MEMS fabrication demands close collaboration

Click here to enlarge image

Nov. 11, 2004 – Our company, Micralyne Inc., is a MEMS foundry and our customers range from startups to Fortune 500 firms. In most cases, they come to us with a computer model or a one-off prototype of a MEMS-based product.

They say, “Here’s one. Now we need you to make a million of them.”

Click here to enlarge image

Over 20 years, we have learned valuable lessons on how to best develop and manufacture a MEMS product. First, our customers must be comfortable with our technical know-how, systems and physical infrastructure — the price of admission for any MEMS manufacturer.

Beyond this, there are three additional factors that will have a significant impact on product development success: design for manufacturing, transfer to manufacturing, and communication.

Design for manufacturing

By using known manufacturing processes, a foundry can shorten development time and ensure higher yields during early fabrication runs.

In fact, our most successful customers take extra steps to change their designs to match our fabrication processes. While that puts the onus on Micralyne to offer well-defined and repeatable processes, the benefit is that our customers are assured their product will work right out of the gate.

In contrast, many customers will nevertheless create a design and process flow, raise money, and then approach a foundry partner. It is often difficult for them to go back to their financial backers when they find out their development timelines and cost estimates are unrealistic because they require new and unproven processes.

Transfer to manufacturing

Most, if not all, potential customers assess our capabilities in terms of both how we develop and manufacture a MEMS product.

However, we have found that the most important aspect of product development is the stage between the two, what we call transfer to manufacturing, or T2M. What happens during T2M will most often make or break the success of a new product development program.

It is during the T2M stage that a foundry and its product development customer have to answer a number of questions:

  • Will the customer’s MEMS process flow result in products that meet target market specifications? It may take several process iterations before the required specification is met, which increases costs and lengthens timelines.
  • Are the fabrication steps repeatable? Can they be completed in a quality-assured way? A manual process step that requires visual inspection by an engineer may never be transferable to high volume manufacturing.
  • Does the cost of the process meet end-product price expectations? For example, expensive process steps such as lengthy deep reactive ion etching (DRIE) may not be suitable for a target market characterized by high volumes and low margins.


It is common sense to say communication is important. However, for successful product development, it is of paramount importance — both internally and with customers.

Internal communication is critical because development engineers and manufacturing personnel have a tendency to look at problems from different perspectives. If strong communication links do not exist, solutions from one side may cause problems for the other.

I am constantly amazed at how problems are solved in formal design reviews, at the coffee machine or on the soccer field at lunch. This would be much more difficult to achieve if our development and manufacturing facilities and people were not in the same building.

Finally, another key aspect of communication is how a foundry educates and works with its customers. Product development companies are faced with high expectations and limited resources, which are constraints that a foundry must excel at managing.

We have a mandatory practice at Micralyne when we start working with a new customer. After a thorough review of the customer’s process flow and development requirements, we draw a diagram on our whiteboard with three boxes. They are titled, from left to right: development, transfer to manufacturing, and OEM manufacturing.

We then ask the customer to indicate where they think they are. We do the same.

Most of the time our mark on the whiteboard is to the left of the customer’s. We may disagree with them about how mature their product is, but the exercise leads to an important negotiation that helps ensure the resulting work plan is both feasible and meets the customer’s needs. If those uncomfortable conversations that align expectations never take place, the project is almost certainly destined for failure.

With these lessons in mind, we can confidently say we can translate prototypes into millions of parts. And the partnership created can ensure this is done within the timelines and budgets that our customers, investors and customer’s customers demand.


Easily post a comment below using your Linkedin, Twitter, Google or Facebook account. Comments won't automatically be posted to your social media accounts unless you select to share.

Leave a Reply

Your email address will not be published. Required fields are marked *


KLA-Tencor announces new defect inspection systems
07/12/2018KLA-Tencor Corporation announced two new defect inspection products at SEMICON West this week, addressing two key challenges in tool and process monit...
3D-Micromac unveils laser-based high-volume sample preparation solution for semiconductor failure analysis
07/09/2018microPREP 2.0 provides order of magnitude time and cost savings compared to traditional sample...
Leak check semiconductor process chambers quickly and reliably
02/08/2018INFICON,a manufacturer of leak test equipment, introduced the UL3000 Fab leak detector for semiconductor manufacturing maintenance teams t...