As a leading internal gear manufacturer, we frequently encounter challenges related to distortion during the heat treatment of heavy-duty internal gears. These components, critical for high-load applications such as reducers, demand precise carburizing and quenching processes to achieve the required hardness, depth of case, and dimensional stability. Internal gears, due to their thin-walled structure and large diameter-to-width ratios, are particularly susceptible to distortion, which can lead to scrapping if not controlled effectively. In this comprehensive analysis, I will share our experiences and methodologies for optimizing the carburizing and quenching of 18CrNiMo7-6 steel internal gears, focusing on distortion mitigation through process modifications, including stress relief annealing, controlled heating, and reduced cooling intensity. Throughout this discussion, the terms “internal gear manufacturer” and “internal gears” will be emphasized to highlight our expertise and the specific components involved.
The material of focus, 18CrNiMo7-6 steel, is renowned for its high bending strength, contact fatigue resistance, surface hardness, and wear resistance, coupled with a tough core. However, its high hardenability often results in significant distortion post-carburizing and quenching, necessitating careful process design. For internal gears, which are inherently prone to dimensional changes due to their geometry, uncontrolled distortion can render the parts unusable after grinding, as the final dimensions may not meet technical specifications. Our goal as an internal gear manufacturer is to implement strategies that reduce this distortion, ensuring that internal gears maintain their integrity and performance under heavy loads.
To address these issues, we conducted a series of experiments on heavy-duty internal gears made from 18CrNiMo7-6 steel. The gears had a module of m = 7 mm, 103 teeth, and a pressure angle of 20°, with carburized case depth requirements of 1.4 to 1.8 mm and surface hardness of 58-62 HRC. The chemical composition of the steel was 0.17% C, 0.81% Mn, 0.21% Si, 0.15% S, 0.15% P, 1.71% Cr, 1.67% Ni, and 0.28% Mo, which aligns with standard specifications for such applications. As an internal gear manufacturer, we prioritized understanding the factors influencing distortion, including residual stresses from machining, internal stresses during heating and cooling, and the geometric constraints of internal gears.
One primary factor contributing to distortion is the residual stress from prior machining operations, such as gear cutting. For internal gears, which undergo processes like broaching or shaping, these stresses can exacerbate dimensional changes during heat treatment. To mitigate this, we introduced a stress relief annealing step before carburizing. The annealing was performed at 500°C for 15 hours in a box-type resistance furnace, followed by controlled cooling. This process helps in relieving machining-induced stresses without forming excessive oxide scale that could interfere with subsequent carburizing. The effectiveness of this step was evaluated by measuring dimensional changes before and after annealing, and it proved crucial in stabilizing the internal gears for further processing.
Another critical aspect is the control of internal stresses during carburizing and quenching. These stresses arise from thermal gradients (thermal stress) and phase transformations (transformational stress). For internal gears, which have large thin-walled sections, rapid heating or cooling can lead to uneven expansion and contraction, causing distortion. We modified the carburizing process by implementing a controlled heating rate of 200°C/h and adding an intermediate hold at 800°C for 1 hour before reaching the carburizing temperature. This gradual heating reduces thermal gradients and minimizes thermal stress. Additionally, we lowered the carburizing temperature from a conventional 920-1050°C range to 930°C and reduced the quenching temperature from 840°C to 800°C. The quenching medium was oil, and we increased its temperature to 70°C (from the standard 60°C) to decrease the cooling intensity, thereby reducing both thermal and transformational stresses. The cooling time was precisely calculated to ensure the gear teeth reached a tempering temperature of 100-120°C after quenching, promoting uniformity.
To quantify the distortion, we focused on the span measurement (over pins distance) of the internal gears, as it directly reflects dimensional changes in the tooth geometry. We tested three distinct process schemes: Scheme 1 involved conventional carburizing and quenching without any modifications; Scheme 2 incorporated the improved carburizing and quenching process with controlled heating and reduced temperatures but no stress relief annealing; and Scheme 3 combined stress relief annealing with the improved carburizing and quenching process. For each scheme, we measured the span before carburizing, after stress relief (if applicable), and after carburizing and quenching. The distortion was calculated as the difference between the initial and final dimensions, and the required grinding allowance was determined based on the final span relative to the drawing specification of 711.65 mm.
