Using FEA to Create Derivative Designs With Successful Base Designs

This presentation was made at the 2019 NAFEMS World Congress in Quebec Canada

Resource Abstract

Tool design engineers often undergo months of trial and error to successfully develop just one size of a specific tool. When this successful design is extended to other sizes, the natural tendency is for the same idea/approach to be used for the new designs. Unfortunately, the new designs do not always perform as well as the first design, so the same trial-and-error process is repeated. What are the secrets of a good design, and how can they be developed in a more structured method that will produce repeatable results across sizes? Can a good design be quantified? An approach for discovering the secrets of a good design quantitatively based on dimensional and finite element analyses to extend one good design to others is discussed.



Dimensional analysis identifies or verifies relationships between physical quantities or features while design parameters are “scaled-up or down” from an existing design to a new design using interdependent values and units of measure; it is a tool commonly used with fluid mechanics. Using dimensional analysis in conjunction with finite element analysis (FEA) to help extend existing structure designs along with two real-world applications is discussed.



The first application involves a ceramic plug used as an isolation tool within oil/gas wells. This type of plug was successfully developed for a 2 7/8 in. size application, but extending it to a larger size (3 1/2 in.) application created issues despite extensive design modifications. Using dimensional analysis, a scaling scheme is identified and used to extend the 2 7/8 in. design to 3 1/2 in., which was confirmed to be sufficient using FEA-based virtual testing and by successfully passing physical testing.



The second application is an end ring design with petals used to hold rubber seals in place within oil/gas well downhole application ns. Because of design constraints, proportional scaling of the successful smaller-size design cannot work. For this case, dominant design parameters are determined, and the connection among parameters from related designs are derived using nonlinear scaling and fundamental mechanics.



The most important aspect of discovering the secrets of good design quantitatively is identifying fundamental mechanics of specific designs. Expressing dominant parameters for physically meaningful dimensionless numbers would enable easy extension of the features to similar designs.

Document Details

ReferenceNWC_19_316
AuthorZhong. A
LanguageEnglish
TypePresentation
Date 18th June 2019
OrganisationHalliburton Carrollton Technology Center
RegionWorld

Download


Back to Search Results