Does AWS D1.1 require defect rate tracking?

AWS D1.1:2025 does not prescribe a specific defect-rate reporting format, but it does require that all rejections and repairs be documented per Clause 8. AISC Certification Appendix A requires fabricators to maintain QC records demonstrating conformance — defect trending feeds directly into that requirement. Savvy QC managers track the data because auditors ask for it.

What defect types most commonly cause high rejection rates in fab shops?

Porosity, undercut, incomplete fusion, and weld profile nonconformances account for the majority of first-pass failures in structural shops. Cracks are less frequent but carry zero tolerance and always trigger root-cause investigation. Porosity spikes often trace to shielding gas coverage failures or electrode moisture; incomplete fusion traces to travel speed, technique, or joint fit-up.

How many defects per linear foot is an acceptable benchmark?

There is no AWS D1.1 benchmark rate. Industry practice in high-performing certified shops typically targets first-pass RT/UT acceptance above 95–97%; reject rates above 5% on a specific process or welder should trigger immediate investigation. The threshold that matters most is your own trending baseline — an unexpected rise in any category is the signal.

Can defect rate data be used to justify WPS changes?

Yes, and that is exactly the intended use. If trending shows systematic incomplete fusion on a specific joint configuration, the QC manager can propose a WPS parameter adjustment (travel speed range, interpass temperature cap, wire extension) and document it as a corrective action. The revised WPS still requires proper re-authorization and, if essential variables change, requalification.

Weld Defect Rate Tracking: CWI-Driven Process Quality Improvement

How CWI and QC managers set up weld rejection rate metrics, spot recurring defect patterns, and document corrective actions for AISC audits under AWS D1.1:2025.

Weld inspection generates a lot of data — daily reports, NDE records, repair authorization logs — but most small and mid-size structural fab shops never aggregate it. The result is that the same defect type keeps appearing on the same process or the same welder's joints, but no one connects the dots until an AISC auditor or an owner's inspector asks the uncomfortable question: "What does your rejection rate look like over the last six months?"

Defect rate tracking is not bureaucratic overhead. It is a diagnostic tool that lets QC managers see process drift before it becomes a project crisis, catch welder performance issues before they compound, and build the documented corrective-action record that auditors expect. AWS D1.1:2025 does not mandate a specific tracking format, but it requires rejection and repair documentation, and AISC Certification standards require demonstration of QC effectiveness. Tracking closes that loop.

What to Track and Why

A useful defect tracking system captures five pieces of information for each rejection:

1. Defect type. Use the AWS D1.1 Clause 8 categories: crack, incomplete fusion, incomplete joint penetration, porosity, undercut, overlap, burn-through, weld profile nonconformance (overfill, underfill, convexity, concavity). These map directly to the visual acceptance criteria in Clause 8 and to the NDE acceptance tables. Consistent categorization is what makes the data useful; "misc" is not a category.

2. Detection method. Visual (VT), radiographic testing (RT), ultrasonic testing (UT), magnetic particle testing (MT), liquid penetrant testing (PT). Tracking this separately reveals where your inspection program catches defects — and whether NDE is finding things visual missed, which suggests an in-process hold-point gap.

3. Weld process and WPS number. SMAW/E7018, FCAW-G/E71T-1C-H4, SAW, GMAW. When a defect type is disproportionately concentrated on one process, the WPS parameters or electrode handling are the first investigation target. See how WPS essential variables affect process behavior for the decision framework.

4. Welder or welding operator ID. AWS D1.1 requires welder identification on production welds; that data already exists in your weld traveler or production traceability system. Linking defects to the welder who made them is not punitive — it is the fastest way to identify whether a rejection cluster is a training issue, an equipment issue, or a WPS issue.

5. Joint location. Connection type, structural member, and project. Defects clustered in one joint geometry (say, single-bevel CJP in tight-access connections) suggest a fit-up or technique problem on that configuration rather than a general process breakdown.

