Most commercial buildings in Australia have moved away from incandescent bulbs to compact fluorescents, which are much more energy efficient and last longer. The limitations of CFLs are slow warm up time, early failure if frequently switched, and high cost for dimmable CFLs. Additionally some speciality bulbs, such as chandelier bulbs, don’t have readily available CFL equivalents. But as CFLs are four or five times more efficient than incandescent in our energy audits we always try to build a strong case for switching to CFLs.

But incandescent may be getting a second life. Australia enacted the first legislation banning  sales of low efficiency lamps (incandescent) and the US followed. With a much larger market than ours this has sparked some innovation in the design of incandescent lamps.

Philips now has a incandescent that is 30% more efficient than a standard incandescent. Osram is shortly coming out with one 25% more efficient.

These sort of efficiency gains still leave CFLs as clearly the superior option, but as there is more research undertaken the incandescent could get even better yet.

If incandescent efficiency can be improved by 20% a year, it will take six or seven years to catch up with where CFLs are now. Which is a long time, unless there is an innovation that provides a quantum improvement in efficiency.

LED lights on the other hand are now getting close to CFL efficiency.

Its great to see all this lighting innovation happening, and hopefully we will soon see screw in and plug in bulbs that are more efficient than CFLs

Research recently undertaken by Siemens says that the lifecycle efficiency of LED lights is equal to that of compact fluorescents.

Measuring the lifecycle efficiency involves looking at the energy to manufacture and dispose of the product, in addition to the energy it uses whilst in operation.

The report was released by Osram, which is owned by Siemans, with the testing taking place by the Siemens Corporate Technology Centre for Eco Innovation, and reported on in the New York Times.

There aren’t many details yet though as to how the research was undertaken or the numbers behind the claims. For example, was the LED light used in the comparison of equivalent brightness to the CFL. 

But it does show that LEDs are getting closer to being the light of the future.

A few months ago I wrote a blog posting about how tighter regulation of electricity supply voltages could save Australia 15 million tonnes of greenhouse gas a year.

However a comment on that posting suggested that voltage reduction may not result in any useful savings.

Below I report on the results of an experiment we undertook to identify how much power can be saved, if any, by operating equipment at a lower voltage.

We measured a variety of single phase loads at different voltages. A variable transformer was used to vary the voltage. A German made Power Tech plus plug in power meter was used to measure voltage, current, power and power factor at the different loads. Loads experimented with included typical single phase lights, computer equipment and a fan.

The experimental set up is shown above. Below is a graph showing the results of the testing.

This graph clearly shows that for common lighting loads power consumption decreases with decreased voltage

• Incandescent lamp (resistive load)
• T8 fluorescent (inductive load)
• T5 fluorescent (electronic ballast)

The reduction in power consumption with the T5 fluorescent (with an electronic ballast) was unexpected.

The fan, with a single phase (shaded pole?) motor, also used less power with lower voltage, interestingly the power factor improved as voltage was lowered, with the power factor the highest at 220 volts.

The PC computer and monitor both showed lowest power consumption at 230 and 240 volts, but power consumption generally did not decrease with voltage. Power factor improved a little at lower voltages.

This experiment shows that for a variety of loads power consumption is in fact less at lower voltage.

For heating or cooling loads equipment may need to run longer when at lowered voltage to reduce the same amount of heating or cooling, with no net energy savings.

Three phase synchronous motors are unlikely to use any more or less power (a theoretical assertion, we don’t have the equipment to test), having the motors run at 230 volts rather than 240 or 250 volts however is unlikely to cause motor damage due to excess current as the voltage difference is only small.

But with lighting and many single phase motors power consumption drops with lowered voltage.

My back of the envelope calculations still come up with a saving of around 15 million tonnes of greenhouse gas if voltages were closer to the 230 volt standard rather than being at 240 to 250 volts.

If high voltage drops in distribution were a problem additional network infrastructure could be used to deliver a more consistent voltage across the network. 2009 is the year of the “smart grid.” A smart grid could mean multitap transformers that can be changed on the fly to deliver a more consistent 230 volts across the whole electrical network.

If LED lighting continues to develop as fast as it has over the last five years, within ten years it may well be the main form of lighting in use across the world. And lighting in new buildings may look radically different to what it does now.

The major advantage of LEDs is that they are a directional light source. Most other artificial light sources on the market radiate light in all directions. Incandescent light bulbs, compact fluorescent lamps, fluorescent lamps and high intensity discharge lamps such as metal halide all radiate light in all directions.

For example in most fluorescent office light fittings typically only 60% of the light produced by the fluorescent tube is emitted as useful light. A great deal of the light is lost because it goes  upwards or sideways rather than down. Using a reflector may increase the amount of useful light provided up to 80%. But even the most efficient fluorescent light fittings on the market rarely have a light output ratio of above 80%.

