Wikia

MEMS Foundry Engagement Guide Wiki

Home

Talk0
2pages on
this wiki
Advertisement | Your ad here
Logo-MIG100x60

Welcome to the WikijaEdit

MEMS Foundry Engagement GuideEdit

The MEMS Foundry Engagement Guide is a “How-To” guide for companies looking for MEMS foundry services; to better connect the MEMS supply chain. This is a joint effort of MEMS Industry Group members. For more information about MEMS Industry Group, please visit [www.memsindustrygroup.org]

SectionsEdit

Background InformationEdit

MEMS Foundries vs. Semiconductor FoundriesEdit

MEMS foundries differ from traditional semiconductor foundries in the fact that MEMS foundries will ultimately tend to specialize in a certain application field. MEMS, by definition, is a generic term that covers an increasing array of micro-devices across a very broad spectrum of applications.

Because the manufacturing processes, while built on the same basic materials and tool sets, vary so widely from application to application, no single foundry can be expected to have expertise and be efficient in producing volumes at a low cost without specialization. Therefore, while a customer may find a company that advertises MEMS foundry services, in general they will have a specific specialization that may or may not be suited to the application space the customer is looking for. Understanding this can help narrow the field of candidates when selecting a MEMS foundry.

Differences between MEMS fabrication and IC fabrication Edit

MEMS fabrication is developed from IC fabrication. Many techniques and materials used in IC fabrication are reused in MEMS fabrication for the advantages of low cost, high reliability and performance. However, MEMS fabrication is still different from IC fabrication at some aspects.

