What this atlas is for

Find the symptom in the table and compare its location, process stage and material with the suggested checks. Preserve the original condition and readings before changing the process.

The six defects

DefectSymptomPossible causesNext checks
BurrRaised metal along a cut or sheared edgeBlade clearance, tool wear, cut parameters, material conditionFirst identify shearing or thermal cutting. Inspect the edge under magnification: compare burr side, blade clearance and tool condition only for shearing; for thermal cutting, review the actual cutting-gas setup and parameters used.
DrossResolidified metal adhering to the underside of a thermal cutCutting gas, power, speed, focus and material conditionInspect the underside along the cut and compare the pattern with the actual cut parameters used.
Angle drift / springbackFinished bend angle differs from the set angle or varies along the partSpringback, material yield strength, R/t ratio (inside bend radius R divided by sheet thickness t), deflection on long bendsMeasure at several positions along the bend line. Record the angle definition: the included angle between the flanges after unloading; keep any loaded-angle readings separate. Compare the trial with the program setting.
Cracking at bendCrack or tear on the outside of a bend; record crack direction separately from rolling directionTight radius, material temper, grain direction, edge conditionInspect the outside of the bend and edges under magnification; check radius and rolling direction against the drawing and record material condition. Record crack direction separately; these fields do not establish a cause.
Oil-canningPanel pops or flexes between supports and will not stay flatResidual stress, unbalanced nesting, cutting heat input, thin sheetLight pressure is screening only: record support points, restraints, popping location and whether restraints were released. Review nesting and cutting sequence. For surface-plate flatness evaluation and acceptance, use the inspection card below and document the required support positions, restraint state and reading reference; the pressed shape is not an acceptance result.
Weld distortionPart bows, twists or pulls out of alignment after weldingHeat input, weld sequence, restraint, fit-up gapMeasure before and after welding against the same reference and under comparable cooling and unclamping states; record those states, weld sequence, heat input and fit-up gap.

Several factors may act together. Use each suggested cause as a lead to check against actual material, setup and measurement records; compare part acceptance separately with the drawing requirement.

  • Burr: record the process, affected face/edge and tool condition. Process variables: Shearing
  • Dross: record underside location and whether it is on a straight edge, corner or hole, together with the actual cutting record. Process variables: Laser cutting
  • Angle drift: retain the positions along the bend, angle definition, loading state and program settings together. Process context: Bending
  • Cracking: retain outside/edge location, radius, material condition, rolling direction and crack direction separately; several factors may be involved and this record does not identify a root cause.
  • Oil-canning: retain support/restraint state and cutting sequence; use the inspection card below for the measurement record.
  • Weld distortion: retain readings against the same reference, sequence, fit-up gap and restraint state. Process context: Welding

Compare one change at a time

  • Quarantine the affected part/batch and record the requirement it fails; a visible feature without a requirement is not yet a quantified rejection.
  • Photograph the location and mark the process stage, material lot, thickness and program/tool revision. Keep a known acceptable comparator if one exists.
  • Choose one suspected cause and compare one variable only in a trial approved by the responsible person under the applicable process and equipment instructions. Keep geometry, support and measurement method the same; retain original and changed settings.
  • Record the conditions and results of this comparison. Whether the symptom improves or not, do not use that result alone to judge a cause established or disproved; review the comparison conditions or consider other factors before judging. Release only after checking the agreed requirement.

For bend cracking, record rolling direction and crack direction separately. Compare material condition, bend radius and their relationship to grain direction when selecting the next trial. FAA chapter 4 explains these material, radius and direction variables. [A1]

How to use the atlas

  • Name the symptom and separate observations, hypotheses and comparison results in the record.
  • Use only approved comparisons as described above; improvement alone does not prove a unique cause.
  • Record material, thickness, tooling and parameters with the defect; another batch may require a different investigation.
  • Refer unclear requirements to the customer or drawing owner.

Where to go next

Machine acceptance concerns the machine and cut-edge checks before production. The inspection card covers part flatness and Ra measurement records: document support positions, restraint state and reading reference for the required method; if these are undefined, resolve them with the drawing owner before acceptance.