Design for micromanufacturability

The growing market for low-cost, portable systems with complex functionality has opened up important new avenues for miniaturization technology. Unlike surface-micromachined and monolithically fabricated MEMS devices of the past, newer microsystems have grown significantly in design complexity and material heterogeneity. To address these issues, alternative production technologies are being investigated.

Microassembly is a pioneering concept that sets new paradigms for manufacturing robust, low-cost and mass-producible microsystems. This concept can be achieved by designing heterogeneous microcomponents followed by precise manipulation and assembly of these components to create complex systems.

Assembling, packaging and testing activities typically account for 85 percent of the cost of many microsystems. This is primarily due to the lack of back-end standards and general methodologies. To help reduce this cost, researchers at the Automation & Robotics Research Institute (ARRI) at the University of Texas at Arlington are focusing on concurrent microengineering and design for micromanufacturability.

Figure 1: DfM² application developed at ARRI.

The goal is to design a microsystem so that it can be assembled, packaged and tested with high yields, at low cost and in a short cycle time. Our process involves simultaneously designing the product, a fabrication process for its constituent parts and the assembly work cell. Special attention is devoted to tolerance analysis, error propagation throughout the process and analysis of the impact of errors on product performance. The creation of a rigorous mathematical framework, residing within a proprietary software tool, has allowed product designers to rapidly evaluate tradeoffs among cost, cycle time and yield.

Assembly-based manufacturing of microsystems can pose several challenges due to factors such as the lack of standards for component design and fabrication, stringent tolerance budgets, workspace constraints and surface effects due to physics scaling. Consequently, selectivity of manipulation systems, sensors, control schemes and automation can produce a large number of iterations while establishing a production cycle with an acceptable yield. This makes the micromanufacturing process extremely expensive and time-consuming.

To overcome these challenges, researchers at ARRI’s Texas Microfactory project take a holistic approach. At the root of this approach is a novel iterative computational model called Design for Micromanufacturability (DfM²). Input parameters from various manufacturing steps—such as part design, part machining, assembly-cell configuration, part fixturing, assembly automation and packaging—are assembled and processed through a collection of correlated mathematical models to create estimates of manufacturing metrics. These metrics include process yield, cycle time, production cost and device performance. Also, a 3-D renderer has been integrated with the DfM² application to offer better visibility of the device configuration and the assembler in real time. (A renderer converts 3-D wire frame models into 2-D images with 3-D photorealistic effects.)

DfM² has been instrumental in evaluating manufacturing metrics for microsystems, including a Fourier transform infrared microspectrometer and a high-resolution linear displacement sensor. Future work includes development of an artificial intelligence-based system that will accelerate a product designer’s search for optimal tradeoffs among cost, speed and product performance. µ

About the authors: Dr. Aditya Das, an ARRI researcher, and Dr. Harry Stephanou, director of ARRI, are conducting research in modular microrobotic systems and micromanufacturing. Both are with the University of Texas at Arlington. Contact Das at (817) 272 5900. Email Stephanou at stephanou@arri.uta.edu.