ISO 4136 Transverse Tensile Testing of Welded Joints

What the test does

A coupon is cut across a welded joint so that it contains parent metal at both ends, the heat-affected zone on either side, and the weld metal in the middle — three regions in series, all carrying the same load. The weld reinforcement is machined flush unless the specification says otherwise, the coupon is machined to section, and it is pulled to failure in wedge grips. The maximum force gives the tensile strength of the joint, and the position of the fracture relative to the weld centreline is identified and classified as weld metal, heat-affected zone, or parent metal.

What it measures, and why it matters

Two things, and the second is often the more important. The tensile strength is compared against the parent material's specified minimum, because the ordinary requirement on a welded joint is that it be at least as strong as the plate it joins. Where the coupon broke then tells you whether that was achieved by the weld being adequate or merely by the plate being weak. A fracture in the parent metal is the outcome a welding procedure is qualified to produce: it says the joint is not the limiting element. A fracture through the weld or the heat-affected zone is a finding however high the load was.

Specimen

The coupon spans the joint, so it contains three different materials in series — parent metal, heat-affected zone, and weld metal. The test finds the weakest.

Orientation

Transverse — across the weldThe weld sits in the middle with parent metal at each end, so all three regions carry the same load.

Reinforcement

Machined flush unless the specification says otherwiseExcess weld cap adds section and moves the failure into the parent metal artificially.

Thickness

Full plate thickness where possibleA thin slice may miss a defect that lies deeper in the joint.

Multiple specimens for thick plate

Through the thicknessA thick joint is welded in passes, and the passes are not identical.

Surface finish

Free of machining marks across the width

Mark the weld centreline before testing

With a scribeDakAfter fracture it can be genuinely hard to tell a HAZ failure from a weld-metal one, and that distinction is the result.

Where it broke IS the result. A high strength with a fracture through the weld metal is a worse outcome than a lower strength with a fracture in the parent plate, because the second says the joint is not the weak link.

Test speed

Rate

Gradual and continuous, to ISO 6892-1ISO 4136 sets no figure of its own; where only force and fracture location are wanted, any ISO 6892-1 option may be used.

Worked through

1,2 mm/min to yield, about 32 mm/min to fractureDakMethod A rates of 0,000 25 s⁻¹ and 0,006 7 s⁻¹ on the 80 mm parallel length a 20 mm wide butt weld in steel gives (Lc ≥ Ls + 60 mm). Method B's 6 to 60 MPa/s is roughly 0,14 to 1,4 mm/min on the same length.

Reported

Tensile strength and fracture location

Acceptance

Usually against the parent material's specified minimumThe joint is required to reach the plate's specification, not a separate weld figure.

Photograph the fracture

Against the scribed centrelineDak

What the machine must be capable of

Substantial force — a full-thickness plate coupon commonly needs between a hundred and six hundred kilonewtons — with wedge grips of matching capacity and enough face length to hold a thick specimen without slipping as it necks. Hydraulic grips hold their clamping force as the section reduces, which mechanical wedges do not always manage at these loads. The frame needs the daylight for a long coupon and alignment good enough that the load is genuinely axial, since a bending component would bias the fracture toward one face and confuse the location finding that is half the result.

ISO 4136 sets no rate of its own. It hands the question to ISO 6892-1, requiring only that the coupon be loaded gradually and continuously, and adds that where nothing beyond the maximum force and the fracture location is wanted — the usual case — any option ISO 6892-1 offers may be used. That is a rule rather than a number, so work it through. The parallel length of a transverse coupon is at least Ls + 60 mm, Ls being the weld width, so a 20 mm wide butt weld in steel gives 80 mm of parallel length. On that length, ISO 6892-1 Method A means 1,2 mm/min of crosshead up to yield (0,000 25 s⁻¹) and about 32 mm/min through to fracture (0,006 7 s⁻¹); Method B's elastic stress rate of 6 to 60 MPa/s works out at roughly 0,14 to 1,4 mm/min for steel of 210 GPa. Set one of those and record it — rate of testing is a required reporting field.

Grips and fixtures for this method

TJ-144

Heavy Duty Hydraulic Grips

Heavy duty hydraulic wedge grips hold a constant clamping force on a thick welded coupon as it necks, which a mechanical wedge does not always manage at these loads.

Specifications

Self-tighteningTJ-15

Universal Parallel Wedge Grips

Universal parallel wedge grips with flat inserts, where the plate thickness and failure load sit within their range.

Specifications

How the test runs

1. 01Cut coupons transverse to the weld, with parent metal at each end.
2. 02Machine the weld reinforcement flush unless the specification requires otherwise.
3. 03Machine the coupon to the specified section and check the surface finish.
4. 04Measure the cross-section and record the original area.
5. 05Scribe the weld centreline on the coupon.
6. 06Fit wedge grips with capacity and face length for the thickness.
7. 07Align and pull to failure at an ISO 6892-1 rate — about 1,2 mm/min to yield and 32 mm/min to fracture on an 80 mm parallel length.
8. 08Record the maximum force.
9. 09Identify the fracture location relative to the scribed centreline.
10. 10Classify it as weld metal, heat-affected zone or parent metal.
11. 11Report strength and location together, with the parent specification.

Calculations

What the report has to contain

• Reference to ISO 4136 and the edition
• Parent material and welding procedure identification
• Coupon orientation, dimensions and original cross-section
• Whether reinforcement was removed
• Where in the thickness the coupon was taken from
• Rate of testing
• Maximum force and tensile strength
• Fracture location, classified
• The parent material's specified minimum for comparison
• Any evidence of lack of fusion or porosity on the fracture face

What goes wrong in practice

The most consequential error is reporting a strength without the fracture location, which discards half the information and the more diagnostic half at that. Leaving the reinforcement on is next: it produces parent-metal failures that look like passes and are artefacts of geometry. Failing to scribe the centreline makes the weld-versus-heat-affected-zone distinction a matter of opinion after the event, when the regions are only millimetres apart on a rough fracture face. And taking one coupon from a thick multi-pass joint can pass a weld whose defect sits in a pass the specimen never sampled.