LED lighting is a fascinating technology. It’s not like incandescent lighting – where you heat up a filament until it glows (with the inevitable result that you generate 90% heat and relatively little light). It’s not like fluorescent either china led light – where an electric current excites a mercury vapor which ultimately leads the phosphor coating on the lamp tube to glow. No, solid state light emitting diodes are a completely different technology. They are semiconductor devices that produce light as a consequence of an electric current. And unlike incandescents and fluorescents, LED lights are directional – you can focus them in one direction. And you can talk to them.

LED lighting has to do with chips, which makes the design, production, and innovation a completely different game. One of the companies that came to the playing field early was Bridgelux, and they have been innovators since the start. Among other claims to fame in the lighting world, Bridgelux was first to create a mass production process for growing LED chips on silicon substrates – LED flood light in usa a development that is now coming to market and will lead to significant cost reductions. The company was also a pioneer in developing what is know as ‘Chip-on-Board’ architecture, which serves as the basis for improving light quality and color control, lowering the production cost and time to market for a wide variety of lighting applications.

Bridgelux has now taken that approach a step further and developed what they call their Vero platform in order to further drive down the cost of LED light (measured in lumens per dollar). The Vero platform also creates increased connectivity and integration capabilities required by the coming growth in smart lighting applications.

Bridgelux then sells to original equipment manufacturers (OEMs), who in turn manufacture products that carry somebody else’s brand. So what Bridgelux does is highly leveraged, and their innovations and productivity improvements matter, as society moves to squeeze more value out of every kilowatt-hour of electricity consumed.

I recently spoke with the company’s CEO, Brad Bullington, LED street light purchaser to try and better understand this highly technical and complicated space, and to elicit a sense of where he sees the industry going. Bullington has been with Bridgelux for four years, and CEO for the last year and a half. He has taken the company through a good deal of change to get where it is today.

Bullington commented that Bridgelux was originally founded to be a capital equipment company in 2002 – they were going to create the tools to develop the wafers and chips necessary for LEDs. They filed some key patents, but eventually came to realize that it might be easier to join the LED fray rather than be a capital equipment company. They soon also realized that if you wanted to beat the incumbent lighting technologies, it was necessary, as Bullington puts it, “to cross over certain price performance levels that will allow new applications to flip to new technologies.”

Part of the challenge of LEDs is the same one that affects other illumination technologies (though here to a lesser degree): the generation of light also creates unwanted heat. Bullington commented that this was the initial technical challenge that had to be addressed.

So the company decided to add “value added packaging,” i.e. taking the chips and creating the best layouts or what is know as “array architecture,” to their portfolio. Bridgelux was particularly keen on developing the capability to create high quality light in general illumination, while resolving thermal and power management challenges presented by LEDs.

The other thing Bullington, the former CEO (Bill Watkins), and the rest of the team did was to focus on the gallium nitride (GaN) chip technology. They selected the GaN technology, specifically with 8” silicon substrates, owing to its lower cost points and ease-of-entry into the well established silicon manufacturing universe represented by Intel, Toshiba and others. The infrastructure to build silicon-based technology was already in place. The challenge was that the gallium nitride application to silicon posed specific technical hurdles.