Why Did My Part Break? Testing
A broken metal part is diagnosed through fracture failure analysis. The laboratory sequence is: secure and section the sample, read the fracture surface with SEM and optical microscopy, run mechanical tests, and verify chemical composition and hardness. The findings are then consolidated into a defensible report judged against acceptance criteria. This article walks through each step.
Principle & mechanism
A fracture begins when local stress exceeds the material's resistance. The path it follows depends on microstructure and loading history. Ductile overload produces plastic deformation before separation. It leaves cup-and-cone features formed by microvoid coalescence. Brittle fracture propagates along cleavage planes or grain boundaries with little deformation. Low temperature, high strain rate, or pre-existing flaws often trigger it. Fatigue fracture is mechanistically distinct. Cyclic stress below the yield strength initiates a crack that advances incrementally, leaving beach marks and striations as a record of growth. Environmental mechanisms such as hydrogen embrittlement and stress corrosion add chemical driving forces to the mechanical ones. Correctly classifying the failure mode is the first analytical decision, because it directs every subsequent test. Overload points toward tensile and hardness verification. Fatigue points toward cyclical loading history and initiation-site examination. The analyst therefore treats each fracture mechanism as a hypothesis. Physical evidence confirms or eliminates it — never assumption.
Failure Modes and Sample Preparation
Sample handling determines whether the fracture surface remains valid evidence. Photograph the failed part in place before removal. Document its orientation, loading direction and fracture location. The mating fracture faces must not be matched or rubbed together. Protect them with clean, dry covering, and ship them without corrosive inhibitors contacting the surface. Cutting laboratory specimens requires abrasive cutting with coolant. Keep the cut well away from the fracture to avoid heat alteration. For grease or dirt contamination, use gentle solvent cleaning. Apply alkaline cleaning or ultrasonic agitation only when necessary, since aggressive methods can destroy fine fractographic detail. Sectioned metallographic mounts are ground and polished through successive abrasive grades to a mirror finish. They are then etched with an appropriate reagent to reveal grain structure, inclusions and any decarburized or transformed layer. Label each specimen against its position on the part. The guiding rule throughout preparation is minimal intervention. Every cleaning or sectioning step risks destroying the feature that identifies the root cause.
Fracture
Surface Analysis — SEM and Metallography
Read the fracture surface first at low magnification with a stereomicroscope to map the overall pattern. Then examine it with a scanning electron microscope for fine detail. Ductile dimples indicate overload. Flat facets suggest cleavage or intergranular separation. Fatigue striations, when resolvable, confirm cyclic crack growth, and their spacing offers a semi-quantitative estimate of crack advance per cycle. The initiation site is the primary target. Ratchet marks, corrosion products, inclusions or machining marks at the origin usually identify the trigger. Energy-dispersive X-ray spectroscopy attached to the SEM identifies foreign deposits, corrosion residues or inclusion chemistry at that site. Optical metallography complements fractography by examining material quality away from the crack. Grain size, banding, carbide distribution, weld penetration and decarburization depth are assessed on the etched cross-section. Microstructural abnormalities — coarse grains, untempered martensite, excessive inclusions — are compared against the specified condition. Together, surface features and subsurface structure locate where the crack started and what material condition allowed it.
Mechanical
Testing — Tensile, Impact, Fatigue
Mechanical testing checks whether the delivered material met its specified properties. It distinguishes material deficiency from design or service overstress. Tensile testing on specimens machined from sound material near the failure measures yield strength, ultimate tensile strength and elongation. These values are compared against the applicable specification. Low ductility in a nominally ductile grade points toward processing faults such as improper heat treatment or hydrogen pickup. Impact testing, typically Charpy V-notch, evaluates toughness and notch sensitivity. It is central when the fracture shows brittle features. Hardness traverses across welds or case-hardened layers reveal gradients that optical methods cannot quantify. When crack growth rate or threshold behavior is at issue, fatigue testing on sound material characterizes the stress-cycle response under controlled conditions. Such data let engineers judge whether the recorded service loads should have produced the observed life. Specimen orientation matters: properties are taken in directions matching the component's principal stress. Interpret all results against the drawing and material standard, not in isolation.
Chemical
Composition and Hardness Testing
Chemical verification establishes identity and rules out substitution, wrong grade or off-specification elements. Spark optical emission spectrometry is the routine method for bulk composition of steels and most alloys. Combustion infrared absorption covers carbon and sulfur, and inert-gas fusion covers oxygen, nitrogen and hydrogen. Hydrogen determination deserves particular attention in delayed brittle failures. Residual hydrogen from plating or pickling can embrittle high-strength steels. Where local verification is needed, such as at a weld or a segregated zone, energy-dispersive or wavelength-dispersive spectroscopy on the SEM gives semi-quantitative to quantitative local analysis. Hardness testing converts a property requirement into a rapid, minimally destructive check. Vickers or Knoop microhardness maps phases and hardened layers. Rockwell or Brinell testing verifies bulk condition. Measured hardness can also be cross-checked against tensile strength through established empirical conversions. This provides an internal consistency test between the two datasets. Composition and hardness together confirm or eliminate the material-condition branch of the failure hypothesis.
Test Reports and Acceptance Criteria
The final report consolidates all observations into a single causal chain: failure mode, initiation location, fracture mechanism, material condition and contributing factors. It should describe the part, service history, tests performed and equipment used in general terms. It should present fractographic and metallographic evidence with annotated images, and state conclusions with their supporting reasoning. Probability language is discouraged; conclusions are graded by evidence strength and stated explicitly. Acceptance criteria are drawn from the applicable product standard, drawing and purchase specification. Composition, mechanical properties and hardness results are judged pass or fail against these documents. Where results conflict, the report must explain the discrepancy rather than suppress it — for example, an anomalous toughness value traced to specimen location. Recommendations address root cause: material substitution, revised heat treatment, design change, surface treatment removal or inspection-interval adjustment. A well-constructed report allows the reader who did not perform the work to verify each conclusion independently. That is the standard a failure analysis must meet.
Frequently Asked Questions
How is the right test sequence chosen for a broken part?
Fractography comes first, because the fracture mode observed on the surface decides the follow-up direction. Ductile overload calls for tensile and hardness verification. Brittle features call for impact testing. Fatigue features call for examination of the initiation site and, where justified, fatigue testing of sound material. Composition analysis runs in parallel for any part whose grade must be confirmed.
What
criteria decide whether the test results pass or fail?
Measured composition, tensile properties and hardness are compared with the values stated in the applicable product standard, drawing and purchase specification, not with general reference tables. The report grades each result as conforming or nonconforming and links every conclusion to specific evidence, such as fractographic features or microstructural condition, so the judgment is traceable.
Which
parts and materials does this failure testing apply to?
The approach applies to metallic components of many types — castings, forgings, machined parts, fasteners and welded assemblies — across steel, aluminum, copper and titanium alloy families. Sample preparation and etching details vary with the alloy. But the sequence of fractography, metallography, mechanical and chemical verification remains the same for any broken metal part.