Reduce residual stress in welds: UIT vs VSR comparison

Why engineers need non-thermal alternatives to PWHT

Post-weld heat treatment (PWHT) is the conventional solution for relieving residual tensile stresses in welded joints, but it is impractical or impossible for a significant share of real-world applications. Large structures cannot be furnace-treated without disassembly; temperature-sensitive alloys risk phase changes or softening; and the energy cost and cycle time of thermal treatment add significant production overhead. According to standards bodies including ISO and ASME, residual stress management is a mandatory engineering consideration for pressure vessels, structural steelwork, and fatigue-critical assemblies — yet the method is left to the engineer’s discretion where PWHT is contraindicated.

Two mechanical methods — Ultrasonic Impact Treatment (UIT) and Vibratory Stress Relief (VSR) — have accumulated substantial peer-reviewed validation across structural steel, stainless steel, and aluminium alloys. Both convert or redistribute tensile residual stresses without thermal input, preserving metallurgical integrity. A third option, low-temperature localised induction heating, addresses cases where mechanical impact is not feasible. Understanding the mechanism, validated performance range, and correct implementation parameters for each method allows engineers to select the right approach for their specific joint geometry, material, and performance requirement.

Ultrasonic Impact Treatment: mechanism, parameters, and validated performance

Ultrasonic Impact Treatment (UIT) reduces residual stress in welded joints by delivering high-frequency impacts at 20–27 kHz to the weld toe and heat-affected zone, causing localised plastic deformation and dislocation multiplication that converts tensile residual stresses into beneficial compressive stresses. The compressive stress layer introduced typically extends 0.5–2mm below the treated surface — directly targeting the stress concentration zone most responsible for fatigue crack initiation.

Validated performance across materials

Performance data across three material classes confirms UIT’s broad applicability. For 316L stainless steel, transverse and longitudinal tensile stresses near the fusion line were converted to compressive stresses, with the stress reduction mechanism confirmed via finite element method (FEM) simulation. For Q345 structural steel, the stress reduction rate reached 80–130%, and tensile strength improved by 17.2–24.3% when UIT was applied during the mid-welding phase. The most dramatic result comes from A7N01S-T5 aluminum alloy, where up to 95% stress elimination was recorded — significantly outperforming mechanical vibration at 35%.

Implementation parameters

Correct implementation requires operating at 20–27 kHz, treating the weld toe plus 5–10mm of the HAZ on each side, and applying the treatment for a material-dependent duration (typically 80–100 seconds per zone, with longer times showing diminishing returns). Coverage must be complete — incomplete coverage leaves stress concentration boundaries that can become fatigue initiation sites.

The practical implication is that before production use, engineers should establish a material-specific force-elimination rate curve by testing 2–3 force levels on representative coupons, then apply the optimised force and verify stress conversion via surface stress measurement. Residual stress should be measured before and after treatment using X-ray diffraction or the hole-drilling method.

Vibratory Stress Relief: frequency, timing, and through-thickness redistribution

Vibratory Stress Relief (VSR) reduces residual stress in welded joints by applying controlled mechanical vibration — typically at sub-resonant or near-resonant frequencies — during or immediately after welding, causing localised micro-yielding in high-stress regions that allows stress redistribution through time-dependent plastic flow. Unlike UIT, which acts on the surface layer, VSR achieves through-thickness stress redistribution, making it the preferred method for large structures and thick sections where surface treatment alone is insufficient.

Frequency selection and timing

The general operating range for VSR is 10–200 Hz, with resonant mode more effective but requiring prior modal analysis to identify the assembly’s natural frequency. Setting the vibration frequency to 50–95% of the natural frequency is the recommended range; the closer to resonance, the greater the stress reduction. For Q345 and 304L stainless steel, hole-drilling measurements confirm 30–50% stress reduction across this range. High-strength steel joints achieve greater than 80% longitudinal stress reduction when frequency approaches 95% of natural frequency.

“Vibration applied after welding is more effective than simultaneous vibration, because material at the weld line continues to yield as it cools — which can erase the beneficial effect of high-temperature vibration.”

