Surface Flatness of Single-Layer Multi-Pass Welds in Automated Flat Cladding

Literature Overview

This study by Quan Guozheng, Shi Ruiju, Lu Shun, Zhang Jiansheng, Dong Xugang, and Zhou Jie (2020, published in Journal of Plasticity Engineering) addresses a critical practical challenge in automated flat cladding operations: achieving acceptable surface flatness in single-layer, multi-pass weld deposits. The research was conducted at the School of Materials Science and Engineering, Chongqing University, in collaboration with Chongqing Jiepin Technology Co., Ltd. and Chongqing Dajiang Jiexin Forging Co., Ltd. The work was supported by multiple national and municipal research programs, including the National Key R&D Program (2018YFB1106502). Surface flatness is a critical quality attribute for cladding applications, particularly in pressure vessel fabrication, where excessive surface irregularities can affect subsequent machining, inspection, and service performance.

Core Technical Content

In automated flat cladding, a single layer is deposited in multiple passes (typically 3–8 passes depending on the required overlay thickness and plate width). Each pass builds upon the previous one, and the cumulative effect of weld reinforcement, thermal distortion, and cooling contraction determines the final surface flatness. Achieving acceptable surface flatness (typically within ±0.5 mm over the cladding width) is essential for subsequent machining operations and for ensuring uniform overlay thickness.

Process Configuration

The automated flat cladding process typically involves:

Welding process: Submerged arc welding (SAW) or flux-cored arc welding (FCAW)
Wire feed: Single or dual wire
Travel direction: Transverse (perpendicular to plate rolling direction) or longitudinal
Pass arrangement: Parallel passes with overlap, or zigzag pattern
Automation: CNC-controlled torch and wire feed with position feedback

Factors Affecting Surface Flatness

The study identifies and quantifies the influence of several process parameters on surface flatness:

Parameter	Effect on Flatness	Optimal Range
Travel speed	Too fast: low reinforcement; Too slow: excessive reinforcement	300–500 mm/min
Wire feed speed	Affects deposition rate and bead width	6–10 m/min
Arc current	Higher current: wider bead, more reinforcement	400–600 A (SAW)
Voltage	Affects bead width and penetration	28–35 V (SAW)
Stick-out length	Affects arc stability and bead profile	15–25 mm
Pass spacing	Overlap ratio affects surface profile	30–50% overlap
Flux coverage	Affects cooling rate and bead shape	Adequate coverage
Plate preheat	Reduces thermal distortion	100–150°C

Weld Reinforcement and Surface Profile

The surface profile of a multi-pass cladding layer is determined by the individual bead profiles and their interaction. Each pass creates a convex reinforcement that is partially filled by subsequent passes. The final surface profile depends on:

Bead width-to-height ratio: Wider, flatter beads produce better surface flatness.
Pass overlap: Sufficient overlap (30–50%) ensures that valleys between beads are filled.
Thermal interaction: Adjacent passes interact thermally, affecting solidification and shrinkage.
Cooling distortion: Differential cooling across the cladding width causes warping.

Experimental Results

The study presents typical surface flatness measurements for various process parameter combinations:

Condition	Surface Flatness (mm)	Bead Width (mm)	Bead Height (mm)	Passes
Low heat input	1.2–1.8	12–15	4–5	5
Medium heat input	0.5–1.0	18–22	3–4	5
High heat input	0.3–0.8	22–28	2.5–3.5	5
Optimized parameters	0.2–0.5	20–25	2.5–3.0	5

The results demonstrate that surface flatness improves with increasing heat input (up to a point), as higher heat input produces wider, flatter beads with lower reinforcement. However, excessive heat input leads to other defects (excessive penetration, distortion, microstructural coarsening).

Surface Flatness Control Strategies

Based on the research findings, the following strategies are recommended for achieving acceptable surface flatness:

Process parameter optimization:

Use higher heat input (within acceptable limits for the material)
Optimize wire feed speed for maximum bead width
Adjust voltage to control bead profile
Use appropriate stick-out length for arc stability

Pass arrangement optimization:

Use sufficient overlap (40–50%) between passes
Consider using a weaving pattern for wider beads
Maintain consistent pass spacing with CNC control
Consider back-rolling for the final pass

Thermal management:

Apply uniform preheat to the entire plate
Use backing plates to control cooling rate
Consider sequential welding patterns to minimize distortion
Monitor and control interpass temperature

Post-weld treatment:

Light machining (0.5–1.0 mm) to achieve final flatness
Induction heating for local distortion correction
Mechanical straightening for severe distortion

Engineering Practice Guidelines

For automated flat cladding operations in pressure vessel fabrication, the following quality control measures should be implemented:

Control Point	Method	Acceptance Criteria
Surface flatness	Straightedge and feeler gauge	≤0.5 mm over 100 mm
Surface roughness	Profilometer	Ra ≤25 μm
Overlay thickness	UT thickness measurement	Nominal ±10%
Surface defects	PT examination	No linear defects >1 mm
Bond strength	Peel test or shear test	≥0.8 × base metal tensile strength

Implementation of PDCA Cycle

The study implicitly follows a PDCA (Plan-Do-Check-Act) approach to surface flatness optimization:

Plan: Identify target flatness specification; select process parameters based on preliminary testing.
Do: Execute welding with selected parameters; monitor in-process conditions.
Check: Measure surface flatness, roughness, and overlay thickness; compare with specifications.
Act: Adjust parameters based on measurement results; implement corrective actions.

Key Questions and Reflections

A significant practical challenge remains the interaction between surface flatness and other quality attributes. Improving flatness through higher heat input may compromise microstructural properties (grain coarsening, reduced hardness). Similarly, increasing overlap to improve flatness increases welding time and cost. Engineers must balance these competing requirements based on the specific application.

Another important consideration is the effect of plate thickness and width on surface flatness. Thicker plates generally exhibit less thermal distortion but may require more passes to achieve the required overlay thickness. Wider plates present challenges related to thermal gradient across the width, which can cause edge effects and uneven bead profiles.

The study also raises the question of whether multi-layer cladding (as opposed to single-layer multi-pass) might offer better surface flatness control. Multi-layer approaches allow for intermediate flatness correction but increase total welding time and cost.

Study Insights and Implications

This research provides practical guidance for engineers dealing with surface flatness challenges in automated flat cladding operations. The key insight is that surface flatness is not determined by a single parameter but by the interaction of multiple process variables, and optimal results require systematic optimization.

The work demonstrates that acceptable surface flatness (≤0.5 mm) can be achieved through proper parameter selection and process control, without requiring extensive post-weld machining. This has significant cost implications for large-scale cladding operations, where machining of several square meters of overlay surface can be time-consuming and expensive.

For future work, I recommend investigating the use of real-time surface profile monitoring (laser scanning or structured light) with closed-loop feedback to the welding parameters. This approach could potentially maintain consistent surface flatness throughout the cladding operation, even in the presence of variable conditions (plate thickness variations, edge effects, etc.). Additionally, the development of predictive models for surface flatness based on process parameters and material properties would enable more efficient process optimization and reduced trial-and-error testing.