  1. Unconventional Materials - MEMS fabrication involves more variety of materials. Besides the conventional materials used in IC fabrication, MEMS fabrication also use other materials. MEMS can also be made from quartz, ceramics, and polymide etc.
  2. Lack of Standard Processes - IC fabrications have converged to certain standard processes, which can be used to implement all kinds of circuit functions, while MEMS fabrication is much more customized and diversified among different applications and different foundries.  For example, some types of pressure sensors and inkjet printing nozzles are fabricated by bulk micromachining, and some airbag accelerometers and micromirror projection arrays are fabricated by surface micromachining. However, many accelerometers and mirror products now also use bulk micromachining processes.  There is no library of design rules that covers all various types of processes available for MEMS, although design rules and/or guidelines exist for specific processes tailored to certain types of devices. Access to these are normally supplied under the NDA by the respective foundry.
  3. Feature Size - Feature size of MEMS fabrication is normally larger than IC fabrication. When IC fabrication feature size has shrink to 25nm, the smallest features of MEMS devices are still in 0.5um to 1um range. This translates to cheaper mask cost than IC fabrication.
  4. Mechanical Properties - MEMS fabrication cares about the mechanical properties (such as residue stress, density, young’s modulus etc.) much more than IC fabrication. Because the purpose of MEMS fabrication is to make micromachines, we more care about their mechanical properties, especially for material forming the structures. The properties of interest include Young’s modulus, yield strength, density, residual stress and stress gradients, and long term stability of these properties. The fabrication process configuration could greatly affect these properties. For example, in a process flow using polysilicon as structure materials, the deposition temperature and annealing process of polysilicon can significantly change its density, Young’s modulus, residue stress, and stress gradient, and thus optimization of process settings to control the mechanical properties of structural materials are crucial in MEMS fabrication.
  5. Unique Unit Processes - MEMS fabrication shares many of unit processes from IC fabrication, however, it also has some unique requirements. MEMS fabrication use the same photolithography, wet and dry etch, oxidation, diffusion, LPCVD, and sputter deposition as IC fabrication. But some unit processes, such as plating, molding, and wafer bonding, are more common in MEMS than in mainstream IC fabrication. MEMS fabrication requires much deeper etch, and thicker deposition of materials (high aspect ratio) for better mechanical performance (such as high sensitivity and better signal to noise ratio). DIRE are conceived for MEMS fabrication to make deep trench, or through holes (up to 0.5mm). For some MEMS process flows, deposition of thick (up to 20-30um) polysilicon layer are crucial, while in IC fabrication, the polysilicon deposition is normally less than 1um thick. For MEMS process using SOI wafer as the starting material, the top silicon layer are normally 10-30um thick, but for SOI electronics, the top silicon thickness is much less than 0.1um.
  6. Release Process - MEMS fabrication needs release process. For MEMS devices with movable parts, a unit process is required to release the movable parts from the substrate. The sacrificial layers are removed by wet etching or dry etching in the release process. During the process of wet etching release, the liquid surface tension of etchant could drag the movable parts to the substrate and cause stiction issues. CPD (Critical Point Drying) wet etch is required for MEMS release to avoid stiction issues, or dry etch can be used to avoid stiction.
  7. Stiction - MEMS fabrication needs to consider stiction problem, and perhaps special layer to prevent stiction, Stiction is caused by strong adhesion forces between MEMS movable structures and the substrate. Once contact is made, the magnitude of these forces is sufficient to deform and attract these structures to each other or to the substrate, resulting in device failure. This type of failure is one of the dominant sources of yield loss in MEMS fabrication. The basic approaches in MEMS fabrication to prevent stiction include increasing surface roughness and/or lowering solid surface energy by coating the structure surface with low surface energy materials (for example, some kind of monolayer).
  8. Doping - For many MEMS applications, the doping of materials doesn’t require very accurate control and the doping level are normally high since the purpose of doping is only to make the material conductive. This could cause compatibility problem with IC fabrication process when two processes are integrated on the same die or share some equipments.
  9. Wafer Bonding - MEMS fabrication usually need wafer bonding to form protective caps/cavities, or wafer level packaging, or implement the integration of ASIC and MEMS transducers. Also some very thick layers or heavy mass can be implemented by wafer bonding. For example, three wafers are bonded together to form bottom, up, and movable capacitor plates for high performance accelerometer devices.
  10. Less Layers - The number of layers/masks in MEMS fabrication is usually less than IC fabrication. Some MEMS processes can be much simpler than IC fabrication. This is partially due to that the interconnection need in MEMS device is less than IC.
  11. Front/Back Side Processing - Some MEMS devices need to be processed on both front side and back side while IC fabrication is focused on one side of the wafer. For example, Pressure sensor need to etch the backside to form cavity and membrane and doping on the front side to form the piezoresistors on the membrane, and the front-backside process alignment is required.
  12. MEMS Package Stress - Thermal and mechanical properties of package are important factors to MEMS fabrication. The package stress can cause deflection and stress MEMS structures and thus change the device behavior while IC is less affected by mechanical stress. There are differences in MEMS packaging, whether wafer- or die-level.
  13. Several orders of magnitude difference in wafer volume
  14. Volume Production - Most MEMS applications are first time through the process – they may not be ready to jump into volume production right after the first prototype wafer run. In many cases, additional wafer runs may be required to refine the process and meet preliminary specifications.
  15. Qualification of Process After process and design lockdown, to transition to volume production, the process must be qualified by Statistical Quality Control & Statistical Process Control, (SQC/SPC) implementation.

Different Types of FoundriesEdit

  • Partially-Captive Foundry: These types of foundries are owned and operated (i.e. captive) by a company that is using them for product manufacturing. Their first priority is to produce parts for the parent company. When excess fab capacity is available, they are willing to accept projects from outside customers. Generally, any outside customers must be producing products that are synergistic with the business goals of the company.
  • Pure-Play Foundry: These foundries perform volume production to their customer's specifications. Customers must have fully developed designs in order to best utilize this type of foundry. Pure-play foundries will not provide any design IP to the customer, but may provide/license process IP. Generally, design services are not available.
  • Foundry with Platform Techology: These foundries possess technology platform IP (sensor architectures) and have in-house capability to customize this IP to a customer's specifications and then fabricate it in volume. Design support is available. The design IP offered by these foundries is generally not transferrable to any other foundry.
  • Prototyping Services: Companies that utilize research/academic laboratories or small production facilities to produce MEMS prototypes. Generally, only small quantities of wafers (< 25 wafers/month) may be produced, quality varies according to facility, but more process flexibility is available. Prototyping services can be a cost-effective way to develop new MEMS designs and prepare them for foundry production.