Timing is a frequently misunderstood variable. Research shows that post-weld application is generally more reliable than simultaneous vibration for the reason stated above. However, for thin sections, mid-welding regulation can be superior. The amplitude range of 0.2–2.0mm should be approached conservatively: start at 0.5mm for 10 minutes and optimise based on measured stress response. Duration of 5–30 minutes is typical depending on joint size and material.

Multi-frequency approach

A multi-frequency strategy — combining ultrasonic vibration at 20+ kHz with low-frequency vibration at 10–200 Hz during welding — yields larger reduction rates than single-frequency methods. This approach is particularly relevant for complex joint geometries where a single frequency may not effectively couple with all stress-bearing regions of the assembly.

Setup configuration

Vibrators should be positioned symmetrically about the weld seam axis and operated in counter-phase to achieve a balanced stress field modification. The excitation direction should be in the plane parallel to the weld seam axis. Material grain size affects VSR efficiency: finer grains in Q235 respond better than coarser grains in Q345, which should be factored into parameter selection when working with different steel grades.

Low-temperature induction heating as a third-path alternative

Low-temperature localised induction heating provides a non-mechanical route to residual stress reduction in welded joints when neither UIT nor VSR is feasible — for example, in materials sensitive to mechanical impact or in geometries that prevent effective vibrator coupling. The method works by controllably heating the compressive stress zones adjacent to the weld, inducing balancing tensile stresses in those zones and thereby relieving the original tensile stresses in the weld zone itself.

The operating temperature range is 100–500°C — deliberately below the metallurgically significant threshold to avoid phase changes or softening. Induction coils operating at 50 Hz to 10 MHz (typically around 90 kHz) are the primary heating method, though defocused laser or electron beam heating can also be used. The target zones — typically 30–50mm from the weld centreline — should be identified via finite element modelling before treatment. According to research published by engineering materials bodies including ASME, the ability to achieve stress relief without exceeding critical temperatures is particularly valuable for high-strength steels and precipitation-hardened alloys.

The achievable stress reduction of up to 50% longitudinal stress is lower than UIT (80–95%) and comparable to the lower end of the VSR range (30–50% for standard configurations). However, for thick sections of 25mm or greater where mechanical methods are contraindicated, induction heating represents a viable and validated alternative. The method should always be preceded by FEM mapping of the residual stress field to correctly locate the heating zones.

Comparing UIT and VSR: selection logic and validation plan

Selecting between UIT and VSR depends on the component’s geometry, the required depth of stress relief, available equipment, and whether quantifiable repeatability is a priority. The comparison below — drawn from validated performance data across multiple material classes — provides the selection framework.

Decision rules

• If fatigue life and surface crack resistance are critical → choose UIT, which introduces a deep compressive layer at the weld toe.
• If the component is very large or thick and through-thickness stress redistribution is needed → choose VSR.
• If you need quantifiable, repeatable results → choose UIT with force control (the GB2617650A approach).
• If cost and simplicity are priorities → choose VSR.
• For critical applications, consider combining both: VSR during welding + UIT post-weld for maximum effect.

Validation plan

Before releasing treated components, a structured validation sequence is essential. According to guidance from ASTM and welding engineering practice, the following seven-step process should be followed:

1. Material characterisation: Confirm base metal and weld metal yield strength and hardness.
2. Baseline measurement: Map the as-welded residual stress field using X-ray diffraction or hole-drilling at the weld toe, HAZ, and base metal.
3. Method selection: Choose UIT or VSR based on the comparison table above.
4. Parameter optimisation: Run 2–3 trials with varying force/frequency/amplitude.
5. Post-treatment verification: Re-measure the stress field and calculate the elimination rate.
6. Mechanical testing: Conduct tensile and fatigue tests to confirm no degradation in joint strength.
7. Microstructure check: Metallographic examination to ensure no surface damage or unintended phase changes.

Engineering risk management requires attention to four specific issues: grain size sensitivity (finer grains in Q235 respond better to VSR than coarser grains in Q345); timing window (avoid VSR during active cooling phase if possible); force calibration for UIT (establish material-specific curves before production); and coverage completeness (100% coverage of critical zones is mandatory). Verification should always use destructive (hole-drilling) or non-destructive (X-ray or neutron diffraction) methods before component release. Neutron diffraction, referenced in standards from NIST, provides through-thickness stress profiles unavailable by surface methods alone.