The Heat-Affected Truth: A Comprehensive Guide to Weld Hardness Testing
Introduction: The Unseen Vulnerability
Welding is a feat of controlled destruction. It melts, mixes, and re-solidifies metal in seconds, creating a joint that can be as strong as—or stronger than—the base material. But within that joint lies a hidden landscape: zones of extreme hardness and zones of dangerous softness. A weld that looks perfect to the naked eye may harbor brittle martensite, capable of cracking under a single winter breeze. Conversely, an overly soft weld may creep and fail under sustained load.
Weld hardness testing is the primary diagnostic tool to assess this hidden landscape. Unlike tensile or fatigue testing, which require destroying a separate coupon, hardness testing can be performed directly on a weld cross-section, revealing the mechanical fingerprint of every thermal cycle the joint experienced. This article explores why weld hardness matters, how to measure it correctly, and how to interpret the results against international standards.
1. The Metallurgical Basis: Why Welds Are Hardness Mosaics
A fusion weld is not a uniform material. It consists of three distinct regions, each with a different hardness signature:
1.1 The Weld Metal (WM)
Melted filler and base metal, then rapidly solidified. Hardness depends on chemistry (carbon equivalent) and cooling rate. Under air cooling, weld metal often forms acicular ferrite—a desirable, tough microstructure of moderate hardness (typically 180–280 HV). However, with high carbon equivalent or fast cooling, it can form martensite, exceeding 400 HV.
1.2 The Heat-Affected Zone (HAZ)
The base metal that was heated below its melting point but above the transformation temperature (typically >723°C). The HAZ is a graded region:
- Coarse-grained HAZ (CGHAZ): Nearest the fusion line. Heated above 1100°C, grains grow enormously. Upon cooling, it can form brittle martensite or bainite. This is the highest hardness zone.
- Fine-grained HAZ (FGHAZ): Heated just above the transformation temperature. Grain refinement occurs, usually giving moderate hardness.
- Intercritical HAZ (ICHAZ): Heated into the ferrite-austenite two-phase region. Partial transformation creates a mixed microstructure that often becomes the softest region.
1.3 The Unaffected Base Metal (BM)
The parent material, with its original hardness (e.g., 150 HV for mild steel, 250 HV for quenched-and-tempered steel).
The challenge of weld hardness testing is to map this mosaic. A single hardness number is meaningless; the distribution of hardness tells the story.
2. Why Hardness Testing Is Mandatory for Welds
Hardness is not a direct design property, but it is a powerful indicator of other critical properties:
| Hardness Observation | Implication | Risk |
|---|---|---|
| >350 HV (approx 37 HRC) in carbon steel | Martensite or bainite present | Hydrogen-induced cold cracking (delayed cracking hours or days after welding) |
| >480 HV (approx 48 HRC) | Un-tempered martensite | Immediate quench cracking |
| >100 HV below base metal | Over-tempered or softened HAZ | Preferential deformation, stress concentration, premature failure |
| Sharp gradients (>50 HV change over 1 mm) | Abrupt microstructural transition | Strain localization, reduced fatigue life |
Thus, hardness testing serves three roles:
- Cracking risk assessment (hydrogen cracking, also called cold cracking)
- Softening verification (for heat-sensitive alloys like duplex stainless steel or 9% Ni steel)
- Heat input validation (too low input → hard, brittle; too high input → soft, weak)
3. Methods of Weld Hardness Testing
Not all hardness tests are suitable for welds. The choice depends on the zone size, required precision, and applicable standard.
3.1 Vickers Hardness (HV) – The Gold Standard
Why Vickers? Its pyramidal indenter maintains geometrically similar indentations at all loads, and it can measure very small zones. For welds, Vickers microhardness (loads from 0.1 kgf to 1 kgf) is the universally specified method.
- Procedure: A polished, etched cross-section of the weld is indented along a predefined traverse line (or grid). Indent spacing must be at least 2.5× the diagonal length to avoid interference.
- Load selection:
- 0.1–0.2 kgf: For narrow HAZs (e.g., laser or electron beam welds) or thin sheets.
- 0.5–1.0 kgf: For typical arc welds in steel (SAW, GMAW, SMAW).
- 5–10 kgf: For coarse-grained materials or when macro-hardness suffices (but resolution is lower).