Product Development StagesEdit

The maturity of your product must be considered when choosing a foundry and before starting work at a foundry. Understanding your product's technological and business maturity is very important to the foundry, as it helps them to more accurately assess risk and pricing. Four different product development stages are described below:

  1. Proof of concept: A design/model exists but the new device has either not yet been built, or built in a research lab only. Wafers must be fabricated to prove that the basic physics of the MEMS device are sound and that a manufacturable process is indeed feasible. Usually fewer than 10 wafers need to be fabricated per batch. Chips from these wafers are used for bench-top demos only. Future wafer order quantities are totally uncertain at this stage. Prototype services companies are best suited for this work.
  2. Functioning prototype: Wafer processing is needed to produce multiple wafers of functioning prototype devices. The prototypes will be used to verify design models, start packaging work, address system integration, and test device performance. Usually 10-25 wafers need to be fabricated per batch. Multiple design-fab-test iterations will occur to further refine functioning prototypes. Future wafer order quantities are uncertain. Prototype services companies or foundries are suited for this work.
  3. Pilot product: The device design and process flow are established and stable, only small tweaks are still needed. Process characterization, control and yield improvement efforts are conducted during this stage. Sequential batches of wafers must be fabricated to refine and improve the process. Chips from these wafers support on-going design refinement, product characterization, and provide sample inventory. Future wafer order quantities are approximate and initial markets are being established through sampling. This work must be done by the foundry who will provide volume production.
  4. Mature product: Device design and process flow is fully qualified. The product is in volume production, with yield improvement as an on-going background effort. Future wafer order quantities are well understood. During this stage, developing a second source foundry may be considered.

How do you know when you are ready to approach a foundry and how to start the conversation?Edit

A self-evaluation checklistEdit

These are questions that a MEMS developer should answer to determine their readiness to approach a foundry. Depending on the number of questions that a device manufacturer/designer is able to answer, the company will be able to determine their manufacturing readiness factor.

Application and Market

  1. What is the description of your application?
  2. What is the description of your target market?
  3. What is the size of your target market?

Device Description

  1. What type of device do you want to produce?
  2. What is the package type?
  3. What is the level of necessary system integration?

Cost Targets

  1. What is the cost target per die?
  2. What is the cost target per wafer?

Volume Projections

  1. What is the volume projection in the development stage?
  2. What is the volume projection in the production stage?
  3. What is the length and slope of ramp?

Risk Evaluation

  1. What is the overall impression of feasibility?
  2. What are the specifications that a foundry must meet?
  3. What is the description of the aforementioned specification?
  4. What wafer and die yield do you expect?
  5. What are your data collection requirements in regards to:
  • SEM
  • IR Microscopy
  • Thickness
  • CD Measurements
  • Number of wafers sacrificed per lot for cross-sectioning (number of measurements per wafer)

Assesment in Product Life Cycle

  1. In what phase of the PLC is your device in regards to
  • Advanced Development
  • Development
  • Pre-production
  • Production

Timing

  1. What is the internal qualification timing of getting your device into production?
  2. What is the external qualification timing of getting your device into production?

Previous Fabrication

  1. Has the design been fabricated elsewhere?

Device Design

  1. What is the device design of each level?
  2. Do you have a cross sectional image?
  3. Are there any functional/electrical,optical, etc. specifications?

Process Flow

  1. Are you able to provide a short written outline of the process flow?

Design/Fabrication

  1. How many photomasks does your device require?
  2. What is your required delivery schedule and conditions of delivery?
  3. How many wafers are required?
  4. What wafer thickness to you require?
  5. What resistivity do you require?
  6. What is the thickness of your device?
  7. Are there any additional wafer requirements?
  8. How many designs per wafer do you require?
  9. How many die per wafer do you require?
  10. Will you supply any materials?
  11. If so, what materials?
  12. Do you have a preferred vendor for starting wafer substrates?
  13. Will you provide any pre-processed materials supplied by another foundry?
  14. Will you use another foundry for post-processing?
  15. Are you working with another foundry for 1st/2nd sourcing?

