Wednesday |
|
|
| Observation Post |
Q&A with James O'Gorman, Founder and CEO, Eblana Photonics, and Ray Milano, VP Optical Technologies, Vitesse Semiconductor.Introduction Eblana Photonics, based in Dublin, Ireland, has its origins in a research program jointly established by Trinity College Dublin, the Irish National Microelectronics Research Centre and Enterprise Ireland. Eblana, founded in December 2000, was subsequently a spinoff from this project. The company has raised a total of $7.75 million in funding from investors including TLCom Capital Partners, ACT Venture Capital and Enterprise Ireland. Eblana develops and manufactures laser modules based on proprietary technology for incorporation into optical subsystems and systems addressing the broadband access, enterprise and metro networking markets. In January 2005, Eblana announced an alliance with electronics company Vitesse Semiconductor for the production of its laser diodes. Through this relationship, Vitesse will manufacture Eblana's laser products using existing integrated circuit toolsets and processes on an established four-inch indium phosphide fabrication line. The relationship is intended bring to market high quality laser devices at a price-point significantly below that possible using traditional laser fabrication techniques. What is the scope of the relationship between Eblana and Vitesse - is Vitesse simply the contract manufacturer? The basis of the relationship is the provision by Vitesse of foundry service for production of Eblana's product. The companies are also considering the possibility of conducting joint reference design work, as Vitesse manufactures a number of products in the PMD (Physical Media Device) arena. The Eblana laser could be combined with such products to offer a more complete solution for customers. The door is also open with regards to further collaborative efforts as the relationship develops in the future. Currently the extent of the agreement is not set in stone. The implication is that Vitesse will market products manufactured through the relationship - how far does the agreement go in this direction? As stated above, currently the relationship is a contract manufacturing one. However, Eblana and Vitesse have an understanding of the way in which their strategic interests overlap in certain areas. The businesses of the two companies do exhibit a strong complementary nature - Vitesse is a leading manufacturer of semiconductor chips that surround Eblana's lasers in systems and subsystems. Ultimately Vitesse produces the electronics while Eblana makes the optical elements that go into end products. Plainly there are ways in which each company can benefit the business of the other; this could be via development of joint reference designs or more broadly through the realisation of new products. As the relationship develops all possibilities will be explored with the aim of delivering benefits to both customers and shareholders. Are there milestones and timescales attached to the relationship? Yes. There is a joint milestone relating to details pertaining to the manufacturing of Eblana's product by Vitesse, primarily concerning the responsibilities of each party. In particular, technical qualification stages relating to the production of the laser device on Vitesse's manufacturing line are expected to be complete before the second half of 2005. Would it be true to say that Eblana is now looking to expand its market, whether through the Vitesse relationship or elsewhere, given that a company of Eblana's size cannot meet global demand for the devices that it develops? It should be noted that Eblana is already producing its laser devices with another contract manufacturer on a two-inch scale foundry. However, augmenting manufacturing activities with a company of Vitesse's standing and capabilities will increase the confidence of systems vendors that are using, or considering using, these devices - Eblana certainly expects to derive a benefit in this respect. In the Eblana-Vitesse press release it is implied that there exists complete compatibility between common aspects of the companies' products and processes - how much of an adjustment was necessary to achieve this compatibility? When Eblana began working with Vitesse it was on the understanding that no new processes would be introduced in the manufacture of Eblana's laser devices, Vitesse would essentially use standard processes as employed in the production of HBTs (Heterojunction Bipolar Transistors). There are certain differences within the production process, for example lasers are etched at a different depth to HBTs, therefore some adjustments to the calibration of control monitors is necessary, but the equipment used is identical. Similarly the process flow used in producing a laser device is different to that for a DHBT (Double HBT) - in laser fabrication the process flow must be tuned for the specific demands of an epitaxial structure. The fabrication of a laser is actually a more straightforward process than that for a transistor as fewer process flow steps are required, which allows use of a simpler mask set. Presumably the Eblana technology possesses unique features, how are these integrated into Vitesse's processes? The features of the Eblana laser device are either embodied in the epitaxial structure or dictated by the dimensions of the device - specific features are defined through the etching and passivation processes. In terms of the equipment used and sequencing of process steps in laser fabrication, these are the same as are used in producing any other high quality semiconductor device. Processes successfully completed in the laboratory with careful monitoring and unlimited technical support are not always easily transferred to a real world, commercial environment, has this factor been an issue for Vitesse? This is the primary