3.2 Rockwell Hardness (HRC/HRB)
Limited applicability. Rockwell’s large indenter (1.6 mm sphere for HRB, diamond cone for HRC) averages over a region too large to resolve the HAZ. Only acceptable for:
- Homogeneous weld metal in large, thick sections
- Soft, non-martensitic welds (e.g., austenitic stainless steel)
- Rapid shop-floor checks where ±3 HRC is acceptable
Do not use Rockwell to evaluate HAZ hardness—the indentation will straddle multiple zones, giving a meaningless average.
3.3 Portable Leeb / Equotip (HLD)
Rebound hardness testers are convenient for field inspection of finished welds (non-destructive, on the surface). However:
- They measure only the surface hardness, not sub-surface HAZ hardness.
- Surface grinding, scale, or paint cause large errors.
- No standard accepts Leeb for weld procedure qualification; it is a screening tool only.
3.4 Knoop Hardness (HK)
For very thin welds (<0.5 mm total depth) or brittle coatings, the Knoop indenter (elongated pyramid) allows closer spacing to the fusion line. However, Knoop results do not convert accurately to Vickers; report only in HK.
4. Standards and Acceptance Criteria
Weld hardness testing is governed by industry-specific standards. The values below are typical; always consult the applicable code.
4.1 General Welding: ISO 9015 (or AWS B4.0)
- ISO 9015-1: Destructive tests on welds in metallic materials – Hardness testing – Part 1: Hardness test on arc welded joints.
- Specifies Vickers method (HV 5 to HV 0.2).
- Requires two traverse lines: one just below the surface (0.5 mm from surface) and one at mid-thickness.
- Indent spacing: typically 0.5 mm for macro, 0.2 mm for micro.
- Acceptance criteria (for unalloyed/low-alloy steel):
- Maximum individual hardness: Often 350 HV (some codes say 380 HV for <25 mm thickness).
- No more than 20% of readings above 350 HV.
- For sour service (NACE MR0175): Maximum hardness of weld and HAZ is 250 HV (22 HRC) for carbon steel, 280 HV for low-alloy steel.
4.2 Pipeline Industry: API 1104
- Appendix A (hardness testing): Mandatory for sour service pipelines.
- Maximum hardness: 250 HV (or 22 HRC) for both weld metal and HAZ.
- Test method: Vickers HV10 or HV5, indents centered on the fusion line, then 1 mm into HAZ and 1 mm into weld metal.
4.3 Pressure Vessels: ASME Section VIII, Div. 1 & 2
- Refers to ASME Section IX (welding qualifications). Hardness limits are specified by the material code (e.g., SA-516 Gr. 70).
- Typical maximum: 200 HB (~210 HV) for normalized steels, 235 HB (~245 HV) for quenched-and-tempered steels after post-weld heat treatment (PWHT).
4.4 Stainless Steels and Non-Ferrous
- Duplex stainless steel (2205): Limits are tight—280–320 HV. Below 280 HV indicates harmful phases (sigma or chi); above 320 HV indicates excessive cold work or nitride precipitation.
- Nickel alloys: No universal limit, but >350 HV suggests severe embrittlement.
- Aluminum alloys: Hardness is not used for weld qualification due to natural aging effects; use bend tests instead.
5. Practical Procedure: Step-by-Step
Performing a weld hardness traverse to ISO 9015 requires meticulous preparation.
Step 1: Sample Extraction
Cut a cross-section perpendicular to the weld length. Include weld metal, HAZ, and at least 10 mm of unaffected base metal on both sides. Avoid thermal damage from cutting (use abrasive saw with coolant, not a torch).
Step 2: Mounting and Grinding
Mount in conductive resin (to avoid edge rounding). Grind sequentially: 120 → 240 → 400 → 600 → 800 → 1200 grit silicon carbide paper. Final polish with 3 µm and 1 µm diamond suspension. The surface must be scratch-free; any remaining deformation will alter indentation geometry.
Step 3: Etching (Optional but Recommended)
Etch with 2% nital (steel) or oxalic acid (stainless). This reveals the fusion line, HAZ boundaries, and grain structure. Mark the intended traverse path directly on the etched surface with a scribe.
Step 4: Indentation Traverse
Set the microhardness tester to the required load (e.g., HV 1). Perform indents along a line perpendicular to the fusion line, typically starting 0.2 mm from the fusion line into the HAZ, then moving outward every 0.3 mm. For full characterization, run a second traverse parallel to the surface at 1 mm depth.
Step 5: Measurement and Recording
Measure both diagonals of each indent. Record HV, distance from the fusion line, and zone identification (WM, CGHAZ, FGHAZ, ICHAZ, BM).
