newsletter link
mondo arc

The year of the LED lighting fixture

Issue 47 Feb / Mar 2009

British government LED adviser, Dr. Geoff Archenhold, reports on new government initiatives that will speed up the progress of LED fixtures in the commercial market.

Unfortunately, the lighting industry is getting to grips with the start of what seems to be a long recession. However there are significant opportunities that point to success for the near-term future of lighting, one of which is high quality LED lighting fixtures.

2009 is a pivotal year for the lighting industry where traditional lighting companies need to invest heavily in the technology to provide quality light for lighting designers. This may not be an easy task for the larger lighting players as they focus on near-term issues such as how to restructure or close conventional product line rather than invest in developing the next generation of technologies – this is an ideal time for innovative newcomers to make big strides in delivering innovative lighting solutions.

This month’s article focuses on new government initiatives that are aimed at supporting new technologies such as LED lighting both in the UK and Europe, as well as discusses a few of the new research breakthroughs that will increase the efficiency and reduce the costs of LEDs from 2010 onwards.

An update on Government legislation and support for new technologies
Unless you were without a TV and didn’t read any newspapers or websites you will have seen the European Union have agreed to finally ban the incandescent light bulb. Although this is a very positive step, it is no doubt a very small one mainly due to the fact that several inefficient or environmentally unfriendly light sources such as Halogens and CFL lamps are essentially left untouched. In my opinion there are many reasons why you have legislation but the key aims should always be to ensure safety (whether its product safety or environmental safety through reduced carbon emissions) and to stimulate an improvement in technology which provides an economic stimulus. Unfortunately, banning the incandescent on its own will not achieve significant gains on either front mainly due to the fact that CFL and Halogens are the main products marketed today. One only needs to go to any of the top five supermarkets in the UK to look at the lighting section to see that it is dominated by CFL’s from two or three companies along with a low cost supermarket own label version.

What does this EU directive actually mean for the Industry?
The EU directive sets minimum requirements for domestic lighting so that from 1 September 2009, high-energy bulbs can no longer be circulated in the EU for use in private households. These bulbs will gradually be withdrawn until 2012. The EU directive defines exactly which bulbs will no longer be able to be sold and when. For example, as of 1 September 2009, there will be no more frosted lamps, except for frosted energy saving lamps (CFLi) with energy efficiency class A. Also light bulbs with 80 watts or more will be banned as of this date.

This new EU directive is part of a package of measures from the EU which aims to significantly reduce the energy consumed by electrical devices. The primary aim of this directive is to protect the environment by reducing CO2 emissions.

It is important to note that these regulations apply to non-directional lamps. Directional (reflector) lamps such as spots will be covered by a dedicated measure at the end of 2009 or in 2010. These new regulations could be an opportunity to start phasing out CFL’s especially those used within directional downlight fixtures where such light sources where approximately 50% of the lamp lumens are lost in the fixture design.

Providing LED fixture manufacturers continue to improve light output quality and consistency then the EU directive will become the first step to banning the CFL’s and Halogens within the next 2-3 years as LEDs become more cost effective especially as they are already proving to be more efficient in certain applications.

What will actually change?

As of 1 September 2009, the first styles of light bulbs (80 watts or higher) will disappear from the shelves. Every year after that, in steps of twelve months, further bulbs and wattages will be affected by the ban. By 2012, all light bulbs of greater than 7 watts will be withdrawn from sale except for certain special applications such as ovens or fridges, traffic lights, infrared lamps.

Also halogen lamps with the efficiency class D will be gradually phased out step by step according to their watt and lumen value from 2009 until 2012.

An update of the lamp energy label (Directive 1998/11/EC) is also planned in 2009 in order to redistribute the energy classes after the phasing out of classes D, E, F, G and also to integrate low voltage lamps and reflector lamps, which are currently not covered by the label.

