The power industry's poor relation is currently experiencing a sea change in attitude thanks to a sudden interest from social media. Can power electronics start to reshape the world?
For decades it has been the sector to ignore. Few, even those doing product design, took much notice of the companies making power electronics and when they did it was often at the last minute - when they needed a power supply to take a product to mass production. It's one of the reasons why so many of our gadgets ship with anonymous wall warts that, once separated from the products they power, become practically impossible to marry up again. But now the industry's poor relation is the one holding the key to future development and even how well the world manages to deal with the problem of excess greenhouse gases.
Andrew Repton, senior engineering director at Dialog Semiconductor, described at the DATE conference in Grenoble in March how even software giants have suddenly taken an interest in power electronics. "Our power management IC technology has been a significant part of our growth. Now we are even seeing input from social media companies wanting to drive what we do. They bring an interesting perspective."
The sudden interest from social media comes not from academic interest but is a reflection of how important the smartphone is to the future of a lot of companies. One big problem for the social-media sector is device battery life. Articles on the Internet describe how social-media apps are some of the biggest drains on the phone's battery because they keep having to phone home. Greater cooperation between the power electronics and the network software can keep the battery going longer and, as a result, prevent what Facebook and others absolutely want to resist: that users switch off their status updates. This change in attitude is a microcosm of what is happening in the rest of the industry and how power electronics could reshape the world.
Semiconductor companies such as Infineon Technologies have bet on the long-term impact of power electronics, shedding its high-end digital silicon lines that needed to be focused on Moore's Law scaling, and its increasing costs, in favour of its own power technologies as well as purchasing one of the original power-electronics component suppliers, International Rectifier, for $3bn.
Infineon has good reasons for believing power electronics will be a strong area for growth. The International Energy Agency found the increase in global demand for electricity is expected to accelerate ahead of the projections for primary energy consumption, although some of that growth will come from technological changes intended to reduce overall energy consumption - such as the electrification of transport.
There is a second reason. The IEA's 450 Scenario, named for its goal of limiting the concentration of CO2 in the atmosphere to 450 parts per million, concluded that in the near term especially, efficiency improvements represent the least costly method of hitting the target and then bringing global greenhouse-gas concentrations down.
Significant savings could be across a variety of areas. A number of surveys have claimed that simply improving the way industrial motors are run could make massive savings. Electric motors account for 9,000TWh/year or 40 per cent of global electricity consumption. According to the World Energy Council, in some developing countries with large industries and outdated'electrical equipment, the share of electricity consumed by motors is even higher. More advanced motors and controllers could save about 1,000TWh, about the same as Japan's annual electricity consumption.
But against that seemingly large saving you have to take into account the share of electricity within total energy usage worldwide - just 15 per cent. Most energy goes into heat - either to heat homes or for incredibly energy-intensive industrial operations such as smelting metals.
Home appliances account for 13 per cent of all electricity consumption and improved motor technology could make similar savings, although the relative reductions are likely to be lower because western homes replace their appliances more frequently than industrial users upgrade their production systems - with the result that energy-saving advanced motor controls are already more widespread.
The savings made possible by switching to solid-state lighting are likely to have a more significant impact overall. Although high-efficiency power semiconductors are needed to convert the 240V AC or 24V DC supply down to a voltage suitable for driving LEDs, the main savings would come from the LEDs themselves. The US Department of Energy estimated in 2012 that a switch to LEDs would halve lighting energy consumption. As lighting worldwide accounts for 19 per cent of global electricity consumption and that the US-reported savings hold for the rest of the world, this would most likely have a greater impact on energy use than either motor replacement.
Does transportation hold the key?
Lighting's share of electricity will be pushed up by a shift away from kerosene lighting in developing nations. But this is likely to reduce overall greenhouse-gas emissions. For similar reasons transportation offers one of the biggest opportunities for power electronics to eat into CO2 production. A move towards electrification could result in massive savings both in terms of overall efficiency and CO2 production by replacing oil as a fuel with electricity derived from cleaner sources. But the situation with electric vehicles demonstrates how delicate the balance can be between moving to higher electronics content and making environmental savings. Not surprisingly, the greatest environmental cost calculated using standard models comes in terms of CO2 emissions during use. Cars with internal combustion engines can last for 20 years and their production costs have gradually come down as manufacturers have made more efficient use of materials.
A 2012 lifecycle analysis put together for the California Air Resources Board by UCLA estimated 96 per cent of a conventional vehicle's total emissions come from usage. This figure drops dramatically for an entirely electrically powered vehicle based on a conventional mixture of electricity sources. The switch to electric traction practically halved the emissions, but of that total, making the battery alone contributed a quarter. Assuming up to a third of the generating capacity will include renewables - which is the aim for California by 2020 - the battery vehicle's emissions dropped to 40 per cent of those of the internal combustion-engined car. Using China's coal-rich generating capacity pushes the total in the other direction.
A hybrid vehicle can use a much smaller battery as most of the energy while driving is still consumed by the internal combustion engine. As a result, although the hybrid was estimated to be responsible for 20 per cent more emissions overall, the manufacturing component was only around a tenth of the battery-driven vehicle.
In these calculations there is a strong sensitivity to the source of electricity, the efficiency of charging and the lifetime of the battery itself. Vehicle batteries lose their ability to carry a full charge over time. Over the typical lifetime of a vehicle it's likely that the battery will need to be replaced to ensure that the vehicle maintains a reasonable driving range, incurring the cost of a second battery, further increasing overall environmental cost. The real impact of those additional batteries depends on how society chooses to deal with them. Although vehicles are quite sensitive to battery capacity, they could be repurposed for grid storage - holding charge temporarily to even out usage peaks.
At the same time, a number of dramatic improvements have been made in fuel efficiency, particularly for hybrid vehicles where the ability to store energy temporarily in a small battery can be used to maximise the yield from the combustion engine. Here, power electronics can make a strong contribution because electrical power conversion and the ability to deliver it efficiently lies at the heart of both electric and hybrid vehicles. Manufacturers have invested heavily in building variants of the basic transistor to deal with the stresses involved with channelling high currents.
Assuming these savings come to fruition, power electronics is not going to decarbonise the world, but it has the potential to help put a massive dent in the 100 quadrillion BTU (100EJ) demand for oil in transportation, potentially cutting it by half. Some of that will have to be weighed against increased losses from grid distribution - which exceed the energy used by transportation. Those losses could increase due to the amount of electricity that will be held temporarily in somewhat lossy batteries, but better power semiconductors and higher-voltage distribution can help reduce those inefficiencies.
In taking those losses into account, it is not unrealistic to expect better power electronics and the systems they enable to make a difference of more than 10 per cent to overall energy consumption and, assuming renewables grow as expected, significantly more than that in terms of CO2 production. It's not solving the problem, but it's a good head start. *