Step 6: Reporting
Present data as a table and a hardness profile graph (distance vs. HV). Highlight the maximum hardness and its location. Note any outliers.
6. Interpretation: Reading the Hardness Profile
A weld hardness profile is a diagnostic curve. Here are four common patterns and what they mean.
Pattern A: Ideal (Low-Carbon Steel, Proper Heat Input)
- BM: ~160 HV
- HAZ: gradually rises to ~220 HV (fine pearlite/ferrite)
- WM: ~180 HV
- Interpretation: Safe. No cracking risk. No softening.
Pattern B: Hard HAZ (Potential Hydrogen Cracking)
- BM: ~180 HV
- HAZ peak (1 mm from fusion line): 380 HV
- WM: ~250 HV
- Interpretation: Martensite present. Risk of delayed cracking. Required: Preheating, low-hydrogen practice, or post-weld heat treatment (PWHT).
Pattern C: Softened HAZ (Overheating)
- BM: ~300 HV (quenched-and-tempered steel)
- HAZ: drops to 180 HV (over-tempered)
- WM: ~280 HV
- Interpretation: Joint will fail by deformation in the HAZ. Reduce heat input or change filler metal.
Pattern D: Hard Weld Metal, Soft HAZ (Dissimilar Fillers)
- BM: ~200 HV
- HAZ: ~210 HV
- WM: 420 HV (high-carbon filler or hardfacing alloy)
- Interpretation: Acceptable for wear-resistant overlays, but unacceptable for structural joints due to notch effect.
7. Common Errors and How to Avoid Them
| Error | Consequence | Prevention |
|---|---|---|
| Indent straddling the fusion line | Measurement averages two zones; meaningless | Etch first; place indent center clearly in one zone |
| Indents too close (<2× diagonal) | Work-hardening alters neighboring indents | Use standard spacing (≥2.5× diagonal) |
| Surface not perpendicular to indenter | Asymmetric indent, low hardness reading | Level specimen; use self-leveling stage |
| Grinding burns (overheating) | Tempered surface, lower hardness | Use light pressure; frequent water cooling |
| Measuring through mill scale | Extremely high scatter | Grind to bright metal before testing (for portable testers) |
| Converting Rockwell to Vickers | Non-conservative errors (up to ±10%) | Always measure in the required scale |
8. Special Cases: Non-Standard Welds
8.1 Dissimilar Metal Welds (e.g., Steel to Inconel)
Hardness gradients are expected and can be extreme (200 HV in steel to 400 HV in nickel alloy). Acceptance criteria must be jointly agreed between designer and metallurgist. Typically, no single standard applies.
8.2 Hardfacing / Cladding
Hardness is the design property (e.g., 55–60 HRC for wear resistance). Testing is performed on the surface using Rockwell C, with multiple indents averaged. However, HAZ softening beneath the cladding must also be checked via microhardness traverse.
8.3 Post-Weld Heat Treated (PWHT) Joints
PWHT reduces hardness to relieve stresses. After PWHT, maximum hardness typically drops to 200–240 HV for carbon steel. Testing is mandatory after PWHT; some codes require both “as-welded” and “post-PWHT” hardness surveys.
9. Case Study: A Failed Pipeline Girth Weld
Scenario: A 24″ API 5L X65 pipeline (sour service) failed during hydrotest. Fracture occurred through the HAZ.
Hardness traverse (HV10) results:
- Base metal: 210 HV
- Weld metal: 225 HV
- HAZ (fusion line + 1 mm): 325 HV
Analysis: The specified maximum per API 1104 (sour service) was 250 HV. The 325 HV reading indicated untempered martensite. The fracture surface showed classic hydrogen-induced cold cracking—initiated 24 hours after welding.
Corrective action: Increase preheat from 100°C to 150°C, maintain interpass temperature, and apply a low-hydrogen welding consumable. Post-weld hardness was retested at 240 HV maximum.
Conclusion: The Weld’s Truth Teller
Weld hardness testing is not a bureaucratic exercise. It is the most direct, quantifiable method to detect the two great enemies of welded structures: brittle martensite und over-softened heat-affected zones. Whether you are qualifying a new welding procedure, investigating a field failure, or inspecting a sour-service pipeline, the Vickers microhardness traverse remains the gold standard.
A successful weld is not merely crack-free to the naked eye. It is a joint where the hardness varies within safe, predictable bounds—where the HAZ is tough, not brittle, and the weld metal is strong, not unpredictable. Hardness testing gives engineers the evidence to certify that invisible truth.