Embedded Mechatronic Systems

Analysis of Failures, Predictive Reliability

Abdelkhalak El Hami, Philippe Pougnet

• Publisher: ISTE Press - Elsevier
• Year: 2019
• eBook ISBN: 9780081019559
• Edition: 2
• Topic: Technology & Engineering
• Subtopic: Mechanical Engineering
• Index: Technology & Engineering

Mechatronics brings together computer science, mechanics and electronics. It enables us to improve the performances of embedded electronic systems by reducing their weight, volume, energy consumption and cost. Mechatronic equipment must operate without failure throughout ever-increasing service lives.The particularly severe conditions of use of embedded mechatronics cause failure mechanisms which are the source of breakdowns. Until now, these failure phenomena have not been looked at with enough depth to be able to be controlled.- Provides a statistical approach to design optimization through reliability- Presents an experimental approach for the characterization of the development of mechatronic systems in operating mode- Analyzes new tools that effect thermal, vibratory, humidity, electric and electromagnetic stresses

1

Reliability-based Design Optimization

Philippe Pougnet; Abdelkhalak El Hami

Abstract

In order to increase and maintain their businesses, mechatronics manufacturers develop innovative products and reduce product development costs. Economic constraints motivate them to reduce the duration of the testing phase and the number of prototypes and to develop simulations. Introducing innovations enables them to meet customers’ expectations and stand out from their competitors. A poor assessment of the ability of these innovative products to function properly in the conditions of use may result in nonconformities during warranty periods that negatively impact profitability. To reduce these industrial and financial risks, reliability must be incorporated into the design process.

Keywords

Accelerated testing; Correlation of digital image; Critical effect of stresses; Design reliability test; ECU reliability; Failure mechanism modeling; Predictive reliability calculations; Reliability-based design optimization; Risk assessment

This chapter describes a reliability-based design methodology for embedded mechatronic systems. The first step in this approach is to define the reliability targets, the risks of failure due to architectural innovations or new conditions of use, and then to evaluate the predictive reliability of the electronics. The FIDES reliability guide, a handbook on predictive reliability based on the laws of the physics of failure, regularly updated according to field returns, provides realistic forecasts of the conditions of use. The objectives of the following steps are to identify the potentially faulty in the life profile conditions and then to determine the distribution of the constraints causing these failures. In order to understand failure mechanisms, experimental characterizations of the effects of mechanical, thermal or electromagnetic stresses are carried out on a few prototypes, and tests are designed to provoke failures. Consecutive failure analysis helps to develop multiphysics failure models. These failure models are optimized and then validated by comparing model responses to thermal or vibratory solicitations with results. Developing metamodels capable of including the variability of the life profile loads and of fabrication enables reliability predictions. The design is then optimized by adjusting the architecture parameters that improve reliability.

1.1 Introduction

The mechatronics business sector is expanding rapidly due to the practice of embedding electronics inside mechanical devices, enabling manufacturers to reduce volume, mass and energy consumption, as well as production costs, which helps to gain market share. To develop new mechatronic systems, manufacturers need to face several challenges. They have to be competitive in terms of production costs and development lead time but must also ensure a high level of performance and correct functioning in increasingly stringent operational conditions for even longer lifetimes. Design and validation need to be firmly controlled.

Most frequently, the design of a mechatronic system is done by a tier one supplier in order to respect the set of requirements provided by the original equipment manufacturer (OEM). The specifications detail the requirements, the obligatory performances, use conditions, (operational and environmental, and storage and transport conditions) and the expected reliability objectives. These elements enable the tier one supplier to define the life profile which is the basis of the mechatronics system architecture [CRO 01].

After analyzing the required functions and constraints defined by the specifications [CRE 03], the designers draw up the functional modules that perform the high-level functions of the system. The performances detailed in the specifications are expressed as objectives to be met for each of the functional modules. These objectives are defined by measurable physical quantities and tolerances. In a mechatronic product design process, for each high-level functional module, the designers specify the necessary sub-functions, the constraints and the physical elements needed to perform these functions [SUH 01]. This process of translating the functional requirements and constraints into blocks in the physical domain is repeated at the various levels of the product architecture down to the basic elements. In a mechatronic system the basic elements are the electronic components. The functions and characteristics of electronic components and the required fabrication assembly processes are also detailed.

To meet the cost and size requirements, embedded mechatronic system architectures are designed without redundancy which means that an embedded mechatronic system may fail if only one of its components is defective.

A mechatronic system is reliable if it performs the required functions under the intended conditions of use for a specified period of time. The reliability of a mechatronic system is defined by the probability of performing high-level functions of the system for a given confidence level. For example, a mechatronic system is reliable if it is able to operate without failure for 20 years with a probability of 0.9 and a confidence rate of 80%. Another reliability objective may be product lifetime as defined by the length of time that a given proportion of products operates without failure for a given confidence level. By definition, reliability can only be assessed when the product is manufactured and used. However, it is obvious that manufacturers of embedded mechatronic systems cannot wait to manufacture systems in large series to prove their reliability.

In the electronic systems industry, designers predict reliability by applying predictive reliability guides. These approaches are based on the definition of a mission profile, i.e. on the definition of the thermo-mechanical and electrical stress cycles applied on the electronic components in use and on aging models based on experience. The major advantage of applying predictive reliability guides is in identifying the electronic components that are critical to reliability because of their high predictive failure rates. However, these predictive reliability handbooks have limitations: the latest technological developments of electronic components are not always included. Some guides like the MIL HDBK 217F are still in use while their component libraries are obsolete. If field returns are insufficiently taken into account, the predicted performance will differ enormously from the operation results.

The first s...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Preface
• 1: Reliability-based Design Optimization
• 2: Non-destructive Characterization by Spectroscopic Ellipsometry of Interfaces in Mechatronic Devices
• 3: Method of Characterizing the Electromagnetic Environment in Hyperfrequency Circuits Encapsulated Within Metallic Cavities
• 4: Metrology of Static and Dynamic Displacements and Deformations Using Full-Field Techniques
• 5: Characterization of Switching Transistors Under Electrical Overvoltage Stresses
• 6: Reliability of Radio Frequency Power Transistors to Electromagnetic and Thermal Stress
• 7: Internal Temperature Measurement of Electronic Components
• 8: Reliability Prediction of Embedded Electronic Systems: the FIDES Guide
• 9: Multi-objective Optimization in Fluid–Structure Interaction
• List of Authors
• Index