Energy Savings Trust – Starts a 2-phase LED Lighting trial
As part of the UK Government’s £400m Environmental Transformation Fund (ETF) setup in April 2008 and designed to bring forward the development of new low carbon energy and energy efficiency the Energy Saving Trust (EST) has secured funding for LED-based project over the next three years. The first of these projects are the planned field trials of LED lighting technology in communal areas of social housing to begin before April 2009.

The attraction of this type of housing is that the lighting is on 24 hours per day, 7 days a week resulting in high energy costs with traditional lighting and considerable potential savings using LED technology.

A previous study undertaken by the Lighting Association for Aberdeen County Council concluded that the new LED products were suitable for installation in the dwellings, replacing either fluorescent or incandescent lighting yielding several benefits such as reduced energy consumption, reduced carbon emissions and reduced maintenance costs.

The new ETF/EST field trial will seek to install LED lighting in a number of communal housing locations throughout England and the phase 1 locations have already been selected.

The first phases of this field trial are important as there are over 4 million dwellings in England owned or managed by Local Authorities (including ALMO’s and RSL’s) and Housing Associations. This represents approximately 26% of the English housing stock so is a considerable market for LED lighting companies to focus upon.

I gather the interest to participate in phase 1 of the trial was substantial with more than 110 housing managers and over 30 LED companies requesting further information.

In Phase 1, the Energy Savings Trust will fund up to a maximum of 80% of the costs of the proposed replacement LED fittings and has a total fund size of ~£350,000. Expressions of interest to take part in Phase II trials (a much larger trial) will commence in April 2009 so if you are interesting knowing how to take part then please do not hesitate to contact me by email at

The sites will be fully monitored before, during and after the installation of the LED replacements so that a complete and robust case study for each site can be published in the future. To my knowledge this will be one of the first times data in the UK will be published by an unbiased source and so will provide a good comparison between conventional and LED equivalents in certain applications.

In addition, the Energy Saving Trust will be adding a performance requirement for domestic LED lighting into the EST’s lamp specification version 6. The first three classes of LED lamps to be endorsed under ESR are:

• Mains voltage reflector lamps such as GU10 replacement lamps ie; 240V and power equivalents to a 20,35 and 50W halogen having a minimum life of 25,000 hours.

• Low voltage for incandescent reflector lamps such as MR16 lamps ie; 12V 20, 30 and 50W LED equivalents having a lifetime in excess of 25,000 hours.

• External LED luminaires to replace typical 250W and 500W Halogen floodlighting.

I believe the work EST are undertaking is vital and exactly what the consumer and lighting industry need. Quality and unbiased performance data on LED fixtures and supporting innovation through the ESR programme.

Further information and presentations on the EST LED initiative can be downloaded from

LED efficiency is continuously improving
The leading manufacturers are continuing to rapidly improve production LEDs enabling high efficiency LED fixtures to be developed. For example, both CREE and Lumileds have single-die LED products that are readily available with minimum luminous flux that are approximately 100 lumens per watt as shown in table 1.

It is clear that there is not much between the leaders of LED manufacturing except for the Warm White Colour Temperature range where Cree has an obvious lead today.

Importantly, there will be some changes during this year and Philips are working on a new technology that is incorporated into the manufacturing of LEDs which will not impact on the total light emitted from an LED die but will significantly improve the efficiency. I have guestimated what this new technology would perhaps deliver for Philips-Lumileds products by mid-2009 when they should be available in quantity. As you can see from the final column in table 1, the lumen per watt figure provides a significant jump in efficiency performance.

It is clear that excluding the great LED improvements that I discussed in the December 2008 issue we will continuously see White LEDs approach the 120 lumen per watt mark at 350mA in the next year which is a significant increase in performance.

In terms of fixture design, this means less number of LED’s are required for a given light output and the thermal management can be much simpler making the heatsinks for the LED lighting smaller which also increases flexibility and reduces cost.

Further improvements of LED dies are continuing at pace in the R&D world and recently Researchers at Rensselaer Polytechnic Institute (RPI) have developed and demonstrated a new type of LED with significantly improved lighting performance and energy efficiency. The research shows that a polarisation-matched LED exhibits an 18% increase in light output and a 22% increase in wall-plug efficiency compared to currently constructed LEDs.