Starting a Conversation with a FoundryEdit

Information your company should you be willing to share with a foundryEdit

The following is information that you should be willing to share with a foundry after initial contact.

  1. Level of product maturity (i.e. has the device been fabricated? If yes, what risks were retired or exposed?)
  2. Device specifics (suggested materials, dimensions, dimensional tolerances, critical/non-critical design elements)
  3. Schedule expectations (e.g. milestones to meet program objectives)
  4. Expected deliverables
  5. Program phase gate success criteria

A foundry qualification checklistEdit

Please follow this link for a compilation of questions you may want to ask a foundry in order to quality it for a project.

MEMS Marketplace - Foundries and their capabilitiesEdit

Please visit http://www.memsmarketplace.com to view a list of MIG member MEMS foundries and their capabilities.

List of questions to ask a foundryEdit

  1. Where is your fab located?
  2. How do you protect IP?
  3. What is the cost for prototype run excluding wafers & masks?
  4. What is the cost to develop the product?
  5. What is the cost per wafer in production?
  6. What kind of design support do provide ? Do you have design kits ?
  7. Are you CMOS compatible?
  8. What is your process for transferring production to another fab?
  9. How flexible are you with development?
  10. What is the financial status of your company?
    1. How profitable are you?

How to write an RFQ (request for proposal)Edit

Checklist /step-by-step instructionsEdit

Navigation through the negotiation process with a foundry.

The following are items that should be


  1. Quote - Understand what is in the quote
    1. What are the recurring vs. non-recurring costs?
    2. What is amount of effort quoted?
  2. Final Product - What is the final product that is delivered to you?
  3. Payment - What is the structure of payment?
  4. Communication - Determine the structure and frequency of communication, reports, meetings, and visits
  5. Experience - Do your research on the foundry experience and stability and match the foundry’s experience specific to similar applications
  6. Feedback - Ask the foundry for feedback on complexity of the design, potential = challenges, and possible solutions to those challenges.
  7. Resources -Ask the foundry about engineering resources and dedicated team for your project.

How to match requirements to capabilitiesEdit

Before entering a supplier customer relationship it is worth to consider the following questions:

QuestionnaireEdit

Product Edit
  1. Which product type is being asked for from the customer: early stage, high value or low cost ?
  2. What is the level of the product maturity?
  3. What are the volume expectations and price projection from independent market research?
  4. What are the market specific requirements like e.g. quality level, FAR time, etc.
  5. Is there a dedicated valuation of the foundries technology to the product:

Like e.g. unique technology competence or IP (WLP, TSV , ...)

  1. Are the critical device characteristics and the corresponding design/material/process correlations known?
CustomerEdit
  1. Does the product fit to the customer?
  2. Is the customer strong enough to enter the projected market regarding
  • technical knowledge
  • market position
  • financial background
  • IP portfolio
  1. What is the MEMS comprehension level of the customer?
  2. To what extent are backup plans and iterative technology development cycles considered?
  3. Beside pricing, what other motivations does the customer have to choose your foundry?
  4. Does the foundry have similar customers to gain on synergies (e.g. similar applications)?
FoundryEdit
  1. Is there a principal technology matching?
  2. Are the technology gaps defined and manageable?
  3. If yes, are they within the scope of the foundries strategical technology roadmap?
  4. What are the costs to close technology gaps, either by new equipment set up or alternative process/module development?
  5. Is the customer aware of these costs? Who covers them?
  6. Is there enough engineering resources available for the development and design in phase?
  7. Is the foundry open to process adaptations or introduction of new processes/materials if irreconcilable difficulties occur?
  8. Is the used facility and technology setup able to achieve reasonable profits also in mid-term?
  9. Is the expected price drop balanced by gradual technology improvements?
  10. Is the foundry capable to fulfil the volume and volume expansion plans?
  11. Are there yield targets and yield improvements targets available?
  12. Is the existing level of quality adequate?
  13. Are there failure analysis procedures with sufficient response time established?

