Engineering Standards and Codes in Failure Analysis

ASTM International (formerly the American Society for Testing and Materials) publishes more than 12,000 technical standards across materials, products, and test methods. For forensic engineering, the key families are those governing mechanical testing of metals (E series), materials characterisation, and examination of failed components.

• ASTM E8 / E8M: Standard Test Methods for Tension Testing of Metallic Materials. Specifies specimen geometry, grip requirements, and loading rate. A failed component tested to E8 produces yield strength and ultimate tensile strength directly comparable to the design specification.
• ASTM E92 / E384: Vickers and microhardness testing. Hardness profiles across a weld heat-affected zone, a case-hardened component, or a corroded surface are direct evidence of microstructural state at the time of failure.
• ASTM E860: Standard Practice for Examining and Testing Items That Are or May Become Involved in Products Liability Litigation. Establishes protocols for documenting, testing, and preserving physical evidence to maintain chain of custody and prevent spoilation claims.
• ASTM E1444: Standard Practice for Magnetic Particle Testing. Non-destructive examination (NDE) standard for detecting surface and near-surface flaws in ferromagnetic components before destructive sectioning.
• ASTM G standards: Corrosion testing protocols (G1, G5, G31, G48, G59 among others). Essential for establishing whether a component met its corrosion-resistance specification and whether the service environment was within the design envelope.

In a product-liability case, ASTM test results do two things: they establish whether the material met its specification (a manufacturing-defect question) and they characterise the condition of the material at the failure surface (a service-history question). An investigator who runs an ASTM test on recovered fragments and finds that tensile strength is 30% below specification has direct, documented evidence of a material defect.

The ASME Boiler and Pressure Vessel Code (BPVC) was first published in 1914 following a series of catastrophic boiler explosions, including the Grover Shoe Factory explosion in Brockton, Massachusetts in 1905, which killed 58 people and injured 150 more. Today, the ASME BPVC is referenced in the laws of all fifty US states, all Canadian provinces, and many other countries. Its twelve sections cover design, fabrication, inspection, and in-service testing of pressure equipment.

For forensic purposes, the most relevant sections are: Section I (Power Boilers), Section VIII Divisions 1, 2, and 3 (Pressure Vessels), Section IX (Welding and Brazing Qualifications), and ASME B31.3 (Process Piping). When a vessel ruptures, the investigator checks: was the vessel ASME-stamped? Were the design calculations for the specific operating temperature and pressure on file? Were the welds made by qualified welders to qualified procedures? Was in-service inspection performed on the required schedule?

NFPA 921 is not a prescriptive code (it does not say 'you must do X'). It is a guidance document, and it describes a scientific methodology for fire and explosion investigation: form a hypothesis only after collecting data, test the hypothesis against all available evidence, and revise it when new evidence contradicts it. What makes NFPA 921 distinctive is that US courts have adopted it as the methodological benchmark.

In the 2005 decision Fireman's Fund Insurance Co. v. Canon U.S.A. (394 F.3d 1054, 8th Cir. 2005), the court affirmed exclusion of expert testimony from fire investigators who failed to apply NFPA 921 reliably: their experimental testing did not replicate the conditions required to produce ignition, and their methodology did not meet the scientific standards NFPA 921 requires. The decision exemplifies how NFPA 921 functions as a Daubert reliability filter for fire investigation testimony. NFPA 921 explicitly warns against this approach. The decision exemplifies how NFPA 921 functions as a Daubert reliability filter for fire investigation testimony.

The engineering content of NFPA 921 covers heat release rates and fire growth, compartment fire dynamics and flashover, arc fault mapping for electrical fire origin determination, and the mechanics of gas and dust explosions. Engineers working on fire investigations use these sections to model whether the alleged ignition source had sufficient energy to initiate the observed fire, a quantitative check that pure origin-and-cause investigators sometimes skip.

ISO (International Organization for Standardization) publishes the international equivalents of many ASTM standards, and in jurisdictions outside North America, ISO standards are typically the ones incorporated into law. For forensic engineering the most relevant families are: ISO 6892 (tensile testing of metallic materials, the international counterpart of ASTM E8), ISO 6506 (Brinell hardness), ISO 6507 (Vickers hardness), ISO 4967 (determination of content of non-metallic inclusions in steel), ISO 148 (Charpy impact testing for fracture toughness), and ISO 9001 (quality management systems for establishing whether a manufacturer's processes were controlled).

Structural design codes set the minimum required performance for buildings, bridges, and civil infrastructure: load combinations, material partial factors, connection design, foundation requirements. For forensic investigations of structural failures, the investigator asks: what code governed this structure's design, was the structure designed to that code's requirements, and did it perform as the code predicted?

A crucial point for litigation: codes are periodically revised, and older structures are generally not required to be upgraded to meet new editions unless a material change of use or a triggering renovation occurs. The relevant standard is the one in force when the building permit was issued, not the current code. This can be significant in long-lived infrastructure: a bridge designed to a 1970s seismic code may be entirely code-compliant for its era while being far below what current standards would require.

This distinction is the one lawyers and engineers argue about most in failure litigation. When a standard is incorporated into statute or regulation by reference, violation is typically treated as negligence per se under US common law: the plaintiff does not need to prove that the requirement was reasonable, only that it was violated and that the violation caused harm. This dramatically simplifies the plaintiff's burden.

Voluntary standards are evidence of the standard of care but not conclusive proof of it. A defendant can argue that the standard was conservative, that departure from it was reasonable given site conditions, or that the standard was not universally known or adopted at the time. Conversely, a plaintiff can argue that even a voluntary standard, if widely known and widely followed, reflects what a reasonably competent engineer would have done.