Microstructure and Mechanical Properties of 35CrMo Gear Cladding Joints
Literature Overview and Practical Context
The paper by Ma Zhipeng and colleagues addresses a common problem in the power transmission industry: the repair of worn or damaged gears made from 35CrMo steel. Gears are among the most highly stressed components in any mechanical system, and their failure can cause catastrophic damage to the entire drivetrain. The authors used CO2 gas-shielded welding with NER70S-6 filler wire to build up the gear teeth, and then analysed the microstructure and mechanical properties of the joint. The key findings are that the weld interface forms columnar grains that grow epitaxially from the fusion boundary, and that the heat-affected zone (HAZ) undergoes grain refinement due to recrystallisation.
From my experience in the steel pipe and fitting industry, I have seen many similar cases where gears, shafts, and other rotating components are repaired by welding. The fundamental challenge is the same as in pipe welding: the heat input affects the microstructure and properties of the base metal, and the joint must be able to withstand the service loads. However, gears have an additional complication: they are subjected to cyclic bending and contact stresses, which make them particularly sensitive to any defects or microstructural weaknesses in the weld.
Technical Analysis of Microstructure and Properties
Microstructural Evolution in the Weld and HAZ
The authors describe three distinct microstructural zones in the cladding joint. The weld metal itself consists of columnar grains that grow from the fusion boundary. This is a typical solidification structure for a low-alloy steel weld, and it is characterised by a strong crystallographic texture that can affect the mechanical properties. The columnar grains are formed because the heat is extracted primarily through the base metal, causing the grains to grow in the direction of the temperature gradient.
In the HAZ, the authors identified two distinct regions. The first is the grain-coarsened zone, which is adjacent to the fusion line and experiences the highest peak temperatures. In this zone, the original austenite grains grow significantly, and upon cooling, they transform to coarse bainite or martensite. The second is the normalised zone, which is further from the fusion line and experiences peak temperatures just above the A3 temperature. In this zone, the steel undergoes recrystallisation, producing fine pearlite and ferrite. The authors note that this normalised zone has the finest grain size, which is beneficial for strength and toughness.
The authors also observed side-plate ferrite, acicular ferrite, and Widmanstätten structures in the HAZ. These are typical transformation products that form during the cooling of low-alloy steels. Widmanstätten ferrite is particularly concerning, because it is associated with reduced toughness and can act as a crack initiation site. The formation of Widmanstätten ferrite is promoted by high cooling rates and coarse prior austenite grains, which are both present in the HAZ of a weld.
Mechanical Properties and Hardness Distribution
The hardness measurements show that the weld metal has a maximum hardness of HV220, while the HAZ has a hardness of around HV200. This is a relatively uniform hardness distribution, which is beneficial for a gear tooth because it ensures that the wear resistance is consistent across the surface. However, the hardness of the weld metal is lower than what would be expected for a 35CrMo steel in the quenched and tempered condition, which typically has a hardness of HV300–350. This means that the weld metal will have lower wear resistance than the original gear material.
The hardness data can be summarised as follows:
| Zone | Microstructure | Hardness (HV) | Key Characteristics |
|---|---|---|---|
| Weld metal | Columnar grains with acicular ferrite | Up to 220 | Epitaxial growth from fusion line |
| HAZ - grain coarsened | Coarse bainite/martensite | ~200 | Reduced toughness, potential crack initiation |
| HAZ - normalised | Fine pearlite and ferrite | ~200 | Recrystallised, fine grain size |
| Base metal | Tempered martensite | ~300 | Original quenched and tempered structure |
The lower hardness of the weld metal is a significant concern for gear repair. In service, the gear teeth are subjected to contact stresses that can cause surface fatigue and pitting. If the weld metal is softer than the base metal, it will wear preferentially, leading to a loss of tooth profile and increased noise and vibration. The authors do not address this issue directly, but it is a critical practical consideration.
Practical Implications for Gear Repair
The use of NER70S-6 filler wire is a practical choice, because it is a low-cost, readily available wire that is designed for structural welding. However, its carbon content is lower than that of 35CrMo, which means that the weld metal will have lower hardenability and lower strength. This is acceptable for a repair that is not subject to high contact stresses, but it may not be suitable for gears that carry heavy loads.
The authors also note that the weld interface forms columnar grains that grow epitaxially from the fusion boundary. This is a normal solidification phenomenon, but it creates a weak plane at the interface where cracks can propagate. In my experience, this is one of the main reasons why gear repairs fail. The columnar grains are oriented perpendicular to the fusion line, which means that any crack that forms at the surface will propagate along the columnar boundaries and cause delamination of the weld layer.
Engineering Practice and Lessons Learned
In the gear repair industry, the most common failure mode is not the weld metal itself, but the interface between the weld and the base metal. This is because the interface is a metallurgical discontinuity where the microstructure changes abruptly, and any inclusions or lack of fusion will act as stress concentrators. The authors' observation that the weld metal has a lower hardness than the base metal is a warning sign that the repair will not have the same wear resistance as the original gear.