A guide to how a foundry selects customersEdit

IntroductionEdit

Foundries employ different methods and philosophies when it comes to determining whether to pursue a foundry relationship with a customer. Some foundries are interested in only technology development and R&D programs whereas others are interested in more complete supply chain management and higher volume opportunities.

Successful Foundry PhilosophyEdit

Successful foundries will be disciplined in the types of projects and customer to engage with. For example, a foundry whose business model is to engage with high volume MEMS manufacturing will be careful to not overload themselves with too many NRE only projects as these types of projects can consume significant resources and do not provide an recurring revenue at the end of the project. Additionally, due to the “one application, one process” difficulty of MEMS, no single foundry can offer to be a volume production facility for any type of MEMS that can be devised, hence most will tend to focus their efforts on a few select markets as opposed to being a general MEMS foundry. However, even high volume foundries will have some level of NRE only projects, even outside of their primary market or application focus, to keep its project portfolio diverse and be able to react to peaks and valleys within the markets they are addressing.

Whereas a foundry targeting high volume MEMS production may entertain smaller, NRE only projects, the converse cannot be necessarily true. Small niche foundries may not have the ability to scale sufficiently to address high volume markets and will focus on appropriately sized markets accordingly.

What factors a foundry uses to determine on when to engage on a project will vary; there are, however, some general guidelines that may be helpful for customers looking for a foundry to understand.

ExpectationsEdit

High fulfillment of customer requirements promises a high grade of customer satisfaction. However for development and production of MEMS devices not all technical and business related risk can be completely controlled in every stage. Therefore customer expectations and supplier performance may differ.

A MEMS foundry is generally partnering already during the product development phase and therefore must take a proactive approach to the customer-supplier relationship. The formation of a mature volume MEMS product takes up an extended time frame. Customer acquisition and implementation may end up in a costly venture, the length of development, prototyping and qualification phase is ambiguous and quantifying the return of investment at the start of a product development is rather a guess with large error marks.

The selective acquisition as well as the elaborate evaluation of potential customer is therefore laying the foundation to beneficial future business and effective growth of customer base. The following outline, considering product attributes, customer type and foundry competence may help a foundry and potential customers to accommodate the right partner with their appropriate products.

General RequirementsEdit

  • Look for customers that understand the opportunities and challenges of MEMS
    • Good understanding of the challenges associated with and initial fab effort
    • Good understanding of tolerances required for their device
    • Realistic cost expectations
    • Acknowledgement of the time required to put together a good proposal
    • Realistic schedule expectations for development and prototypes
  • Look for customers that have similar corporate/business cultures
    • Similar communication styles
    • Similar approach to problem solving
  • Look for customers that have good understanding of the maturity of their technology
  • Look for a good match of customer fab needs and fab capabilities
  • Look for a good match of forecast production volume to fab capacities
  • Look for customers who appreciate the “value add” provided by the foundry
  • Look for a good match of T’s and C’s expectations
  • Look for a good match for IP ownership expectations
  • Look for a desire to build long term relationships
Customer Expertise/ExperienceEdit

Is this a new field for the customer or is the application in a market the customer knows well? Projects with customers that have good knowledge and experience in the end application are typically less risky and easier to manage.

Necessary foundry core competence:Edit

How well does the foundry know the processes or technology involved? Is this new or do processes and know how already exist that can be applied to the customer’s project. This is important in calculating the risk profile of the project that the foundry should take into account when developing a proposal for the customer.

  • Focus on highest volume and low cost
  • Standardized MEMS process modules
  • High automation grade and batch processing
  • Fast ramp up capability, option for capacity expansion
IP PotentialEdit

A foundry considers several IP ownership solutions. Examples of potential scenarios are below:

  1. Will the foundry contribute IP during the project?
  2. Will new IP be developed during the development?
  3. Will IP conceived during development be owned by the foundry, the customer or both based on the how and when the IP was derived?

The ownership of IP is always clear when the foundry or the customer contributes pre-existing IP to the program. When IP is developed during the program, it is advised to identify conditions of ownership prior to the start of the program. It is also important to identify rights the customer needs to IP owned by the foundry that may be embedded in the final product.


