LED light bulb 'achieves the impossible'

The challenge was laid down by the Department of Energy and Climate Change and funded through Small Business Research Initiative grants from the Technology Strategy Board. Its £0.5m grant was for the development of a British, energy efficient LED light bulb that could win a chunk of the 60W light market that is worth in the region of $60billion globally – approximately 40% of the total light bulb market. The prime driver was to find a low power LED based equivalent, typically around the 10W mark, to the conventional 60W incandescent bulb. However, combining this with other specifications presented a daunting task. Shadbolt commented: "They said you need a colour temperature of 2700K, a power factor of 0.9 or better, harmonic distortion, what lumens we would need to get, 360° light; all within the 60W bulb form factor. We spoke to a lot of people and they said it was not physically possible. They said the Government doesn't know what it is talking about – a stupid idea." It appeared initially that everything would be a problem. For example, when a feasibility project was launched in 2009, there were no LEDs available that would provide the specified 2700K colour temperature – warm white – at 100 lumen/W. The core driver was efficiency, which itself comes in different forms. One is the luminous efficacy – what the eye perceives. The 2700K specified is towards the red end of the light spectrum, where the photographic response of the eye is at its least efficient. This is why the light output is both important and difficult to achieve. The other metric is system efficiency, the wallplug efficiency, which concerns heat loss within the LED die. "Typically, LEDs were only 25% efficient, but they are getting closer to 40%. That's still not good enough, because the 60% you are losing needs to be managed as heat," said Presanna Vijaykumar, technical director. This heat needs to be pulled out of the LED because performance and longevity suffer with excess temperature – and the most common method of removing heat is through conduction. "That is where heat management of the Life Bulb is critical," claimed Vijaykumar. One of the challenges was to maintain the form factor and be close to the weight of a conventional 60W bulb, so excessive surface area and weight for heatsinks were not options. "A key aspect was providing the largest surface area and using convection as a means of getting the thermal efficiencies that would allow us to achieve the LED life figure," said Vijaykumar. Shadbolt added: "What most LED lamps are missing is using the surface area. At the moment, the standard competitor bulbs on the market have a big metal heat sink with LEDs slapped on top of it and a diffuser covering that prevents convection. That approach works primarily from a manufacturing point of view, however it doesn't use the heat sink area effectively. "We use the entire surface area of the bulb, keeping the form factor intact and allowing for convection, which not only cools the LEDs, but also the drivers within the system. One of the challenges with this type of lamp is not that LEDs fail quickly – they have a known degradation curve – it is things like capacitors used within the driver that fail more quickly, so we had to manage thermals within system requirements, not just at the LED level." While the rest of the mechanical design was done in SolidWorks, Zeta used computational fluid dynamics (CFD) software from Flotherm for thermal modelling and simulation. By taking the .STP file from SolidWorks and inputting power specifications, materials and ambient conditions, a simulation can show the thermal coefficient of the metal required to remove the necessary heat from the LED, as well as the thermal resistance between ambient and the LED junction. The latter is used to ensure the efficient operation of the LED. The resulting casing design, with its large surface area and air vents, is one of the patented features of Life Bulb. Vijaykumar observed: "You can't simulate environmental variables, like a spike in temperature, you can only do a steady state analysis. But real life is not steady state, so once we made the prototype, we would use thermocouples to see how hot the case got. In general, it does not exceed 50 to 55°C." The typical temperature difference between ambient and the LED junction is typically 15 to 20°C – at 70°C, the LED's performance starts to droop.The measured values are within tolerance to provide a long life solution. Dimming performance Another problem associated with LED bulbs relates to dimming. Most dimmer circuits are TRIAC based, which offers compatibility problems because they use the resistive load of the incandescent bulb. There is a DIAC (diode for alternating current) within the dimmer circuit to trigger the TRIAC. This relies on a time constant based on the AC frequency, in combination with the RC time constant, and part of that RC time constant comes from the incandescent bulb's resistance. An LED bulb is of a completely different nature because the power supply used within the LED system has EMI filters, required to pass the CE marking. Vijaykumar explained the problem: "These are, in other words, just big capacitors that sit across the line. In effect, the dimmer starts to get a mixed RC time constant, which it would not have got with an incandescent bulb, and starts to misfire. Within the TRIAC, instead of firing once per cycle, you will have multiple firings within the cycle, which causes flickering if it is not handled well. It could also limit the life of the TRIAC. "We have designed this out by increasing the resistance within the bulb's driver, but have had to do this efficiently – you don't want to put a big resistor across the line, kill the efficiency and defeat the purpose of having an energy efficient bulb. "This circuit features a dynamic pulsed resistor, which provides the resistance required within the cycle, to make the DIAC function correctly and not to misfire. The circuit is clever enough to map the ability to manage the TRIAC dimming, along with power factor correction, which requires dynamic impedance." Bright future? Having delivered on the technical challenges, is the Life Bulb ready for commercialisation? Not yet, said Shadbolt, who believes the market is too price sensitive for a product with a premium price tag. "We are now thinking that we want to make a more intelligent product, so you can control it from a phone or an iPad. We are hoping to sell a premium product for £20, rather than compete with a volume product at £10, which we will never do."