly a nice way of doing things. You can remote a temp sensor from you