Ken Ibbs, director of marketing at JDSU, believes the photonics industry is reaching a tipping point
The laser industry today is at a crossroads in its history, and has an opportunity for a period of rapid growth, founded on the adoption of manufacturing and design standards and best practices applied to the core technologies driving a new generation of laser systems. At JDSU, we have described this new industry paradigm as a transition from 'design excellence' as the sole source of competitive advantage to integrating 'process excellence' into any new business model. Developing process excellence as a route to sustainable competitive advantage is facilitated by the rapid emergence of fibre lasers across a range of applications, replacing many conventional diode pumped solid state (DPSS) lasers in commercial markets.
The laser industry has evolved from a number of small, highly specialised niche suppliers to a healthy business with some well-known, world-class companies. This process of evolution, which is well documented in the business press, is generally accompanied by broad industry-wide initiatives, such as the establishing of industry groups, standards and best practices that facilitate a widespread adoption of the technology. The same has not occurred to any great extent in the laser industry. With the notable exception of eye safety standards, dating back to the early days of laser usage, there are still few design standards, optical assembly practices, or even standards of measurement that would support a broader adoption of laser-based sources and measurement systems in commercial markets. The lack of such standards is, at least in part, responsible for the perception that our industry is still at an immature stage of development despite its almost 50-year history. Most lasers remain highly complex, component-rich devices requiring skilled assembly.
I believe an opportunity exists to apply business models and technologies developed in high volume telecommunications and semiconductor industries to a new generation of lasers, leveraging more process-intensive manufacturing techniques:
- Semiconductor manufacturing technologies based on an independent foundry model; and
- Telecommunications manufacturing technologies integrating fibre assembly and pig-tailing of diodes and other devices
Semiconductor diode manufacturing
Conventional wisdom is that unlike, for example, the semiconductor industry, the laser 'market' consists of a broad range of application segments with widely differing performance and support requirements, and therefore does not have the benefits of high volume manufacturing that tend to drive the transition to process excellence.
In fact, our industry is in many ways similar to the semiconductor business in using a few core technologies that may be applied generally to a range of final form, fit, function product designs - and which depend much less on highly trained specialists in the assembly of complex products.
Today there is an opportunity to focus on such initiatives in the most basic components of almost all new lasers and systems: semiconductor laser diodes. One of the factors that contributed to driving the cost and performance of semiconductor logic devices was the emergence of independent 'foundries' supporting an array of device manufacturers that create competitive advantage through their custom design capabilities. The foundries have the wafer-scale volumes to drive highly process-driven and capital equipment intensive manufacturing; and have enabled generations of small, specialised chip design companies that have kept this industry vigorous in the face of enormous device processing costs. By contrast, in today's laser industry, every major manufacturer of laser systems has its own foundry for laser diode sources; limiting the potential for smaller start-ups to innovate at higher levels of integration, and keeping the basic costs of the diodes high. Each of the major laser companies, as well as a number of independent diode manufacturers, has the chip capacity to support the entire industry several times over. This massive redundancy in capital equipment and expertise is a serious drag on the industry's ability to innovate, as well as limiting cost reduction opportunities. I believe something similar to an independent foundry model for diode manufacturing is likely to emerge within the laser industry through extensive consolidation. We should ensure that the benefits are felt throughout the industry and do not lead to monopolistic pricing by a single system manufacturer.
Fibre laser systems
Although most new laser technologies are already based on diode pumping, conventional DPSS lasers are still component rich and require highly-skilled precision opto-mechanical assembly, resulting in products that lack the uniformity and reliability inherent in more process-intensive manufacturing. A wide dispersion in performance and reliability for any given product design has contributed to the belief that lasers are inherently 'hard to get right', and are either unpredictable or require a large performance overhead to support manufacturing Cpk values greater than 1. Efforts to narrow the manufacturing process windows to create a more predictable outcome after assembly have driven component costs ever higher. What is required is a system architecture that is more amenable to automated assembly, and robust in accepting relatively wide process windows to achieve a narrow and predictable performance result. Hence, the transition to process intensive manufacturing is accelerated by the most recent of a series of disruptive technologies driven by investment in the telecommunications industry - namely the rapid development of fibre lasers and their pump sources. Fibre laser systems not only hold out the potential for high efficiency and performance, but the more integrated 'self-aligning' designs are amenable to automated assembly.
At JDSU, we would welcome a broader industry dialogue that would encourage structural changes to the industry that foster more competition and innovation.
Please send any comments/feedback to the editor at firstname.lastname@example.org