LED lights on the other hand produce all their light in a single direction. Light fitting designers can take advantage of this to efficiently direct light exactly where its needed, with very little lost or wasted light.

Go into a progressive hardware or electrical store and you can already see a variety of LED lights being sold.

Fluorescent lamps are the most commonly used lamp in the world. LEDs however are not yet competitive with fluorescents for three main reasons:

1. Energy efficiency is similar but not yet better. A high performance fluorescent tube will produce 100 lumens per watt. Put it in an energy efficient fitting, with a light output ratio of 80%, and the overall lighting efficiency is 80 useful lumens per watt of electricity. The best white LEDs on the market (that we are aware of) produce 75 to 80 lumens per watt. This is good, but not yet better than, fluorescent.
2. Reliability. Unlike fluorescent tubes, which are generally reliable no matter who makes them, LEDs are often unrealiable. We have purchased LEDs from many different manufacturers, and over half have failed within the first year of use.
3. Price. LEDs are still expensive.

This, however, is changing. Energy efficiency is improving, the major lighting manufacturers are increasing their focus on LEDs, and prices are dropping.

Energy efficiency of LEDs has increased markedly in recent years, in 2006 the best LEDs were approaching 60 lumens per watt, by the end of 2008 they were up to 77 lumens per watt. 

Reliability. The three main global light manufacturers – Osram, Philips and Sylvania – are all now selling LED lights. As major global brands they are unlikely to risk the cost and reputational damage of supplying unreliable products. As LED products become more main stream we can expect reliability to improve.

Costs are now starting to decrease as well. Whilst it is difficult to purchase a LED fluorescent substitute light for less than $80, only two years ago the price was $100.

When LED lights are achieving energy efficiencies in excess of 120 lumens per watt, lamp costs of less than $2 a watt, and low failure rates (less than say 1%) lighting as we now know it will be superseded. It will be possible to retrofit LEDs and cut lighting energy use by 50% or more in almost any building. New buildings, with lighting designs built around LEDs, may well be providing office – standard illumination for 2 watts of electricity use per square meter or less (current best practice is around 5 watts per square meter).

These new lights may look very different. Light fittings may become panels with hundreds of LEDs on them. Or ceilings may end up with stripes of LEDs across the ceiling. Or ceilings will end up with sockets into which panels of LEDs can be plugged in, so that its easy to move LEDs around in response to the lighting needs of a room (more above a desk, less in the corridors).

If control and sensing technologies can become sufficiently low cost buildings may well be set up to provide lighting whose intensity varies with occupancy and usage.

The rapid development of LEDs is exciting. It gives me hope that, when it comes to lighting, humanity will be able to greatly reduce its carbon footprint in the not too distant future.

In October last year CarbonetiX started an independent evaluation of LED lights as a substitute for fluorescent lighting. LEDs, standing for light-emitting diodes, have previously been commonly used for other purposes such as for low energy indicator lights on household equipment, but have not yet been widely used for general commercial lighting.

The evaluation is being undertaken by CarbonetiX in partnership with the Sustainability Fund, managed by Sustainability Victoria, and with the support of Frankston City Council. Eight months on and the trial is now nearing conclusion.

176 fluorescent tubes were replaced with solid state LED lamps in the Mahogany Neighbourhood Community Centre in the City of Frankston.  Users of the facility were surveyed before and after the upgrade and noted either no change or an improvement in the lighting. An illumination assessment showed that illumination levels after the upgrade were around the same as before. Yet power consumption has dropped from over 40 watts per lamp down to 18 watts.

The trial has involved firstly a desk-top evaluation of LED products, then selection of lamps from those six manufacturers who appeared to have the best products. These were then tested by CarbonetiX for light output and power consumption. The best performing lamp was then sent to a NATA (National Association of Testing Authorities) certified laboratory for photometric testing.

It was disappointing to have the only Australian lamp fail during our in-house testing. However the overall testing result was  surprisingly good: the useful light provided by the best lamp in a standard office troffer was similar to that of a used halo-phosphor tube, whereas our earlier program of testing indicated the LEDs were just not bright enough to be used as a fluorescent substitute.

This means that where a building is currently lit by halo-phosphor lamps, which are still quite common fluorescent tubes, and where the illumination levels exceed those specified in AS1680, that the best LED tube could be used as a fluorescent substitute. 

Another concern was the reliability of the product. Barney Mezey, our energy auditor who ran with the project, was concerned about the headache that failure of the lamps would cause. Fortunately all of the lamps are still working three months after they were installed. Obviously this is nowhere near long enough to establish whether or not the lamps will operate for 50,000 hours or not as claimed by the manufacturer. But it is a good start..

LEDs are still expensive, with CarbonetiX estimating a twelve year return on investment. But this trial indicates that if the technology continues to evolve and prices drop that LEDs could help halve the use the energy used by lighting in commercial buildings.