Last Modified:                                                                                                

Homepage / Publications & Opinion /

Peter Cochrane's Uncommon Sense: The Right Technology For The Right Job
Want to send Peter a business plan? Make sure it's not based on this technology...
For a decade now business plans have arrived on my desk, in increasing numbers each year, but only a small percentage find their way to commercial success. Remarkably I see very little correlation between plans.

There is, however, a major exception where the same plan seems to arrive on my desk every year. Each time it is presented as revolutionary and ground-breaking. But it is always the same technology and it is always as flawed and as misconceived as the original I first saw a decade ago.

News reports often included to amplify the case are along the following lines:

XYZ Company is proud to report broadband internet trials are underway using a ground-breaking technology that will revolutionise radio, TV, cable TV, internet and data services to the home. Existing electricity power cable can supply all of your digital services at speeds up to 50Mbits. Extensive laboratory trials have proven this technology, and testing with customers is at an early stage. If fully successful a commercial rollout is planned within the next three months.

It is then customary to include a CEO interview that says something along the lines of:

"All the obstacles have been overcome. The technology is now proven, stable and economically viable, and we are in a position to revolutionise the last mile. We also predict this technology will see the demise of the telco in the next decade..."

Partner companies usually keep their names secret in anticipation of further announcements to be made later that year and the technical press always seem convinced it is all true. But about 6-12 months after the announcements the companies involved quietly say they are ceasing trials and development because some alternative technology has been discovered. It then goes very quiet and nothing more is heard.

I wouldn't find this so upsetting if it only happened once but to my knowledge there have been dozens of false dawns. If only the people involved would visit my office I could save them a small fortune. At a modest estimate, over $200m has been expended to date and no one has been successful in transmitting significant amounts of data over power cables to the home.

I can guaranocations the power companies have installed optical fibre along their power lines for telemetry related to the ctee no one will get this technology to work as advertised. It might be appealing and economically attractive but I'm afraid the basic laws of physics cannot be sidelined. For anyone contemplating the waste of another $5-30m here is my 'don't do it' shortlist:

- Power cables employ low-grade plastic that is unfriendly to high-frequency signals as the absorption per unit length is very high. This alone precludes transmission of high-speed data over significant distances.

- Power cables are not physically symmetrical and are therefore very effective antennas. They radiate energy from high-speed data signals which becomes a source of interference for wireless services including broadcast radio as well as emergency, maritime, aeronautical, military and navigation services. By reciprocity they also suck in energy from every local radio source which further degrades data signals.

- As signals propagate along cables they become weaker but the switching transients from washings machines, refrigerators, vacuum cleaners, electric drills, light switches and other appliances are huge, do not decay at the same rate and swamp data signals.

- Switching transients on power grids with generators going on and off line, dynamic load sharing, fault and maintenance work, all induces massive transients that also swamp data signals.

- Cable joints, transformers, power meters, the on/off nature of electrical appliances and the topology of power grids create large load changes and multiple signal reflection points. This creates a dynamic echo environment where the transmitted signal is further corrupted.

- Real time communications of any kind - whether by telephone, radio or TV - are taken out by the huge voltage transients inherent to power lines and ultimately the data rates achievable for non-real time are also very low.

- Transformers and power meters require a workaround as they present an absolute block to any high frequency signals.

This is a short disaster list that says this technology will not work. And the real nail in the coffin? Telephone and cables were designed to carry far higher frequencies than 50Hz power cables and in every aspect offer superior performance for all data applications. And more recently wireless technology is becoming so low cost and so high performance that signal processing requirements for data over power cables, even if it were possible, would be prohibitively expensive in comparison.

In many lontrol of power distribution. Because their data requirements are so meagre huge amounts of bandwidth are available. So it does make sense for wireless technology to be used at that end point, in a distribution mode to attack the last mile.

Despite all of this there are more than 20 power companies currently active across Europe planning or conducting trials. The reported field performance results are very poor, as expected! Installation costs are higher than the telcos and cablecos. Yet they still seem determined to become the biggest source of radio interference on the planet.

And the ultimate decider? It looks as though the politicians will be asked to decide between the power and wireless lobbies on the legality of the interference levels.

Contrary to a widely held belief, all the technology necessary to transform the local loop is to hand, and sporadic solutions such as data over powerline are a bit of an economically driven red herring.

So please, will the next candidate thinking of sending me another business plan with a revolutionary technology that will exploit power cables please employ someone who understands Maxwell's equations, data transmission and holistic economics.

This column was compiled in a hotel lobby between meetings and despatched via a Wi-Fi link that appeared without identification - thanks to whoever decided to provide this charitable service.