Applied Reliability for Industry 2 E-Book

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 280

Veröffentlichungsjahr: 2023

Beschreibung

Applied Reliability for Industry 2 illustrates the multidisciplinary state-of-the-art science of experimental reliability. Many experts are now convinced that reliability is not limited to statistical sciences. In fact, many different disciplines interact in order to bring a product to its highest possible level of reliability, made available through today's technologies, developments and production methods. These three books, of which this is the second, propose new methods for analyzing the lifecycle of a system, enabling us to record the development phases according to development time and levels of complexity for its integration. Experimental reliability, as advanced in Applied Reliability for Industry 2, examines all the tools and testing methods used to demonstrate the reliability of the final mechatronic system.

Leseprobe

Reliability of Multiphysical Systems Setcoordinated byAbdelkhalak El Hami

Volume 17

Applied Reliability for Industry 2

Experimental Reliability for the Automobile, Aeronautics, Defense, Medical, Marine and Space Industries

Edited by

Abdelkhalak El HamiDavid DelauxHenri Grzeskowiak

First published 2023 in Great Britain and the United States by ISTE Ltd and John Wiley & Sons, Inc.

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address:

ISTE Ltd

John Wiley & Sons, Inc.

27-37 St George’s Road

111 River Street

London SW19 4EU

Hoboken, NJ 07030

UK

USA

www.iste.co.uk

www.wiley.com

© ISTE Ltd 2023The rights of Abdelkhalak El Hami, David Delaux and Henri Grzeskowiak to be identified as the authors of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act 1988.

Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s), contributor(s) or editor(s) and do not necessarily reflect the views of ISTE Group.

Library of Congress Control Number: 2022945338

British Library Cataloguing-in-Publication DataA CIP record for this book is available from the British LibraryISBN 978-1-78630-692-0

Foreword

Predicting and then guaranteeing the reliability of an industrial system is a major challenge for manufacturers in the automotive, aeronautics and defense industries, as well as for those in the rail, telecommunications, nuclear and medical sectors, among many others. But above all, it is a major challenge for us, who use this equipment on a day to day basis, who must have absolute confidence in the information transmitted and the decisions made in real time. The increasing development of connected objects (autonomous cars, home automation, etc.) will lead to a drastic reduction of human intervention in favor of the intervention of mechatronic systems. All these systems can only be deployed if users have absolute confidence in the reliability of the equipment.

These forms of equipment are divided into two main functionalities: a hardware part mainly composed of electronic boards (coupled with mechanical systems) and a real time software part enabling the implementation of the equipment and the realization of the expected tasks.

Predicting and guaranteeing the reliability of electronic equipment is a huge and endless task. On the one hand, the number of component types used to build these boards is very high and on the other hand, the new functionalities in innovative equipment require numerous reliability and robustness tests.

Thanks to the support of the Astech and MOV’EO competitiveness clusters, the NAE aeronautics industry, the Normandy and Ile de France regions, and the Rouen and Versailles Chambers of Commerce, programs have been implemented and have produced exceptional results (AUDACE, FIRST-MFP, PISTIS, CRIOS, SYMSTECK, etc.).

Given the richness of the results of these three books that have been published, I would like to thank very warmly Abdelkhalak El Hami, David Delaux and Henri Grzeskowiak for their remarkable work in the implementation of six volumes in total as well as all the authors who have devoted many hours to shaping all their results. This has meant that this essential information does not remain the property of few but shared by the largest number of engineers, technicians, researchers and students.

The first volume is devoted to Predictive Reliability, for estimating or predicting the reliability performance of systems.

The second volume is dedicated to Experimental Reliability using tools and test methods to demonstrate the reliability of systems.

The third volume presents Operational Reliability; it verifies the reliability performance of systems in real life by way of an analysis of the field data.

I would like to thank all those involved in this program as well as the financiers (State, Regions) without whom these projects could not have succeeded and I wish that the members of the hard core set up in these programs (AUDACE, FIRST-MFP, PISTIS, CRIOS, SYMSTECK) continue their association because their skills will be essential for guaranteeing the reliability of the very many technologies which are in the process of emerging and which will equip the mechatronics systems of tomorrow.

Philippe EUDELINE

President of Normandie AeroEspace (NAE)

October 2022

Preface

March 2020 was a turning point in our lives, for industry worldwide, etc., when Covid-19 struck the world! This pandemic has highlighted the fragility of our industries while exacerbating the paradigms of development, innovation, commercial competition and the acceleration of research phases.

More than ever, we need to be agile in our American, European, Asian and even global, business plans, while at the same time cementing the use of our products in their operational use. It is clear that the challenges of combining the concepts of variability and stability, standardization and acceleration of product development are enormous. The “post-Covid” world is pushing all industries to the very limits of reliability engineering.

It is true that the word “reliability” is often used as a marketing cosmetic without any real care for its underlying scientific substantiation! Nonetheless, behind this word lies an applied science, validated by analysis, research and talented people.

Which type of the reliability approach can a designer, an engineer, a manager, use in their professional environment? Whatever the industrial field (aeronautics, space, defense, automotive, home automation, etc.), how is it possible to design a reliable product more quickly, all the while upholding safety measures?

