July 23, 2012 — The market for refurbished semiconductor manufacturing equipment has grown steadily — estimated at $6 billion (Semico report, 2010). Fabs looking to increase capacity quickly and minimize costs commonly acquire secondary equipment.
Various companies, from refurbishers to brokers, sell secondary semiconductor equipment at a lower cost than new equipment. While finding used tools is easy, semiconductor manufacturers must consider the additional installation, refurbishment, reconfiguration, warranty, service, support, and other factors involved in using secondary equipment.
Fabs typically need a number of tools, requiring a fab to approach different vendors who handle each individual tool. Vendors may only be capable of handling part of the fab’s needs rather than a full turnkey project, necessitating a project manager at the chipmaker to oversee the process. Adhering to timelines, and ensuring quality and reliability, present significant challenges.
Flexibility creates stability and predictability
Instead of providing individual tools or services, a different model handles the fab’s needs as a single, seamless turnkey project.
In the common refurbisher/broker model, where neither party is fully integrated with the other, there is a potential for delays and cost over-runs as negotiation for core tools and configuration design issues arise. Refurbishment should be just one step of a refurbishment project, wherein a team of experts ensure quality and reliability at every stage — delivering on time, ramping to schedule and staying within budget are equally important.
The turnkey services model integrates each step as part of a process rather than a single service or product sold to a fab. This can be called a "project-process model" (as compared to providing a single step solution in which the user seeks out the supplier of the next step in the process elsewhere).
Secondary equipment options
The project-turnkey process aims to transform a complex series of tasks, each with a high degree of variability, into a predictable outcome — on time and within budget. To achieve this, the project begins with analysis of the fab’s needs, building a plan to deliver this at the highest quality, lowest risk, and lowest cost.
Once the fab’s configuration needs are understood, numerous options are available: for equipment there is anything from ‘as-is’ to Tier II process demonstration. Purchasing secondary equipment ‘as-is’ is the cheapest solution but it is also the riskiest. It is often difficult to accurately understand the condition of the equipment and if it will match the configuration needs of the fab. There are indeed no guarantees that the equipment will work. Some fabs will take this risk, as they have the in-house engineering capabilities to work with the tool.
On the other hand, a full Tier II installation eliminates the risk of the unpredictable at an additional cost. The turnkey model dictates that the project does not end at installation but rather all equipment is installed under warranty with the option of expert escalation, should the need arise.
Another way to minimize costs is for fabs to work with their equipment source to identify equipment running at other facilities, which can be de-installed, shipped and re-installed at the fab (this is known as a relocation project). From a cost and time to ramp perspective, this is normally the most effective way of getting tools in to production. However, the shutdown, documentation and de-install of the tools are critical steps in this process. It is not uncommon for fabs to purchase operational equipment only later to have it shut down incorrectly or without fingerprinting. This makes start up challenging and creates needless expense. When handling a turnkey project where tools from another facility are utilized, the best method is to shut down tool adhering to a well-documented iterative process. This helps ensure the same performance in the new location. Occasionally, relocation procedures may facilitate opportunities for repair or modification before install.
Whether its a complete Tier II turnkey approach or a blended approach with relocation tools, the aim is effective tool acquisition at minimal cost, for predictable, stable fab expansion within the company’s ramp timing.
 |
| Figure. RED Equipment warehouse. |
Buying ahead of the market
Semiconductor manufacturing equipment is expensive. Without an acquisition model, buying a tool could easily be seen as a given expense — a certain tool is needed, it costs a certain amount, and there is nothing much that can be done about it. The turnkey project-process takes a different approach. Utilizing a more flexible model, users can obtain more, at lower cost.
Turnkey equipment suppliers can hold a large inventory, and buy the tools for customers’ future needs ahead of their budget availability. Often, fab tools become available when a fab’s capital expenditures (capex) budget has not yet been approved.
Buying ahead of the market can ease both cost and availability concerns. The market for used equipment is just as dynamic as the rest of the semiconductor industry, with its own cycles. When the market is good, tools become scarce and their prices go up. Often, tools are available at a lower cost in a weak market, but chipmakers do not have budget to acquire them. A forward-thinking partnership between semiconductor manufacturer and secondary equipment supplier can manipulate the ups and downs of semiconductors to their advantage. This approach requires a very close working relationship between the turnkey provider and fab.
A close working relationship with the equipment supplier may also yield alternate procurement models to the ones the fab envisioned, in the event that the requested core tools are not available in the market; the OEM no longer exists or no longer supports the tool; or parts or consumables are highly priced. In these cases, an alternate model may be a better option. The turnkey project model takes these situations into consideration, while a ‘one tool, one service’ model does not.
Engineering services
Usually, semiconductor fabs need more than just equipment for a new line or expansion. The engineering services required are unique to each fab and project, based on maturity of technology, experience of the chipmaker’s engineering team, and fab location. However, there is often an ongoing need for post-install engineering services, such as spares, training, or technical support.
Where appropriate, an engineering team from a turnkey equipment source will suggest cost-saving options during the design phase of the project. Ongoing spare parts costs can be reduced by qualifying new sources during the ramp. For example, alternate quartz in a furnace or strip tool, or a different type of ceramic in a process chamber. Installing tools to Tier II qualification provides the fab with the opportunity to fast-track part changes into their facility, avoiding months of process requalification.
Conclusion
The turnkey provider is motivated to ensure that a fab’s expansion project happens. The single tool provider is focused on selling the tool type that they specialize in without considering the whole view.
The turnkey model sees refurbishment from the fab’s point of view rather than that of the product or service provider. Refurbishment is more than just tool selection and adjustment. Frustration and expense can be mitigated by planning ahead and seeing the whole picture — handling the refurbishment as a project rather than simply providing a single tool or service. By equipment supplier/fab partnership, the resulting purchases meet technical requirements and time constraints, and minimize risk for the fab.
Carl McMahon is VP of North America sales at RED Equipment. He holds a BS in Information Technology and Economics from The Open University, United Kingdom.
Also read: Refurbished equipment: heating up and coming of age
Visit the Semiconductors Channel of Solid State Technology!
