Recreating Failure: How Reproduction Testing Confirms Root Cause in Product and Equipment Investigations

In product failure investigations, visual inspection and materials characterization establish what happened. They do not prove why. In this whitepaper, Steven Counts examines how to convert failure hypotheses into demonstrated root cause findings with reproduction, challenge, and verification testing. Based on his experience with forensic investigations and metallurgical analysis, he shares what makes test evidence defensible in litigation, insurance, and regulatory proceedings. 

What Reproduction Testing Actually Does 

Most failure investigations stop too soon. The evidence is examined, a mechanism is identified, and the conclusion is written using language like “consistent with” or “indicative of.” That language is honest when testing has not been performed. It is not sufficient when testing is possible and the stakes are high enough to require a defensible answer. 

Reproduction testing recreates the service conditions that caused the failure in a controlled laboratory environment. The objective is specific: introduce the suspected causal condition and determine whether the observed failure results. When it does, the hypothesis becomes a demonstrated mechanism. When it does not, the investigation needs a different direction. 

Three questions drive the test design: 

• Does the suspected mechanism generate the failure, or does it only correlate with it?
• Does the failure occur under normal operating conditions, or does it require something abnormal?
• Is the root cause in the design, the materials, the manufacturing process, or the service environment?

These are not questions that materials analysis alone can answer. Testing is what separates a probable cause from a proven one. 

 

Three Types of Failure Testing 

The testing approach should follow from the hypothesis, not precede it. Choosing the wrong test type wastes time, consumes evidence, and produces data that cannot be interpreted correctly. 

Reproduction Testing 

The most direct evidentiary tool. The goal is to recreate the reported failure by introducing the suspected cause under controlled conditions. This is the test that converts a hypothesis into a finding. In contested product liability cases, appliance fire investigations, and equipment incidents, reproduction testing is typically the step that resolves the dispute. 

Challenge Testing 

Rather than reproducing a specific failure, challenge testing characterizes the margin between acceptable performance and failure. Loads, temperatures, pressures, or cycle counts are increased systematically beyond reported service conditions. The result tells you how much design margin exists and whether realistic variation in service or materials could have driven the product across the failure threshold – useful in R&D failure analysis and product liability cases where the claim is that normal use caused the failure. 

Verification Testing 

This comes after root cause is established and a corrective action is proposed. Verification testing confirms the fix works and that it does not create a new failure mode. Engineers under schedule pressure often skip this step. That is how redesigns end up failing in the field for a different reason than the original failure. 

 

What Makes a Test Defensible 

In investigations that feed into litigation, insurance claims, or regulatory proceedings, methodology is evidence. A result that cannot be explained in terms of what was controlled, what was measured, and what criteria were applied is a result that can be challenged – and it usually is. 

A defensible test plan is written before the test runs, not reconstructed after the fact. It names the hypothesis, defines the variables being controlled and the ones being deliberately varied, specifies the instrumentation and environmental conditions, and states the pass/fail criteria. Instrumentation should be calibrated to traceable standards per ISO/IEC 17025.[1] 

Footnote 1: ISO/IEC 17025: General Requirements for the Competence of Testing and Calibration Laboratories. International Organization for Standardization. Available at: https://www.iso.org/standard/66912.html 

Documentation during the test creates the record the conclusion stands on. Pre- and post-test photographs, video of dynamic failure events, and continuous data logging of temperature, resistance, load, and pressure allow another investigator to review and, if necessary, reproduce the test. When opposing experts challenge the methodology, that documentation is what survives scrutiny. 

 

Four Investigations: How Testing Resolved Each One 

The following cases illustrate how the methodology works across the most common product failure categories in forensic engineering and product liability. 

Case Study #1: Appliance Fire - Resistive Heating at a Crimped Electrical Terminal 

A kitchen appliance caught fire during normal use. Fire pattern analysis following NFPA 921[2] methodology identified the origin within the heating assembly. Internal examination found oxidation and discoloration at a crimped terminal connector on the wiring harness. Three mechanisms were possible: a loose connection, heating element failure, or an external ignition source. 

Footnote 2: NFPA 921: Guide for Fire and Explosion Investigations. National Fire Protection Association. Available at: https://www.nfpa.org/codes-and-standards/nfpa-921 

A production appliance was instrumented with thermocouples and thermal imaging. Terminal clamping force was reduced to simulate assembly process variation. Under sustained electrical load, contact resistance climbed and temperature at the terminal rose progressively. Insulation reached ignition temperature. The failure sequence matched the field unit exactly. 

The cause was not a design defect or a material problem. Insufficient clamping force during assembly allowed a connection to degrade under thermal cycling until it ignited adjacent insulation. Testing directed the corrective action to the crimp process specification, not the heating element. 

Case Study #2: Pressure Regulator - Contamination Causing Intermittent Over-Pressure 

A commercial gas regulator was blamed for an over-pressure event that damaged downstream equipment. The manufacturer disputed the finding. The diaphragm was deformed, and particulate contamination was present in the valve chamber. Materials testing confirmed the elastomer met specification. 

A new regulator on a controlled test bench showed stable regulation under clean inlet conditions. When debris matching the type found in the failed unit was introduced into the inlet stream, the valve could not fully seat. Outlet pressure spiked intermittently. The deformation pattern in the test unit matched the failed unit. 

The material did not fail. The valve failed because contamination prevented it from closing. Without the contamination simulation, that distinction – which determined liability – would not have been provable. 