Importantly, the new LED device achieves a notable reduction in ‘efficiency droop’, a phenomenon that allows LEDs to be most efficient when receiving low-density currents of electricity, but then to lose efficiency as higher density currents of electricity are fed into the device. Essentially, this type of new research would enable LED manufacturers to build LEDs that are highly scalable in terms of driving them at higher forward currents whilst still maintaining the lumens per watt efficiencies seen at the typical 350mA as shown in figure 1. Current LEDs can lose up to 30% of the increased light output when they are driven at 700mA rather than 350mA. This research will help LED manufacturers make up the 30% loss when driven at high currents.

Further Fixture Level Improvements have just a significant role to play as LED emitters in the latest high quality LED fixture design. Today, thermal management is one of the most important system design issues so any new technologies that come along which can reduce the issue enables improved performance.

Once such new technology is graphite PCB’s which can replace the more commonly used Metal Core PCB (MCPCB) adopted by the high power LED industry. Strictly speaking graphite PCB’s are not new technology and they have been used in military applications for many years however their costs have made them rather exclusive in application. However a new Taiwanese company, Polo Tech, is now marketing graphite PCBs (GPCB) for the LED industry.

The new GPCB contains 99.5% graphite and offer a thermal conductivity greater than 400w/mK which is better than most MCPCB’s. Another advantage of GPCBs are they are very lightweight with a density of 1.5 g/cm3. The improved thermal performance means that more heat is transferred from the back of the LED to a heatsink potentially reducing the junction temperature of the LED die and so improving performance and efficiency.

Remote Phosphor Technology – the potential and pitfalls
I thought it would be a good time to briefly discuss an LED technology that is starting to penetrate the current LED marketplace known as Remote Phosphor Technology (RPT). RPT was first pioneered in the LED sector some nine years ago but it was researchers at the Lighting Research Centre, Rensselaer Polytechnic Institute in the US that provided a practical way forward. They developed and patented a new packaging method for white LEDs called Scattered Photon Extraction (SPE) which enables higher luminous efficiency by placing the phosphor at a remote location from the die by shaping the lens surrounding the die to extract a significant amount of the back-scattered light from phosphor before it is absorbed by other components within the LED emitter. Even back in 2005, the SPE technique showed a 20-25% improvement in efficiency compared to standard white LEDs which was an incredible feat considering how bright LEDs have become in four years!

The most popular way to create a white LED is to use a Blue LED chip which is coated with a phosphor and there are several common methods used in production as shown in figure 2. Referring to figure 2, the majority of LEDs are created using either (a) or (b) where the phosphor is either conformally coated around the LED chip or dispersed throughout the coupling medium respectively. Conformal coating is considered the better technique as the quality of light output is improved and there is less of a CRI or CCT variation with beam angle.

Philips Lumileds have recently developed a new method, Lumiramic, mentioned in previous articles where the phosphor is placed within a small disk above the die to improve quality and consistency.

The first iterations of remote phosphors placed the phosphors away from the LED chip but still within the emitter package as shown in figure 2c and 2d. Although this showed an improvement there would still be significant issues with LED temperature especially as phosphor efficiency and lifetime degrade as the ambient temperature increases.

A further issue is that when blue light which is absorbed by the phosphor it converts the light into a different wavelength commonly know as phosphor conversion. Unfortunately, phosphor conversions scatters light in all directions (not just forwards) and hence studies by the Lighting Research Center (LRC) have shown that up to 60% of the lighting created by the phosphor is scattered back toward the LED die and lost due to absorption.

Even when there is a highly polished reflector used as shown in figure 3 the arrangement did not show much of an improvement in terms of extracted light output over traditional packages where the phosphor was placed close to the die. The LRC uses the SPE technique to extract the back-scatter light from the remote phosphor before it is absorbed by using a secondary optic that efficiently transferred the blue light to the remote phosphor whilst recycling the back-scattered white light from the system so it is not lost as shown in figure 4.