Design methods are constantly evolving alongside a growing trend to incorporate highly innovative components into new products, even in cases where user experience (UX) feedback does not yet exist or is insufficient. The boundaries between the areas where feedback exists and the unexplored areas are continuously shifting. The “conquered” areas are those in which a sufficient feedback of user experience has been obtained, enabling us to apply one of the forms to translate or exploit this UX feedback: ranges for typical values of failure under given conditions of use, characteristic values for parameters intervening in degradation laws, etc. For the unexplored domains, it will be necessary to use validation approaches based on experimentation, and consequently, physical tests; this will be the case where one of the following criteria characterizes the product under development: new components, new design, new conditions of use, new manufacturing processes and/or the alteration thereof.

The validation actions that respond to this context of a product in rapid evolution, which we could call a “product in revolution”, are based on experimentation, and there is no alternative. Forgetting this will inevitably lead to the reality of the situation being brought to the attention of the decision-maker when the time comes at a very high cost in terms of finance and loss of brand image.

This book gives an overview of the techniques that can be implemented to move toward a mature product from the start of the product’s lifecycle:

– Chapter 1 presents the aggravated tests, which make it possible to increase the performance design margins, and for which there are great stakes to increase. This must remain a selective approach to be implemented in a circumspect way.

– Chapter 2 presents the results of a study carried out on film capacitors, which are widely used in products, and for which there is insufficient UX feedback.

– Chapter 3 presents the methodology for accelerated tests, which for at least two decades have replaced dedicated reliability testing, taking into account the costs and delays induced by the latter, and which have become incompatible with the current requirements.

– Chapter 4 presents the AFNOR NF 50-144-1 collection of six standards, describing the methodology for taking the environment into account while developing. This consideration allows for a complete maturation of the product by guiding the product development so that it is able to operate and endure the maximum values of the normal environments of the specified life profile. However, it should be noted that despite the best application of actions presented in Chapters 1, 3 and 4, experience shows that at the beginning of product exploitation, the reliability predicted by reliability engineering is not immediate. It takes time for the end user to learn how to use the product and to correctly assimilate the operating instructions, to adapt to the ergonomics of the product and, above all, to get as close as possible to the conditions of use taken into account during development and described in the specifications and high-level assembly specification of the product. It is only at the end of this learning period that the product will have the reliability predicted by reliability engineering.

– Chapter 5 presents the development of vibration specifications for powertrain components.

– Chapter 6 presents a work that takes into account the mechanical environment, and in particular, the laboratory simulation of stationary but non-Gaussian vibrations, as required by the 2019 IEC 60068-2-64 Standard.

A model can only be used in a domain validated by feedback. Calculations are a set of operations performed on symbols representing physical quantities. Simulations (or modeling) are all demonstrations whose aim is to evaluate, with the help of a computer, the expected performances for a given definition of the product, a definition generally described by a numerical and/or an analytical model, under a given environment (e.g. the evaluation of the aerodynamic performances of an aircraft at a given flight attitude; the dynamic behavior of an electronic circuit board subjected to sinusoidal vibrations, etc.). The models enable us to represent:

– Either the product itself:

- the partial or total representation of the product;

- simplifications compared to the real product;

- the type of representation (3D, 2D, 1D, etc.);

- the physical–chemical constitution of the product (materials, chemical composition);

- the discretization of the product (geometry, mesh, sampling).

– Or the phenomena to which the product is subjected to:

- the nature of interactions with the environment;

- models concerning the spatial distribution of the phenomenon (uniform, punctual, law of variation, etc.);

- models concerning the temporal evolution of the phenomenon (stationary, harmonic, transient, evolution law, initial conditions);

- the boundary conditions applied to the product; that is to say, the behavior of the product when it is subjected to these constraints;

- the behavior law and associated equations (linear, nonlinear, laminar or turbulent flow, deterministic or probabilistic, etc.).

The term “tests” includes all demonstrations based on the use of specimens of the product, more or less in conformity with the definition to be qualified, and implemented in conditions more or less representative of the operating conditions of the product.

The current practices of validation during development are made up of the back and forth between design and simulation, with the inclusion of real data or even real elements in the validation loops: this is known as “Hardware In the Loop” (HIL). For example, the car manufacturer BMW used this technique in the development of its “active steering”. They used it to simulate lane changing in a virtual vehicle, and automatically correct the wrong trajectory in real time. In the end, it saves time and money! Depending on the degree of representativeness of the specimens to be qualified, we distinguish:

– development tests using mock-ups to demonstrate the validity of a technological choice, to specify the setting of certain parameters and to detect design defects as early as possible;

– qualification tests using prototypes representative of the definition to be qualified. Depending on the rank, in the product tree, of the component concerned by the test, we distinguish:

- system tests, on the complete system or on a large part of the system,

- product tests, on one of the components of the system, whatever its rank within the product tree. According to the extent of the functional domain to be tested, we distinguish: elementary tests, concerning the verification of one or a few functional characteristics (e.g. reliability testing, the testing the drafting of control commands according to inertial characteristics, etc.).