reason that Eblana picked Vitesse. Eblana approached all contract manufacturing companies in the U.S., plus some outside of the U.S., that offered the capacity to take on the work of producing its laser device. Eblana had drawn up a list of core criteria to help decide which company to choose - asking questions such as could the company manufacture the product using standard processes, as mentioned previously. The overriding factor for Eblana when selecting a partner was commercial focus, specifically a proven capability with respect to the manufacture of commercial III-V microelectronics. This requirement was intended to ensure that the laser device would be produced on the commercial fabrication lines of the partner's foundry, as opposed to through the use of a separate, specialised procedure. A significant percentage of companies offering innovative technologies claim these will disrupt the industry being addressed; however, change within established industries generally occurs at a much steadier pace than these claims would suggest. Could you comment on this in the context of the Eblana-Vitesse relationship? One observation I would make is this - Intel was once a small company in Silicon Valley surrounded by orange groves! As a manufacturer operating an Indium Phosphide foundry, Vitesse has been approached by all manner of people over the past few years aiming to make use of this facility. The majority of these approaches are rejected because those companies are seeking to produce what is basically a research and development curiosity with no obvious application in the market. However, Eblana came to Vitesse asking how to develop the answer to a pertinent question in the optical space - how to manufacture its laser devices on an IC process line. In the optical industry, the typical approach to laser manufacture is to stick to the traditional technique of using a two-inch wafer fab and regrowth steps; the inevitable result is low product yield and a high price-point. Eblana brought to Vitesse a flow process conceptualised and demonstrated on a two-inch fabrication facility; Vitesse immediately realised that the process could be easily ported to a four-inch fab line with 5X steppers, combined with the use of statistical process and manufacturing controls. Such a manufacturing process delivers high yields and this is why it will be disruptive in the market place - it is a departure from the standard way of producing lasers. Is it possible to quantify the cost reductions that may be achieved over the next twelve months using this manufacturing process? Eblana's product disrupts the market in a number of ways, firstly in terms of the lower component prices that it allows the company to bring to the market and its customers. With regards to this point, Eblana will shortly be issuing an announcement concerning the pricing level of components. However, this is only one part of the story. One of the key challenges facing the optical industry today relates to the costs associated with taking a component and incorporating it into a subsystem or system. This is principally due to the fact that the industry has yet to learn the lesson from silicon - that the way to reduce the cost of complex systems is through the use of more consistent, better specified, higher quality components. The optical industry today still approaches cost reduction by use of cheap, low quality, poorly specified components. Eblana's model is based on the efficient production of high quality devices to be sold in high volumes. This approach leads to a number of additional benefits for the customer, over and above lower price. For example, Eblana's lasers' beams do not steer or wander with temperature, therefore when the module is coupled to a fibre the coupling efficiency does not vary with temperature; Eblana's lasers produce a circular beam, allowing the use of simple, cheap lenses rather than expensive aspheric lenses; the lasers are highly tolerant of light reflected back into the device in comparison with DFB lasers, this removes or reduces the cost of isolators used in the transceivers. The end result is a lower priced, higher quality component facilitating a reduced bill of materials in a system, and improved yield for the system vendor. For the markets that Eblana is targeting it is estimated that these factors together will enable cost reductions of between 100% and 200% in an optical system. Breaking down this potential cost saving, what proportion can be attributed to each factor - the technology, the size of wafer and the manufacturing process? This is an extremely complex question to answer because the reduction in costs is not a single-point pricing exercise. When a system vendor buys Eblana's product that company is able to reduce the price of its own product, which is likely to lead to higher sales and so increased production volume, which in turn means that it will be buying in components in larger volumes at a lower price, and so on - the industry enters a virtuous circle. The key point is that until the industry adopts the silicon model it is impossible to get into this virtuous circle. In the electronics world increased volume leads to improved quality, and the industry is totally dependent upon this relationship. This scenario does not represent a one-off, single price-point issue, and this is why it is disruptive. If it were a single price-point issue the result would be a gradual change over time, as was noted earlier. At the opposite extreme, if the degree of change is too great and too disruptive it becomes unmanageable and fails to take hold at all. A disruptive change that allows an industry to progress to a virtuous circle as cited above effectively pulls the whole industry into a more efficient mode of operation, as has already taken place in the microelectronics industry.  |
||||||||||||
