Rob talks diodes...
Light emitting diodes – LEDs. Small devices that turn electricity into light. Since their commercial introduction in the 1960’s LEDs have been squeezing ever more light from each bit of electricity that passes through them. The rate of improvement has been astronomical; their efficiency has been increasing exponentially since the '60s – a trend referred to as Haitz’s Law.
LEDs are now approx. one million times more efficient than they were in the '60s.
To put that into context, if the bottom of the range Ford motor car were to have improved its efficiency (miles per gallon) at the same rate, since 1960, you would now be able to achieve 33 million miles per gallon. That’s to the moon and back over 1000 times, on a single tank!
Despite these gains, it took developers over 30 years to design a successful white LED; the very first LEDs in the '60s were expensive and most could only achieve an infrared light –useful in some transmission devices but not visible to the human eye. The '70s saw mainstream introduction of LEDs in the visible light spectrum, in various reds and yellows. At around 5 cents per unit these were the first to be commercially viable. In ’94 the Nichia Corporation finally demonstrated a blue coloured LED, with a novel design for which the developers won the Nobel Prize in physics. By coating this blue LED with a special phosphor, the first white LED soon followed.
LEDs are fast becoming the best way we have of creating light. They are brighter, less power hungry and live for much, much longer than traditional solutions. It is what has made GravityLight - delivering just a tenth of a watt - feasible
GravityLight and its LED
LED testing and selection has formed a large part of our recent development work at The GravityLight Foundation. When trying to select GravityLight’s LED, there were a number of physical properties that were important to consider. One such property was its ‘forward voltage’. It’s not necessarily the focus for most developers, but for us it dictates our dynamo’s speed. The larger the LED’s forward voltage, the faster the dynamo spins. The faster the dynamo spins, the faster the bag descends to the ground. We have been searching for the LED with the lowest forward voltage on the market, with the aim of maximising drop time.
A second, more subjective aspect of LED selection has been honing in on what tone of white we should choose. It’s something like trying out various, but very similar, shades of white paint for your walls.
Most LED models will have a vast range of subtly different ‘colour temperature’ options. The differences between them in terms of cost and efficiency are trivial, so it’s up to us to discern what feels best. We’ve been doing this in a ‘dark room’ in our basement, adjacent to the workshop. Here, we to-and-fro between the various options. Superseding our own preferences are those of the end users in our target markets – primarily sub-Saharan Africa and India at launch.
It’s only in recent times that cultural variations in colour temperature preferences have been examined. As such, they’re not yet fully understood. It is believed that a culture’s aesthetic judgement is influenced by a wide range of drivers such as local geography, landscape and architecture. Some suspect a predominant factor to be climate, with light colour temperature being influenced specifically by night-time climate.
In India for example, the temperatures remain extremely hot throughout the night. Subsequently, a cooler, bluer light is preferred in the evenings. In contrast to this temperatures in sub-Saharan Africa, although sweltering throughout the day, can plummet at night, with people there preferring a warmer light with more of an orange hue. In time, with several regional assembly hubs, it may possible to vary the LED’s colour in accordance with local preferences.
Thanks to the relentless improvements in LED technology, GL02 is able to employ an LED which is over twice as efficient as that used in GL01. Meaning we have achieved a doubling of the light output simply by updating the LED!
It’s unclear where development will go in the future. What is clear is that as LEDs improve, so will the products that use them. Commentators have warned that the rate of advancement will soon plateau out as we approach some theoretical physical barriers. They have been saying the same thing for 50 years now.