How Forensic Engineers Investigate Fires, Explosions, and Combustion Events

When a fire, explosion, or combustion upset shuts down a facility, the clock starts ticking. Claims decisions, liability questions, and repair plans all depend on a clear, defensible answer to a simple question: what really happened. 

That answer lives in the evidence, the equipment, and the operating conditions. My job, as a forensic mechanical engineer, is to turn those facts into objective analysis. This article explains how forensic engineers approach these events, the types of evidence that matter most, and patterns we often see. I also share case examples that show how small oversights during commissioning, maintenance, or operation can lead to major losses.

Why Combustion Events Demand Careful Investigation

Combustion is not simply “fuel plus spark.” Real-world systems involve fuel trains and regulators, purge and interlock sequences, burners, heat exchangers, exhaust paths, and control logic. Small variances in pressure, tuning, or sequencing can change a stable flame into an unstable one. 

Explosions add another layer. A deflagration is a rapid combustion event that propagates at subsonic speed through a fuel and air mixture. A detonation moves faster than the speed of sound and produces a shock wave. The same room can see either outcome depending on mixture, confinement, and ignition timing.

Different aspects of investigation focus on different questions:

• Fire origin and cause: Where and how a fire started.
• Combustion system failure: Was the system design adequate?  If so, why did the equipment not behave as designed?
• Explosion analysis: How an ignitable mixture formed and why ignition produced overpressure.

Often, a single loss involves all three. Industry standards such as NFPA 921, fuel gas codes, and manufacturer procedures set the accepted standard of care, with our job to apply sound engineering to the facts and explain them in plain language.

The Investigation Process

When called in to investigate, our engineers follow a disciplined process designed to protect evidence, clarify facts, and produce accurate findings.

1) Scene assessment

Goal: Preserve and document the condition of structures, equipment, and controls.

Typical methods: Site walkdown, documentation of fire patterns and physical damage, high resolution photography, 3D capture where helpful, labeling of valves and switches, non-destructive imaging for damaged internals when appropriate.

Coverage and liability impact: Early documentation can confirm whether interlocks and safeties were functional or bypassed, and valves were open or closed. These details often sit at the center of fault allocation.

2) Evidence collection and testing

Goal: Secure the components that tell the story.

Typical methods: Collect failed parts, retain sections of fractured pipe or heat exchanger tubes, secure control hardware, archive controller programs and logs, preserve gas or residue samples for laboratory analysis when applicable.

Coverage and liability impact: Laboratory and fractography results help separate pre-existing flaws from overload, thermal damage from mechanical abuse, and failure from post-event breakage.

3) System and records review

Goal: Compare “as designed,” “as built,” and “as operated.”

Typical methods: Review P&IDs, equipment data sheets, burner manuals, commissioning reports, maintenance logs, and change orders. Interview operators and contractors about recent tuning, changes in utility services, or modifications.

Coverage and liability impact: System design or manufacturing defects are sometimes found. Records often reveal whether work was performed outside manufacturer specifications or without a required re-commissioning. These support subrogation against an installer, manufacturer, or service provider when appropriate.

4) Reconstruction and testing

Goal: Recreate conditions to evaluate credible failure modes.

Typical methods: Re-create startup sequences, analyze fuel and air supply variations, review purge timing and effectiveness, evaluate pressure regulator performance, and simulate stresses or temperatures where needed.

Coverage and liability impact: Demonstrations that a system could or could not reach a dangerous state under normal operation are key in allocating responsibility among design, manufacture, installation, and operation.

5) Root cause determination and reporting

Goal: Provide a clear, defensible answer to what happened and why.

Typical methods: Synthesize physical evidence, data, and engineering analysis into findings, with clear separation between facts, assumptions, and opinions.

Coverage and liability impact: A concise causal chain supports coverage decisions, informs reserve setting, and guides subrogation strategy.

First 48 hours: What to Preserve

• Lock out energy sources safely, then record the position of valves, switches, and setpoints before changes are made.
• Secure controller hardware, programs, and logs from building management or process control systems.
• Protect fracture surfaces and heat-affected areas from handling and corrosion.
• Retain commissioning reports, recent work orders, and any emails or texts related to modifications.
• Photograph burner assemblies, regulators, piping, and air intake arrangements before removal
• Document environmental and surrounding area conditions.

Common Failure Modes We See

• Design and specification: Equipment sized or tuned without accounting for real operating ranges or simultaneous loads.
• Installation and integration: Misapplied regulators, misrouted or obstructed purge air, or improper wiring or safety settings safeties.
• Operation: Running outside the manufacturer’s tuning envelope, skipping purge steps, or assuming a system is “ready” after upstream changes.
• Maintenance: Disabled interlocks, forgotten limit switches, sticking valves, compromised seals, or corroded piping.
• Materials and components: Cracked welds, fatigued fasteners, degraded gaskets, or heat exchanger tube failures.