Market PotentialEdit

This can be one of the most important factors depending on the business model of the MEMS foundry. Most foundries desire to move into recurring production as soon as possible. The market potential (and length of the market horizon, or time to production) is of key importance here. However, large market potentials may not fit with all foundries. Consumer markets have high potential revenues but also require significant scale (high volumes) due to significant price pressures and compressed margins, whereas specialty MEMS may have similar market size in terms of revenue but require significantly less volume but more specialized know how. Again, this must be weighed against the foundries capabilities and business model.

Supply Chain ScopeEdit

This defines who is involved in the final delivery to the customer. Is only the foundry involved, or are the others in the supply chain that the foundry must manage to deliver a turn-key solution to the customer. The more steps in the supply chain that must be managed mean more complexity and resources required from the foundry side.

Supply Chain TypeEdit

This defines what is being delivered. Foundries can deliver either untested whole wafers, wafers that have been tested but not diced, diced and tested wafers, tested/diced/packaged components, etc. The more backend processing required drives up the complexity, incurs yield and quality control costs and in general increases the risk and complexity of the project.

Product DifficultyEdit

This is an assessment whether the foundry can reuse existing process technology or new processes or designs must be created to meet the customer’s specifications. For pure-play foundries, this may not apply as much as the customer is expected to bring the design and packaging solutions to the foundry. However, if the foundry also provides MEMS design and packaging services, this will be factored into the overall project complexity.

Market Risk/Customer ExpectationEdit

This is related mostly to quality requirements for the end product and can be specified as either passive measurement (low criticality), consumer devices (medium criticality), medical (high criticality), or aerospace/military (extreme criticality). The assessment of this factor can impact pricing and resource requirements for managing the project and the supply chain over the long term.

InvestmentEdit

Does the project require the foundry to make an investment in process technology or infrastructure to meet the requirements of the customer? This will typically need to be justified by an ROI calculation, but also factors into the project complexity assessment as well.

Estimated Development BudgetEdit

This is an estimate of the total cost of the development up to production (this does not include any costs incurred during production, such as COGS), or for an NRE only project the estimate cost of the project. Projects with larger development budgets are deemed to be more complex.

Early Stage MEMS ProductsEdit

Product attributesEdit
  • Commercialization of new technologies, products or applications
  • New system concept based on new technology, material or design
  • High price, high margin, high risk, low technology readiness, low market readiness
  • Various iterative product development phases, may result in high development costs
  • Uncertain or delayed volume ramp up
  • Unique IP protection
Customer type:Edit

Startups can be the most challenging customers for a foundry to engage with. This is not meant to be disrespectful of startups. Typically the technology is leading edge and specifications are ill defined. There may or may not be working prototypes. Typically, startups with prototypes may have produced them in a university lab environment and scaling that to a volume production process can be difficult. Startups will require more resources, not only in the technology development, but also on the customer management side. It is important to understand if the customer is funded, by whom, when and by how much. How much is allocated to the MEMS development? What is the business plan to go to market? What other factors are involved in the success of the device in the market place (ASIC, packaging, clinical trials, etc). Who are the potential customers and what level have they been engaged. These are among the riskiest of projects as startups typically have tight budgets, have high failure rates and go out of business and the foundry may never get paid. Also, the likelihood of a product moving into production is generally slim.

  • University spin offs - A foundry may choose to engage in these types of programs to explore new application areas jointly as part of a University partnership or a Government Research agency. This allows access to new technology, shared risk, and potential licensing of generated technology for use in extending the foundries capabilities and customer offering.
  • Venture capital based companies
  • Fabless companies
    • Tier 2 Customers – These are typically defined as customers that may produce a niche product or does not have a large market share within a certain application space, but represents a low risk but low reward opportunity for recurring revenue. There is a lower level of resource and infrastructure commitment necessary to manage the customer relationship with this customer profile.
    • Tier 1 Customers – These represent market leading customers and are either high volume customers or very high value added customers (highly specialized technology, high margin, low volumes). For high volume Tier 1 Customers, a foundry must have the scale necessary to manage volume requirements. For high value technology customers (defense, military, aerospace), the foundry must have the necessary technical expertise to meet the highly specialized manufacturing requirements.
Necessary foundry core competence:Edit
  • Profound expertise in MEMS process development and device integration
  • High flexibility, short term process adaptations
  • Strong allocation of resources to R&D tasks
  • Open to new material, process and handling procedures
  • No financial pressure if prolonged time to market