Case Study #3: Lithium-Ion Battery -   Ruling Out the Alternatives 

A lithium-ion battery caught fire during charging. CT imaging of cells from the same production lot revealed internal electrode winding irregularities. Three hypotheses were on the table: internal short circuit, charger fault, and external short circuit. 

Element’s battery testing team evaluated cells under controlled overcharge, external short circuit, and mechanical deformation scenarios per IEC 62133[3] protocols. External short and overcharge produced distinct thermal signatures that did not match the field failure. Internal short circuit triggered rapid exothermic heating, cell rupture, and gas release consistent with the field unit. 

Testing eliminated two of the three hypotheses with documented evidence. In battery fire investigations, the charging system is almost always blamed first. Test data is what changes that conversation. 

Footnote 3: IEC 62133: Safety Requirements for Portable Sealed Secondary Lithium Cells and Batteries. International Electrotechnical Commission. Available at: https://www.iec.ch/homepage 

Case Study #4: EV Charging Connector - Identifying Contact Pressure as the Driver 

A Level 2 EV charging connector experienced partial melting of the polymer housing at the power contact interface. Contact resistance on the failed unit was elevated compared to new hardware. Contamination, contact pressure degradation, and current overload were all considered. 

A controlled charging test with deliberately reduced contact force reproduced the heating pattern under normal current levels. Four-wire resistance measurements confirmed elevated contact resistance consistent with the failed unit. Temperature at the interface exceeded the thermal limit of the housing material without any overcurrent condition. 

Contact force degradation was the failure mode. That answer shapes the corrective action differently than contamination or overcurrent would have, and the test data made it indisputable. 

 

The Analytical Tools Behind the Testing 

Reproduction testing is designed based on what the prior analytical work has established. The tools used in the analytical phase determine what the test needs to prove – and in what sequence. Skipping analytical work and going straight to testing means guessing at what to test. 

1. Industrial Computed Tomography (CT)

CT provides non-destructive internal examination before anything is cut. For sealed assemblies, battery cells, cast components, and electronic devices, it finds internal voids, cracks, misalignment, and foreign material that surface examination cannot reach. In battery investigations, CT can identify electrode winding anomalies and dimensional defects consistent with manufacturing variation. Because it preserves the evidence intact, CT is typically the right first step when internal defects are a possibility. Practices follow ASTM E1441.[4] 

Footnote 4: ASTM E1441: Standard Guide for Computed Tomography (CT) Imaging. ASTM International. Available at: https://www.astm.org/e1441-20.html 

2. Scanning Electron Microscopy with EDS (SEM/EDS)

SEM resolves fracture surfaces at the magnification levels where failure mechanisms become visible. Fatigue crack growth produces identifiable initiation sites and growth patterns. Overload failures show ductile tearing. Stress corrosion cracking and hydrogen embrittlement each produce distinguishable features. The interpretation follows well-established fractographic principles documented in ASM Handbook Volume 12.[5] EDS adds elemental analysis of corrosion products, surface contamination, and unexpected material compositions – often distinguishing long-term degradation from a single overload event. 

Footnote 5: ASM Handbook, Volume 12: Fractography. ASM International. Available at: https://www.asminternational.org/handbooks 

3. Metallurgical Evaluation

Cross-section examination of metallic components shows whether a part was manufactured to specification and whether material condition contributed to the failure. Grain structure per ASTM E112, hardness per ASTM E18 and E384, and inclusion content per ASTM E45 are standard evaluations. Microstructural changes caused by elevated service temperatures are readable and documentable – independent evidence of a component's thermal history that is difficult to obtain any other way. 

4. Mechanical and Functional Testing

Tensile testing per ASTM E8/E8M, fracture toughness per ASTM E399 and E1820, and fatigue crack growth per ASTM E647 establish whether a material or component met its performance specification. Functional testing evaluates product performance under service-representative conditions: current-carrying capacity, pressure cycling, vibration, and thermal cycling. These methods are most effective when they follow the analytical work and are targeted at a specific mechanism the earlier investigation identified. 

 

Why Independent Testing Carries More Weight 

Every party in a product failure investigation has a position. Internal testing conducted by or for a manufacturer, insurer, or equipment owner will be challenged on the basis of that interest, regardless of whether the methodology was sound. That is how technical evidence works in contested proceedings. 

Independent laboratories have no stake in what the results show. In multi-party disputes, that neutrality is often the only way to establish a shared factual foundation. Laboratories accredited under ISO/IEC 17025 operate under quality management systems with documented requirements for impartiality and conflict of interest management. 

Chain of custody is a requirement of forensic engineering practice and a prerequisite for admissibility of physical evidence in legal proceedings. It records the receipt, condition, handling, and disposition of evidence throughout the investigation. Evidence handling procedures should be in place before examination begins. 

Specialized capabilities matter too. Industrial CT systems, scanning electron microscopes, and mechanical test frames are significant capital investments. Independent materials testing laboratories make those capabilities accessible for a single investigation without the infrastructure overhead – and without the institutional interest that comes with using a manufacturer's internal lab.

Conclusion

Most failure investigations produce a probable cause. The ones that matter in product liability litigation, recalls, insurance subrogation, and regulatory proceedings produce a demonstrated cause. 

Reproduction testing is the step that makes that possible. When it is designed around a specific hypothesis, executed with documented methodology, and interpreted by investigators who understand what the results mean, it produces findings that hold up. A failure that has not been tested has not been fully investigated. 

For further detail on how Element approaches failure analysis investigations – including materials characterization, forensic engineering, and litigation support, see Element's failure analysis and materials testing capabilities. To learn more about Element's experience and laboratory accreditations, visit the About Element page.