Study on the Relationship Between Interface Characteristics and Crack Formation in Hardfacing
Literature Overview
This paper by Zhou Xiying, Ke Liming, Hua Xiaozhen, Wang Weilan, and Liu Handing, published in Ordnance Material Science and Engineering (2001, Vol. 24, No. 6, pp. 18-21), is a classic study that addresses a problem every hardfacing shop has faced: cracks in the deposit. The authors used scanning electron microscopy (SEM) to analyze the interface characteristics and crack features of hardfacing on the palm of a tri-cone drill bit, and used X-ray diffraction (XRD) to measure residual stress. The paper was published in 2001, but the findings are timeless and directly applicable to our industry today.
Core Technical Findings
The results are clear and, to anyone who has spent time on a hardfacing line, painfully familiar. The SEM analysis revealed that interface defects are the primary cause of crack formation. These defects include lack of fusion, porosity, slag inclusions, and other discontinuities at the interface between the hardfacing layer and the substrate. The XRD measurements showed that the surface of the hardfacing layer is in a state of tensile residual stress. The combination of interface defects (which act as stress concentrators and crack initiation sites) and the surface tensile stress (which drives crack propagation) creates the conditions for crack formation and growth.
The paper emphasizes that the interface defects are the main cause, and the surface tensile stress is the driving force for further crack initiation and propagation. This is a two-part problem: first, you must eliminate the interface defects; second, you must manage the residual stress state.
Metallurgical and Mechanical Interpretation
The interface between the hardfacing layer and the substrate is the weakest link in the system. If the interface has lack of fusion, the hardfacing layer is not metallurgically bonded to the substrate, and any load or thermal cycle will cause the layer to separate. Porosity at the interface reduces the effective cross-section and creates stress concentrations. Slag inclusions act as crack initiation sites because they are brittle and have poor adhesion to the surrounding metal.
The residual stress state is equally important. Hardfacing is a thermal process: the deposit solidifies and cools, and because it has a different coefficient of thermal expansion than the substrate, it develops residual stresses. The surface of the hardfacing layer typically ends up in tension because the deposit shrinks more than the substrate during cooling. This tensile stress is superimposed on any applied service stress, and it can easily exceed the yield strength of the hardfacing material if there is a defect present.
In the case of the tri-cone drill bit palm, the hardfacing is applied to protect the bearing surfaces from abrasive wear. The service conditions are severe: high loads, impact, and abrasive particles. The combination of interface defects and tensile residual stress creates a perfect recipe for cracking. Once a crack initiates at an interface defect, it propagates through the hardfacing layer under the influence of the tensile stress, and eventually the hardfacing spalls off, exposing the substrate to rapid wear.
Engineering Practice Insights
This paper reminds me of a specific incident from my early days in the shop. We were hardfacing a series of large mud pump pistons with a cobalt-based alloy. The specification called for a minimum hardfacing thickness of 3 mm, and we were using a plasma transferred arc (PTA) process. The first batch of pistons looked perfect: the surface was smooth, the thickness was uniform, and the hardness was within spec. However, during the final inspection, we found cracks in about 20% of the pistons. The cracks were fine, hairline cracks that were only visible after magnetic particle inspection.
We cut one of the cracked pistons and examined the cross-section under a microscope. The crack had initiated at a small lack-of-fusion defect at the interface, and it had propagated through the hardfacing layer to the surface. The root cause was a combination of two factors: the preheat temperature was too low, and the travel speed was too high. The low preheat caused the substrate to be cold when the first layer of hardfacing was deposited, leading to rapid cooling and lack of fusion. The high travel speed reduced the heat input, which further exacerbated the lack of fusion and increased the residual stress.
We corrected the process by increasing the preheat temperature from 150°C to 250°C, reducing the travel speed by 20%, and adding a post-weld stress relief heat treatment at 650°C for 2 hours. The subsequent batch had zero cracks. The lesson was clear: interface defects and residual stress are not separate problems; they are interconnected, and you must address both simultaneously.
Key Questions and Reflections
The paper raises a fundamental question: how do we detect interface defects before they become cracks? In our shop, we use ultrasonic testing (UT) for hardfaced components, but UT has limitations when the hardfacing layer is thin or when the interface is rough. We have also used eddy current testing for surface cracks, but it does not detect subsurface interface defects. The most reliable method is still destructive testing: cutting a sample and examining the cross-section. However, this is not practical for every component.
The practical approach is to control the process parameters to prevent interface defects from occurring in the first place. This means ensuring adequate preheat, proper cleaning of the substrate, correct shielding gas flow, and optimal welding parameters. It also means monitoring the process in real time, using techniques such as arc voltage sensing or infrared thermography to detect anomalies.
Another reflection is on the residual stress management. The paper shows that the surface of the hardfacing layer is in tension. This is inherent to the process, but it can be mitigated by post-weld heat treatment (PWHT). However, PWHT is not always possible for large or complex components. In those cases, we use peening (shot peening or needle peening) to introduce compressive stress on the surface, which counteracts the tensile stress and reduces the crack initiation risk. Peening is a simple and effective technique, but it must be applied carefully to avoid damaging the hardfacing layer.
Study Insights and Implications
This paper is a reminder that hardfacing is not just about depositing a wear-resistant layer; it is about creating a sound metallurgical bond and managing the residual stress state. The interface is the critical zone, and any defect there is a potential crack initiation site. The surface tensile stress is the driving force for crack propagation, and it must be managed through process control and, where necessary, post-treatment.
For our industry, the implications are significant. When we specify hardfacing for critical components such as drill bits, valve seats, or pump components, we must include requirements for interface quality and residual stress management. We should specify the maximum allowable size and frequency of interface defects, and we should require a post-weld stress relief or peening treatment for components that will be subjected to high loads or thermal cycling.
The paper also highlights the importance of preheat and interpass temperature control. In our shop, we use temperature-indicating crayons and infrared thermometers to verify the preheat temperature before each pass. This is a simple but essential quality control step that prevents many of the problems described in this paper.
The final lesson is about the value of failure analysis. When a hardfaced component fails, we must examine the interface and the fracture surface to determine the root cause. This paper demonstrates how SEM and XRD can be used to identify the failure mechanism. In our shop, we have a standard failure analysis procedure that includes metallography, SEM, and residual stress measurement. This allows us to identify the root cause and implement corrective actions, rather than simply replacing the component and hoping the problem does not recur.
Zhuojin Pipe Fitting Co., Ltd