High Value MEMS ProductEdit

Product attributes:Edit
  • High end and high price markets like space, medicine, safety, industrial or automotive (in some cases)
  • Advanced but proven technology for existing or new products or applications
  • Distinct unique selling proposition, feature, performance or quality level
  • High application specific requirements of product functionality, reliability and robustness
  • Strong IP protection
Customer type:Edit
  • Focused on specific application, strong technical background
  • Profound experience on market, pricing and product specification
  • Balanced risk / performance work plan
  • Sequential projects phases with allocated NRE
Necessary Foundry core competence:Edit
  • Production line with superior quality level and established quality control system
  • Experienced and skilled engineering team
  • Awareness of dedicated market conditions and regulations
  • Focus on high yield and defect-free production

Matured MEMS ProductEdit

Product attributes:Edit
  • Standardized and fixed technology and design
  • Technology is close to generic semiconductor processes
  • High volume market like consumer like e.g. consumer electronics or telecommunication
  • Constant cost reduction rate (up to 20% per year)
  • Base technology is more or less open, IP only on specific design or architecture
Customer type:Edit
  • Target: conventional outsourcing strategy to several suppliers
  • Venture capital based companies /spin off in late product development stage
  • Fabless OEM with major market share turning to disruptive MEMS technology
  • OEM closing down own manufacturing line


How to navigate through a processEdit

The following points should be kept in mind when navigating through a process.

  1. Comparison of process flow provided vs. the foundry’s approach, if any, that can improve the device performance or make the process more production friendly
  2. Specifics questions on process capability, as well as questions on differences in equipment or process employed by the foundry
  3. Capability on design for manufacturing
  4. Constant Communication
  • Hold a face-to-face kickoff meeting
  • For programs longer than one year, hold quarterly face-to-face meetings
  • Hold monthly face-to-face for major gate reviews or completion of major milestones
  • Hold weekly telecons
  • E-mail communication as required, don't wait for a telecon to bring up issues/concerns

MEMS Industry Group's Foundry Engagement Guide Steering CommitteeEdit

MEMS Industry Group would like to thank the following people for the initial data to populate the foundry engagement guide.

Committee Chair: Steve Wilcenski, MEMSCAP

  • Richard Allen, NIST
  • Lew Boore, Wispry
  • Joe Brown, MEMS Exchange
  • Dave Busch, SVTC Technologies
  • Troy Chase, Akustica, Inc. A Subsidiary of the Bosch Group
  • Kevin Chau, MEMStaff
  • Theodore Chi, Innovative Micro Technology (IMT)
  • Amy Duwel, Draper Laboratory
  • Alissa Fitzgerald, AM Fitzgerald and Associates
  • Buzz Hardy, MEMSCAP
  • David Harris, Plures Technologies
  • Gavin Ho, Gavin K Ho
  • Wan-Thai Hsu, Discera
  • Nicole Kerness, Maxim Integrated Products, Inc.
  • Jim Knutti, Acuity Incorporated
  • Abdul Lateef, Plasma-Therm
  • Pinyen Lin, Touch Micro-system Technology Corp.
  • Patrick Long, Maxim Integrated Products, Inc.
  • Peter Merz, MEMS Foundry Itzehoe GmbH
  • Christopher Milne, Honeywell
  • Eric Pabo, EV Group
  • Magnus Rimskog, Silex Microsystems
  • Gerold Schropfer, Coventor
  • Joe Seeger, InvenSense, Inc.
  • Mark Stark, Applied Materials, Inc.
  • Brian Stephenson, Tronics Microsystems
  • Craig Trautman, Innovative Micro Technology IMT
  • Bin Wu, Freescale Semiconductor

Latest activityEdit

Around Wikia's network

Random Wiki