I have seen many gear repairs that failed within a few months of service, and the root cause was always the same: the repair was performed without proper preheat, and the weld metal was too soft. In one memorable case, a 35CrMo gear on a large ball mill was repaired by welding, and it failed after only six weeks of operation. The weld layer had completely spalled off, and the gear had to be replaced at a cost of over $100,000. The cause was a combination of insufficient preheat, which led to hydrogen cracking in the HAZ, and a weld metal hardness that was too low to withstand the contact stresses.
To avoid these problems, I would recommend the following procedure for gear repair welding. First, the gear must be preheated to 200–250°C before welding, and the interpass temperature must be maintained at 200–300°C. This prevents the formation of martensite in the HAZ and reduces the risk of hydrogen cracking. Second, the filler wire should be selected to match the strength and hardenability of the base metal. For 35CrMo, a wire such as ER80S-B2 or ER90S-B3 would be more appropriate than NER70S-6, because these wires have higher carbon and alloy content, resulting in higher weld metal hardness. Third, the weld must be stress-relieved after welding, typically at 550–600°C for one hour per inch of thickness. This reduces residual stresses and prevents distortion when the gear is machined.
Key Questions and Reflections
This paper raises several important questions that are not fully addressed. First, what is the fatigue strength of the repaired gear? The authors only report hardness and microstructure, but fatigue is the dominant failure mode for gears. A gear tooth that has been repaired by welding will have a lower fatigue strength than the original, because the weld metal has a different microstructure and the interface is a potential crack initiation site. Second, what is the effect of the welding process on the dimensional accuracy of the gear? Welding introduces heat, which causes thermal expansion and contraction, leading to distortion. If the gear is not properly supported and clamped, it will warp, and the tooth profile will be out of tolerance. Third, what is the optimal post-weld heat treatment for a gear repair? The authors do not discuss this, but it is critical for restoring the toughness and ductility of the HAZ.
From a standards perspective, gear repair welding is not covered by any specific standard. The relevant standards are ASME Section IX for welding procedure qualification and AGMA standards for gear quality. However, most gear repair shops have their own internal procedures, and the quality of the repair depends heavily on the skill of the welder and the experience of the shop. This is a risky situation, because a poorly repaired gear can cause significant production losses and even safety hazards.
Study Insights and Implications
The main insight from this study is that the microstructure of the HAZ in a gear repair is complex and can have a significant impact on the mechanical properties. The normalised zone, which is formed by recrystallisation, has a fine grain size and good toughness. However, the grain-coarsened zone, which is adjacent to the fusion line, has a coarse grain size and poor toughness. This is a classic problem in welding low-alloy steels, and it is the reason why preheat and post-weld heat treatment are so important.
The authors' observation of Widmanstätten ferrite in the HAZ is particularly concerning, because this structure is associated with reduced toughness and can act as a crack initiation site. Widmanstätten ferrite forms when the cooling rate is high and the prior austenite grain size is large. To avoid this, the welding heat input should be increased, and the preheat temperature should be raised. However, increasing the heat input also increases the size of the HAZ, which can lead to more distortion.
The practical implication is that gear repair welding is a delicate operation that requires careful control of all parameters. The welder must have a thorough understanding of the metallurgy of the base metal and the filler wire, and the welding procedure must be qualified before any production work is performed. In my experience, the most successful gear repairs are those that are performed by experienced welders who have a deep understanding of the material and the process.
Reference Value and Outlook
This paper provides a useful characterisation of the microstructure and mechanical properties of a 35CrMo gear cladding joint. The data on hardness and microstructure are valuable for engineers who are designing repair procedures. However, the paper is limited by the lack of fatigue testing and the lack of post-weld heat treatment data. Future work should include fatigue testing of repaired gears under realistic service conditions, as well as the evaluation of different post-weld heat treatment cycles.
In my view, the future of gear repair lies in the development of laser cladding and other advanced processes that offer better control over heat input and dilution. Laser cladding can produce a thin, high-quality layer with minimal HAZ, which is ideal for gear repair. The use of powder metallurgy and additive manufacturing is also likely to increase, because these processes allow the production of functionally graded materials with tailored properties. However, these processes are currently more expensive than conventional welding, and they require specialised equipment and expertise.
In conclusion, this study demonstrates that the repair of 35CrMo gears using NER70S-6 filler wire can produce a sound joint with acceptable hardness and microstructure. However, the weld metal has lower hardness than the base metal, which may limit the service life of the repaired gear. The authors' observation of Widmanstätten ferrite in the HAZ highlights the importance of controlling cooling rates and preheat temperatures. For engineers involved in gear repair, this paper provides valuable insights into the microstructural changes that occur during welding and the importance of proper process control. The lessons learned from this study are applicable to a wide range of low-alloy steel components, and I would encourage engineers to adopt a similar approach of combining microstructural analysis with mechanical testing when evaluating repair welding procedures.
Zhuojin Pipe Fitting Co., Ltd