A Designer’s Guide to Simulation with Finite Element Analysis

Adams, V

Published June 2008
Softback, 138 Pages

This book is an overview of methods and best-practices for mechanical design engineers and designers who are using or plan to utilize Finite Element Analysis (FEA) to validate design concepts or predict and correct product failure. The target readership is part-time users of CAD (Computer Aided Design) embedded analysis packages or those whose interface relies primarily on geometric references. This should not imply that the information presented herein is not important to users of other, more complete, analysis packages or that these users are by default any more capable of successfully simulating product performance. The assumption, right or wrong, is that with the greater investment required by purchasers of more complete tools, comes more investment in training and time to learn proper modelling practices. On the other hand, many users of design analysis tools receive little training beyond a basic software introduction, if even that. These design engineers or designers are typically reliant on whatever educational material is provided by their vendor and much of this is tool related. This book should pick up where that material left off.

Should Designers Be Doing FEA?

Articles in the media and marketing from the software vendors suggest that anyone capable of learning CAD can and should perform FE based simulation on their products. However, in some circles, primarily consisting of more full-time analysts, the concept of FEA for the masses is met with scorn and some go so far as to suggest that putting simulation in the hands of non-specialists is akin to handing a child a loaded gun. As with many things in life, reality is somewhere in the middle of these extreme viewpoints.

Addressing the first point, it is important not to lose sight of the fact that FEA is still an engineering tool, requiring engineering level insight into failure and material characteristics and engineering decisions based on the results. The results of even a properly constructed simulation are not always what they seem. As will be discussed later in this book, data reported by a finite element solution requires interpretation and this is a skill that goes beyond one’s ability to use the software. A user that denies or shirks responsibility for understanding the product, material, and/or software specific variability in any given scenario probably shouldn’t be using analysis. This is why podiatrists don’t typically perform open-heart surgery. Suffice it to say that any product development professional that is willing to learn the technology, in addition to the tool, can obtain value from simulation. Simply learning the software isn’t enough.

Is leaving FEA to the specialists the answer? There are two important reasons why the product development industry will never revert to that paradigm. First of all, in a quest to drive innovation into products and reduce time to market, initial decisions in the design process need to be validated at the earliest possible opportunity and lessons learned need to be incorporated immediately. The obvious derivative from this line of thinking is that the people making those decisions need to be enabled with tools to enable validation. Currently FE based simulation is the tool best positioned to provide this. Secondly, the swing towards designer or design engineer simulation has started and has so much momentum, it may be irreversible even if it didn’t have merit (and it does) in light of the big picture. Design engineers are already convinced they can and should be doing FEA and many feel, rightly or wrongly, that they are fairly successful with it. It would not be a trivial task to convince even the ones using it poorly that they should give it up.

Consequently, the question, “Should designers be doing FEA?” is moot. They are doing FEA and their numbers will continue to swell as additional physics such as fluid flow simulation, electro-magnetics (E/M), and manufacturing simulation tools are incorporated into the various CAD packages and the software continues to become easier to use. The price of entry, historically a barrier to widespread FEA use in the design ranks, is also dropping dramatically, even disappearing, as some CAD packages incorporate limited analysis capabilities at no charge. Many major software providers are committed to this direction and it is inconceivable that this will change.

There is an important caveat to the above conclusion. While the analysis performed by a part-time user is important, even critical, to the improvement of innovation and speed in the design process, it shouldn’t replace traditional validation. When the cost or risk of proceeding to parts or tooling is great, it is recommended that a more experienced analyst be consulted to either review the design analysis conclusions or repeat the analysis with more full-featured tools, if that adds value.  A comprehensive project report will help facilitate this “final check.”

Purpose of This Book

Consistent with their founding mission, "To promote the safe and reliable use of finite element and related technology,” NAFEMS welcomes the growing ranks of “Design Analysts.” To aid them in making every simulation meaningful, this text is targeted at filling the gaps in knowledge left by basic software training. This book will attempt to summarize the collective body of knowledge as it pertains to proper modelling methods, analysis assumption construction, and results interpretation. Realistic expectations for the value of design analysis will be set that, while possibly tempering the blind enthusiasm some get when they are first exposed to FEA, should not discourage anyone who is serious about getting it right versus getting it done.

This is not a theoretical exploration of finite elements or their mathematical basis but a practical guide to using the technology effectively. The primary focus will be on CAD-embedded analysis tools or those that exclusively rely on geometric references for meshing and boundary condition input. However, much of the material will also be relevant to design engineers using more full-featured FEA systems. Where applicable, the limitations of the current offering of CAD-driven tools will be noted so that purchasers of more complete systems can justify their decision or investigate augmenting their more cumbersome tools with analysis products that integrate better with their CAD.  Similarly, users of more limited systems can determine if they should be considering an upgrade once they understand the full potential of the technology. Based on the development of the technology over the last 5 years, it is reasonable to speculate that much of this discussion will still be applicable 5 years out. However, should a new paradigm or breakthrough technology, this author will be more than happy to offer a retraction!

Contents

1. Introduction

1.1      Should Designers Be Doing FEA?
1.2      Purpose of This Book

2. The Role of Simulation in Product Development

2.1      Applicability of Simulation At Various Stages In Design
2.2      Business Benefits of Early Simulation
2.3      Capabilities and Limitations of FEA
2.4      Defining the Scope of the Problem
2.5      Chapter Summary

3. Basic Analysis Concepts

3.1      Current State of the Technology
3.2      Engineering Prerequisites for FEA
3.3      Key Assumption Categories in FEA
3.4      Assumption Sensitivity
3.5      Chapter Summary

4. Material Properties

4.1      Properties Required for a Simulation
4.2      Stress-Strain Curves Basics
4.3      Input Properties
4.4      Failure Properties
4.5      Guidelines for Commonly Used Materials
4.6      Chapter Summary

5. Meshing

5.1      What is a Finite Element ?
5.2      Convergence
5.3      Mixing Element Types
5.4      Chapter Summary

6. Boundary Conditions

6.1      Building Blocks of Boundary Conditions
6.2      Guidelines for Determining the Model Boundary
6.3      Checking Boundary Conditions
6.4      “Unrestrained” Models
6.5      Chapter Summary

7. CAD Model Construction

7.1      Linking Design and FEA Geometry
7.2      Impact of CAD Construction on Validation Efficiency
7.3      Chapter Summary

8. Basic Solution Types & Their Limitations

8.1      Linear vs. Nonlinear
8.2      Dynamic vs. Static
8.3      Drop Test
8.4      Chapter Summary

9. Simulation Model Verification

9.1      Pre-Analysis Checks
9.2      Post-Analysis Checks
9.3      Chapter Summary

10. Closing the Loop: What Does it All Mean?

10.1         Choosing Meaningful Output Quantities and Displays
10.2         Safety Factors and FEA Results
10.3         Factoring in Sensitivity and Uncertainty
10.4         Correlation to Test
10.5         Chapter Summary

11. Introduction to Optimization

11.1         Basic Optimization Concepts
11.2         Robust Versus Accurate Optimization
11.3         Model Preparation for Optimization
11.4         Chapter Summary

12. Project Reporting

12.1         Importance of Reports as a QA tool
12.2         Documentation for Posterity
12.3         Minimum Content of a Project Report
12.4         Quick and Dirty Reporting
12.5         Chapter Summary

13. Where to Go For More Help…

13.1         Brief Discussion of the FE World Outside of CAD Software
13.2         Overview of NAFEMS Documentation
13.3         Internet Resources
13.4         Other Reference Sources
13.5         Importance of Mentoring or Access to Expert
13.6         Summary of Document