Simple Metrics That Drive Action

You do not need a sophisticated software system to get value from defect tracking. A shared spreadsheet with the five fields above, updated daily by the CWI, generates three useful metrics within a week:

First-pass acceptance rate (FPA%). Of all welds examined, what percentage accepted on the first inspection? Track FPA by process and by project. A dip in FPA is the leading indicator of a quality problem; it almost always precedes a project-level rejection crisis.

Defects per 100 linear feet (or per 100 joints). Normalizes rejection counts against production volume, so a busy week doesn't look better or worse than a slow week just because of volume. Target values vary by shop and project type, but a sudden 50% increase in any single defect category is a trigger for investigation regardless of the absolute count.

Repair rate by welder. Of a given welder's joints, what percentage required repair? When one welder's rate diverges significantly from the shop average on the same process, investigation — equipment check, technique observation, WPS parameter review — is warranted before the problem grows.

Connecting Defect Trends to Root Cause

The value of tracking disappears if the data doesn't drive action. When a trend emerges, the investigation follows a structured sequence:

Step 1: Isolate the defect type and location pattern. Is the porosity confined to root passes? Is the incomplete fusion appearing only at joint roots below 70 °F ambient temperature? Spatial and temporal clustering narrows the hypothesis space fast.

Step 2: Review WPS parameters against as-run parameters. Pull the production parameter logs for the implicated process and period. Are travel speeds within the WPS range? Are current and voltage within limits? Are preheat temperatures documented before arc start? Parameter drift outside WPS limits is a common cause of clusters of similar defects.

Step 3: Check consumable handling. Porosity and incomplete fusion both correlate with electrode moisture, shielding gas contamination, or flux reconditioning failures. Low-hydrogen electrode conditioning requirements are an early casualty in a busy shop — check the baking and issuance log.

Step 4: Check fit-up tolerances. Groove opening, root face, and alignment at the time of welding drive root-pass quality. Joint fit-up tolerance requirements under AWS D1.1 define the acceptable range; if as-built fit-up is not being recorded at the inspection hold point, you lose the ability to distinguish a WPS problem from a fit-up problem.

Step 5: Document the corrective action. Write a brief nonconformance report or corrective action record that states: the defect type observed, the period, the probable root cause, and what was changed to prevent recurrence. This document is what AISC auditors read when they ask whether your QC system is effective.

Corrective Action Documentation for Audits

AISC Fabricator Certification requires that the QC system include mechanisms for identifying and correcting nonconforming conditions. When an auditor reviews your defect records, they are looking for evidence of a closed loop — detection, root cause, correction, and verification that the correction worked.

The corrective action record does not need to be elaborate. A one-page form that captures defect description, investigation findings, corrective measure taken, responsible person, and target close date satisfies the requirement. What auditors flag as deficient is the absence of any documentation — a shop that repairs welds without written corrective action gives auditors nothing to verify.

Pair your defect tracking data with the weld nonconformance report process to create a traceable audit trail from rejection through repair through acceptance. When defect trending consistently feeds corrective action documentation, the audit conversation shifts from "do you have a QC system?" to "here's evidence the system is working."

When Trending Triggers a WPS Review

If corrective action at the process level — consumable handling, preheat verification, parameter monitoring — does not bring the defect rate back to baseline within a defined period (typically two to four weeks of production data), the QC manager and the person who approved the WPS need to revisit the procedure itself.

A WPS review after persistent defects should examine whether the parameter ranges are realistic for the joint configurations and positions being produced. Wide amperage or travel-speed ranges provide operator flexibility but also create opportunities for out-of-position technique to diverge from lab-tested conditions. Tightening ranges, adding position-specific parameter sets, or requiring pre-production mockup welds on a problematic joint configuration are all legitimate corrective responses documented as WPS revisions — with the engineer of record's authorization when essential variables are involved.

Tracking defect rates costs very little. Missing the pattern they reveal costs a project.

Want a WPS and welder qualification system that keeps defect traceability built in from day one? See what WPS Welding includes →