A single loss can include more than one mode. The analysis connects the dots.

Case at a glance: Burner not re-tuned after upstream changes

System: Dual-fuel industrial burner

Condition: Manufacturer performed initial burner setup and identified the need for upstream changes before final tuning.

What went wrong: After changes were made, the installer assumed the burner was fully commissioned. It was not. Corrosion and flame impingement developed, followed by damaging fuel oil events.

What the investigation showed: The missed re-tuning step led to the failures, not a design defect in the burner.

Case at a glance: Freeze damage after well-intended airflow changes

System: Large commercial HVAC with water coils

Condition: During a public health response, more outside air was introduced, bypassing freeze protection.

What went wrong: In cold weather, coils and piping froze and ruptured.

What the investigation showed: Failures were the result of unvetted field modifications, not flaws in the original design.

Case at a glance : Explosive overpressure during startup

System: Process heater with forced draft burner and ducting

Condition: Operators attempted repeated restarts after nuisance trips.

What went wrong: Purge sequences were bypassed. Unburned fuel accumulated and ignited, damaging the ductwork.

What the investigation showed: Evidence and controller data confirmed purge was skipped, distinguishing operational issues from design.

Red Flags that Warrant Early Engineering Review

• Burners left in “temporary tune” after construction
• Gas trains, regulators, or supply pressures changed without re-commissioning
• Safeties or interlocks bypassed to get equipment online
• Frequent flame failures or short restart cycles
• Evidence of flame impingement, soot in unexpected areas, or melted seals
• Condensation from exhaust plenums or stacks
• Pops or bangs when burners shut down
• Abnormal behavior of pressure regulators or valves
• New operating requirements layered onto legacy equipment

When to Bring in a Forensic Engineer

• After any fire, explosion, or near miss involving a burner, boiler, heater, chiller, or gas handling system.
• When field changes, load growth, or code compliance work created a new operating regime.
• If nuisance trips, alarms, or performance problems preceded the loss.
• When the parties dispute commissioning completeness or control logic.
• When you need to separate equipment design issues from operation and maintenance.

Early involvement protects evidence and can shorten the path to closure.

Why Objective Analysis Matters

Attorneys need findings that withstand expert challenge. Adjusters need defensible answers that support coverage determination or subrogation. Objective analysis provides both. My approach is to explain what the hardware shows, what the data confirms, and how those facts point to specific failure modes. I then tie those modes to the actions or decisions that made the outcome probable. 

EDT backs this work with mechanical, electrical, civil, and materials expertise across the country. When a case demands deeper visualization or measurement, our teams can apply tools such as digital X-ray for non-destructive internal views, finite element analysis for stress, fluid flow, or thermodynamic evaluation, and advanced materials analysis for fracture or corrosion questions. The result is a single, coordinated answer to what happened and why.

Frequently Asked Questions

When should we call you to the site? 

As soon as the scene is safe and preserved. Early documentation of controls, valve positions, and equipment condition is often decisive.

How long does a typical analysis take? 

It depends on access, evidence condition, and the need for testing. Many matters resolve with a site inspection, records review, and a concise report. Some require lab work or reconstruction that extends the schedule.

What is the difference between origin and cause, and fault allocation? 

Origin and cause identify where and how a fire started, or a combustion upset occurred. Fault allocation assigns responsibility among design, manufacture, installation, maintenance, and operation. Both are important, and they rely on different types of evidence.

What documentation should we preserve? 

Controller logs and programs, commissioning and maintenance records, change orders, operator notes, and photographs of equipment condition. Protect fracture surfaces and sensitive components from handling and corrosion.

Do you only work on combustion systems? 

No. Our expertise spans combustion, boilers, refrigeration and HVAC, rotating equipment, electrical control systems, and gas handling systems. Many losses cross those boundaries, and cross-disciplinary experience is helpful.

Fires, explosions, and combustion events are complex, but the investigative path is clear when systematic evidence preservation and examination are employed. Objective forensic analysis provides attorneys and insurers with the clarity they need to resolve claims and disputes.

About the Author

David S. Williams, P.E., CFEI is a consulting engineer with our Seattle-Tacoma Office. Mr. Williams provides consultation in the areas of mechanical component failure/fracture analysis, combustion systems, boilers, rotating power equipment, fluid flow control and waste gas handling systems, motor control, and sheet metal stamping equipment. You may contact David for your forensic engineering needs at dwilliams@edtengineers.com or (253) 345-5187.