The SPE technique shows a 60% improvement in light output and efficacy compared to similar commercial white LEDs, where the phosphor is placed close to the LED die. Although this research was undertaken in 2005, when traditional white LEDs had an efficacy of 30 lumens per watt at 350mA, the SPE system showed an efficacy of approximately 50 lumens per Watt as shown in Figure 5. If we applied the SPE or similar light recycling techniques to the latest blue LEDs we would be looking at efficiencies of 150-160 lumens per watt!

More recent studies by the LRC indicate how the CCT of an SPE fixture varies with different phosphor densities as shown in Figure 6.

The variation in phosphor density and thickness is one of the key issues for RPT based white LED fixtures hence why manufacturers have taken a long time to begin to introduce such products to the market.

There are several research groups around the world that have been working on RPT technologies where they have mixed two or more different phosphors together enabling a variable colour temperature or CCT to be achieved by using the relative intensities of two different wavelength LEDs. To my knowledge I know of one company that has commercialised this into a product (KPT of Korea) and I must say it has impressed all that have viewed it.

The early-to-market RPT products such as the Philips Fortimo and the much advertised alphaLED products need to be effectively tested for CCT and CRI consistency between fixtures and their lifetime and performances monitored especially if they are using standard blue LEDs illuminating a single phosphor plate.

It is very difficult to ensure the phosphor materials are placed across a 25mm to 100mm plate with a homogeneous density and consistent thickness. If either of these variables change significantly then the CCT and CRI of the light fixture will vary and although it is possible to replicate one or two samples for research it is much more difficult to scale up for higher volumes.

This said, the possibility of significantly improved white LED luminaire efficacy and lifetime which certain remote phosphor technologies have shown, paves the way towards the 150 lumen per watt LED fixture using today’s LED technology.

It will be a matter of time before high quality, consistent and efficient RPT LED fixtures will reach the market with the majority of research required being in the production capability of producing thin consistent layers of phosphor on plates.

The LED fixture manufacturers need to be honest

One of the largest issues that the LED industry needs to face in 2009 is the misleading use of data on specifications sheets. Every time I speak with lighting designers they say that they are constantly being disappointed by LED companies that claim a certain light output and fixture efficacy but when they see the technology it falls way too short.

The main issue is that most of the LED manufacturers quote LED lumens on their datasheets that does not account for thermal degradation, optical efficiency or LED efficiency. For example, I see many specification sheets that claim a fixture with 9 White LEDs can produce 810 lumens ie; 9 x 90 lumens minimum = 810 lumens. Then the manufacturer will state that the fixture takes, for example, 12 watts of power so that the fixture luminaire is ~70 lumens per watt which is totally misleading.

I implore all LED fixture manufacturers to state realistic and actual operational figures on their datasheets so that the industry maintains a professional standard. I believe the UK Government should adopt a strategy similar to that employed by the Caliper programme in the US which tests commercially available lighting products and compares their actual performance against their datasheet stated performance. This way we are able to name and shame organisations that set out to mislead purchases!

Despite this plea, we will see LED fixtures accelerate into the general lighting market and system costs will plumment as the new technologies in R&D filter through in one or two years!

Next issue
In the next column I will look at the next big issue to hit the LED industry that of system reliability and what is perhaps the weakest link, LED drivers.
If you have any questions or comments that you would like answering then do not hesitate to contact the editor of mondo*arc or myself.


Table 1: The minimum luminous flux LEDs and efficiencies available today and during 2009

Figure 1: A significant improvement in forward current droop (shown in Red) of a blue LED die commonly used in White LEDs. Source: Martin F. Schubert et al

Figure 2: Common methods of creating a white LED
Related Articles


Follow us on…

Follow Mondo Arc Magazine on Twitter Follow Mondo Arc Magazine on Facebook Follow Mondo Arc Magazine on Linked In

mondo arc india

darc awards DWLF IALD PLDC LRO