More and more often, tests, disparaging or not, are completed or validated by a virtual phase in the laboratory. Will these virtual tests replace real testing one day? What tools do we need to carry out an accurate testing campaign? Responses from a test specialist and a software designer: “a model can only be validly applied in a field that has been validated by UX feedback”. Calculations are a set of operations performed on symbols representing physical quantities. Simulations (or modeling) are all demonstrations, whose aim is to evaluate, with the help of a computer, the expected performances for a given definition of the product, a definition generally described by a numerical and/or analytical model, under a given environment (e.g. the evaluation of aerodynamic performances).

To conclude, there is a before and an after Covid. The reliability challenges for our industries to address during this new period are significant. The global context, but also the green, sustainable and carbon-neutral goals, are the next steps in a fierce economic competition that will span the next decade. The authors hope that Applied Reliability for Industry 2 on predictive reliability will impart meaningful reflections and help to guide readers toward the best practices to encourage efficient design and effective decision-making.

This book is dedicated to all those who have lost a loved one during the Covid-19 pandemic worldwide.

October 2022

Abdelkhalak EL HAMI

David DELAUX

Henri GRZESKOWIAK

1Aggravated Testing

The idea of aggravated testing is already a few decades old and is part of the family of proactive testing: it aims to get the most out of technologies at a given moment in time, and approaching this means getting closer to the international state of the art or product excellence. This chapter describes the methodology for aggravated testing at the level of principles. Please refer to [AST 10] for a detailed description of the highly accelerated life tests (HALT) or highly accelerated stress screening (HASS) approach that performs this service, which was conducted in a laboratory. This chapter concludes with a comparison between accelerated and aggravated testing.

1.1. Introduction to aggravated (or highly accelerated) testing

An aggravated test (or a highly accelerated test; although this term is somewhat an abuse of language since it is not an accelerated test at all) is a test of short duration, during which the applied stresses are progressively increased to values much higher than the specified values (which are usually related to the conditions of use). Essentially, the objective is to explore the limits of functioning and the destruction of a product in order to push them, through appropriate actions, to the limits imposed by the technology of its constituents.

1.2. Background

The origin of aggravated testing goes back more than 40 years, when Mr. Hewlett and Mr. Packard, the founders of HP, asked their engineers to significantly improve the operational reliability of their workstations. For the old-timers who can attest to that period in time, the annual maintenance cost of workstations was often 10% of their purchase value, given the poor reliability observed in the field. The method chosen by HP engineers to remedy this situation became known as STRIFE, resulting from the contraction of the word “strengthen” with the word “life”. It consisted of applying electrical stimuli (ON/OFF cycles, variation of the clock frequency, variation of the supply voltage, etc.) or mechanical stimuli (vibrations, shocks, static solicitations, etc.), going beyond the real conditions of use. When a defect appeared during these stimuli, we tried to correct it by working depending on the case, the design, the technologies, or even the restrictions for its terms and conditions of use. Next, the testing process was resumed until the technological limits intrinsic to the product (i.e. the technologies of its components) were effectively reached. These achievements were revealed by the appearance of a large number of defects that were clearly no longer reasonably correctable, in terms of cost and time. The aggravated tests, in the current sense of the term, are closely related to STRIFE, as their main objective is to make the most of the state of the art of the technologies implemented. In this respect, they are not intended to replace other more traditional types of testing: development tests, reliability growth tests, accelerated tests, qualification tests, etc.; however, they are complementary, helping to reveal anomalies so that they can correct all assignable causes of defects (the term “assignable” here denotes the presence of a deviation from the state of the art). The indications for the realization of aggravated tests can depend on the customer (it is a requirement of Airbus, for example, with regard to its equipment suppliers) or the choice of the developer to seek to increase margins (of design or manufacture). This generally implies a prior analysis identifying the performances with insufficient margins for which there are stakes to increase.

The 2002 and 2004 ASTEFORUM and the 2003 ASTELAB were used to stage sessions dedicated to aggravated testing and for the exchange of thoughts and information on all issues related to testing.

The ASTE1 commission on “Operational Safety, Environment and Testing”, which at the same time constituted the GTR252 of the lMdR-IFDS3, was at the origin of Project 4/99 “Recommandations dans l’usage industriel des essais hautement accélérés” (“Recommendations for the Industrial Application of Highly-Accelerated Tests”) conducted under the aegis of the Institut de sûreté de fonctionnement4 and which served in particular as a basis for the development of the BNAE-RG0029 recommendation “Les essais aggravés” (“Aggravated Tests”), still available for purchase on the BNAE website.

An ad hoc group of ASTE committees published a booklet in 2006 entitled “Highly-Accelerated Environmental Stress Screening”, which is still on the ASTE sales list (also available in French5).

At the international level, the IEC 62506 Standard defines in particular the HALT and HASS approaches.

1.3. General approach

Many factors require companies to improve their product performances in view of increasing responsiveness to market demands:

– particularly demanding customers, in terms of product quality and durability;

– increased specifications from customers, especially in terms of quality and reliability;

– imperative reduction of new product development cycles to avoid being overtaken by competitors;

– a downward review of development, production and after-sales costs to remain competitive on sales prices.