SAMPE Journal - March/April 2011 - 4

Technical Director’s Corner
Dr. Scott W. Beckwith SAMPE Technical Director SAMPE Journal Technical Editor swbeckwith@aol.com

Conducting Failure Analyses – What Options Exist for Identifying Potential Causes
Whether we are talking about a materials failure, a breakdown in a manufacturing process, a design insufficiency leading to a structural failure, or a combination of the above potential events, there are a number of techniques available for conducting as failure analysis. Sometimes the chosen material for a component does not withstand its ultimate environmental conditions to which it is exposed. Sometimes these materials were never tested under reasonable design conditions. Other times it may be that the part is exposed to conditions never intended for its service environment. Often, the materials used in a manufacturing process change over time. The vendor may often move their facilities (or owners) and/or manufactures the end product with different ingredients or percentages. In any case, the road to establish potential causes should be as thorough as possible. In an initial design of plastic or composite material structures, many companies pursue – or are required – to conduct what often are defined as “Failure Mode Effects Analyses” or FMEAs. Aerospace and military end products are often required to conduct such analyses prior to entering a production phase on a product that will eventually be provided to a governmental customer where critical performance is required. The intent is to assure materials will be what they are required to be with fully defined specifications, specifications upon purchase which will be “flowed” down to every level of subtier material suppliers, and recognized certifications (or “certs”) provided upon purchases that are not just “commercial purchase orders.” If a part fails – or a process goes wrong that parts do not now satisfy contract requirements – there often are several paths that responsible manufacturing engineers and scientists follow in order to identify possible reasons and causes for such failure. One of the keys to a respectable investigation into a failure is to develop sound engineering and scientific practices such that the first possibility is not immediately chosen at the first instance and the failure investigation process halts on the very first possibility. Unfortunately, that often happens. It typically might happen if the investigation team is too narrow in their skill set and cannot cover all of the interacting areas such as: design, materials, quality, manufacturing, testing, environment, chemical, etc. A failure investigation that stops too soon can have serious effects on the continuation of production of a potentially needed part. Typically, a failure investigation will involve the establishment of several key parties with a variety of necessary skill sets. Tools such as FMEAs, Fault Tree Analysis (FTA), MIL-STD-1629A, Event Tree Analysis (ETA), Preliminary Hazards Analysis (PHA), Mean-Time-Between-Failure (MTBF), Hazards and Operability Studies (HAZOP) and others can provide an excellent data base at the start of any product design prior to production. At the same time, these same tools are available for a company, contractor, subcontractor or end-user to use when a material, process or component fails to perform. If one is serious about finding the “root cause” of a failure in any of these areas, it is important to use responsible engineering and scientific principles in defining the approach, reviewing records and carrying out serious failure investigations. The tools are there. Often potential problems or failure modes are identified prior to entering a production phase. Sometimes they are not – and that then becomes the goal of the failure investigation “team”.

SAMPE Journal, Volume 47, No. 2, March/April 2011

SAMPE Journal - March/April 2011

Table of Contents for the Digital Edition of SAMPE Journal - March/April 2011