Failure Analysis of Chemical Pipeline Elbow Cracking During Welding

Literature Overview

The paper by Zhai Hongbo, Dong Leiyun, Shi Zhexiong, and Jiang Xiaodong from East China University of Science and Technology, published in Chemical Equipment and Pipeline in 2010 (Vol. 47, Issue 4, pp. 64-67), presents a detailed failure analysis of a 20 steel elbow that cracked during the welding process. The investigation employs a comprehensive suite of characterization techniques including macroscopic crack observation, chemical composition analysis, mechanical property testing, fracture surface analysis, and metallographic examination, all correlated with the elbow's hot-push manufacturing process. The key finding is that the elbow contained pre-existing (original) cracks from the manufacturing stage, and the welding thermal cycle caused these cracks to propagate and coalesce into through-thickness cracks.

Manufacturing Background and Failure Mechanism

The elbow was manufactured by the hot-push (hot bend) process, a widely used method for producing butt-weld fittings from pipe blanks. In this process, a straight pipe section is heated uniformly and then bent around a mandrel to the desired radius. The critical quality factors include heating temperature uniformity, bending speed, mandrel design, and cooling rate. If any of these parameters are poorly controlled, residual stresses and micro-cracks can develop, particularly at the inner arch (intrados) where compressive deformation is most severe.

The failure occurred during the subsequent welding operation, when the elbow was being joined to a straight pipe section. The welding thermal cycle introduced additional thermal stresses into the already-stressed elbow material. The pre-existing cracks, which may have been sub-surface or shallow surface defects, acted as stress concentrators. Under the combined effect of welding thermal stresses and residual stresses from the bending process, these cracks propagated rapidly, resulting in through-thickness cracking that rendered the elbow unusable.

Characterization Results

The investigation followed a systematic approach consistent with standard failure analysis protocols. Macroscopic examination revealed the crack origin and propagation direction, which were critical for identifying the failure sequence. Chemical composition analysis confirmed that the material met the specification for 20 steel (GB/T 8163 or equivalent), with no abnormal segregation or contamination. Mechanical property testing showed that the base material properties were within the specified range, indicating that the failure was not due to material non-conformance.

Fracture surface analysis (SEM) provided the most definitive evidence. The crack initiation sites showed features consistent with manufacturing-induced defects such as micro-cracks from cold working or thermal fatigue during the hot-push process. The crack propagation region displayed characteristics of both quasi-cleavage and ductile fracture, depending on the local constraint and temperature during crack growth. Metallographic examination revealed the presence of original cracks near the inner arch surface, confirming that the defects existed prior to welding.

Root Cause Analysis and Process Traceability

The root cause is identified as a manufacturing defect in the hot-push bending process. The inner arch of the elbow experiences the most severe plastic deformation during bending, and if the heating temperature is insufficient or the bending speed is too fast, the material may not deform plastically enough to accommodate the imposed curvature, leading to the formation of micro-cracks. These cracks may be too small to be detected by standard non-destructive testing methods such as magnetic particle testing (MT) or ultrasonic testing (UT) at the time of manufacturing, especially if they are subsurface or oriented unfavorably for detection.

The welding process then acted as the trigger for failure. The thermal cycle of welding introduces rapid heating and cooling, which generates thermal stresses. In the presence of pre-existing cracks, these stresses concentrate at the crack tips and drive crack propagation. The fact that the cracks became through-thickness indicates that the original cracks were already significant in depth, and the welding thermal input was sufficient to propagate them completely across the wall thickness.

Engineering Practice Implications

This case study carries important lessons for the fitting supply chain and welding operations. First, incoming inspection of butt-weld fittings, particularly elbows produced by the hot-push process, should include enhanced non-destructive testing beyond the minimum requirements of standards such as ASME B16.9 or GB/T 12459. For critical applications, supplementary testing such as dye penetrant testing (PT) on the inner arch surface, or even phased array ultrasonic testing (PAUT), may be warranted.

Second, welding procedures for elbows with known or suspected manufacturing defects should include pre-heat and controlled cooling to reduce thermal stresses. A pre-heat temperature of at least 100°C is recommended for 20 steel elbows, and interpass temperature control should be maintained. Post-weld heat treatment (PWHT) may be beneficial for relieving residual stresses in critical joints.

Third, the hot-push bending process itself requires tighter quality control. Process parameters such as heating temperature (typically 900-1100°C for 20 steel), bending speed, and cooling rate should be strictly controlled and documented. Statistical process control (SPC) of these parameters can reduce the probability of manufacturing defects.

Study Insights

This failure analysis is a textbook example of how a manufacturing defect can remain latent until a subsequent process step triggers catastrophic failure. The two-stage nature of the failure—manufacturing defect followed by welding-induced propagation—underscores the importance of understanding the full process chain. In modern quality management frameworks such as FMEA (Failure Mode and Effects Analysis), this scenario would be classified as a high-severity failure mode with a potentially high occurrence rating if manufacturing quality is not controlled. The detection rating would depend on the adequacy of incoming inspection procedures.

From a materials perspective, the case reinforces the principle that the mechanical integrity of a fitting depends not only on the final material properties but also on the manufacturing history. A fitting that passes all standard material tests can still fail in service if it contains undetected manufacturing defects. This is particularly relevant for hot-push elbows, which are among the most commonly used butt-weld fittings in chemical and petrochemical piping systems.