The microstructural analysis and hardness gradients were assessed using metallographic techniques and microhardness testing. The microstructure was evaluated according to JB/T 6141.3-1992 standards, and the effective case depth was defined as the depth where the hardness reached 550 HV10. We used optical microscopy and a Vickers microhardness tester with a 10 kg load for these analyses. The results showed that all schemes met the microstructural requirements, with martensite and retained austenite at level 3, carbides at level 2, and core ferrite at levels 2-3, indicating satisfactory heat treatment quality. However, the distortion behavior varied significantly among the schemes.
In terms of hardness gradients, the microhardness profiles revealed that Scheme 1 achieved a case depth of approximately 2.0 mm, with a surface hardness of 661 HV10 at 0.4 mm depth. Scheme 2 and Scheme 3 showed slightly deeper case depths of 2.2-2.4 mm and 2.2 mm, respectively, with surface hardness values of 668 HV10 and 672 HV10 at 0.4 mm depth. This indicates that the modified processes did not compromise the hardening effectiveness, and the grinding allowance could be managed within 0.4 mm to maintain the required surface hardness. The deeper case in Scheme 2 and 3 was attributed to the extended time at lower temperatures during the controlled heating phase, which allowed for some carburizing to occur earlier.
To present the distortion data clearly, we have compiled the results in the following table, which summarizes the span measurements and calculated distortions for each scheme. This table highlights the impact of the process modifications on the dimensional stability of internal gears.
| Scheme | Initial Span (mm) | After Stress Relief (mm) | After Carburizing (mm) | Stress Relief Distortion (mm) | Carburizing Distortion (mm) | Grinding Allowance (mm) |
|---|---|---|---|---|---|---|
| 1 | 709.77 – 709.82 | — | 708.75 – 710.56 | — | -0.79 to 1.05 | 1.09 to 2.90 |
| 2 | 709.81 – 709.94 | — | 709.15 – 709.33 | — | 0.61 to 0.73 | 2.32 to 2.50 |
| 3 | 709.75 – 709.95 | 709.66 – 709.74 | 709.38 – 709.51 | 0.09 to 0.24 | 0.30 to 0.47 | 2.14 to 2.27 |
From the table, it is evident that Scheme 1, the conventional process, resulted in the highest and most unpredictable distortion, with span changes ranging from -0.79 mm to 1.05 mm. This led to excessive grinding allowances of up to 2.90 mm, which risked compromising the surface hardness after grinding. In contrast, Scheme 2 showed a more consistent distortion pattern, with spans decreasing by 0.61 to 0.73 mm, indicating a uniform shrinkage of the internal diameter. The grinding allowance was reduced to around 0.4 mm per side, ensuring that the hardness requirements could be met. Scheme 3, which included stress relief annealing, exhibited the least distortion, with carburizing-induced changes of only 0.30 to 0.47 mm and a stable shrinkage trend. This demonstrates the synergistic effect of stress relief and process optimization in controlling distortion for internal gears.
As an internal gear manufacturer, we also considered the efficiency of these processes. Scheme 1 required the shortest time of 23 hours, primarily for carburizing. Scheme 2 took 30 hours due to the additional controlled heating and cooling stages, while Scheme 3 extended to 45 hours with the inclusion of stress relief annealing. Although Scheme 3 offers the best distortion control, Scheme 2 provides a balanced approach for mass production, where time and quality must be optimized. For internal gears in critical applications, we recommend Scheme 3 to ensure minimal distortion and reliable performance.
The underlying mechanisms of distortion can be modeled using stress-strain relationships and thermal equations. For instance, the thermal stress during heating can be approximated by the formula for thermal expansion: $$\sigma_t = E \alpha \Delta T$$ where $\sigma_t$ is the thermal stress, E is the Young’s modulus, $\alpha$ is the coefficient of thermal expansion, and $\Delta T$ is the temperature gradient. By reducing $\Delta T$ through controlled heating, we minimize $\sigma_t$. Similarly, the transformational stress during quenching relates to the volume change associated with martensitic transformation, which can be expressed as: $$\epsilon_v = \beta (V_m – V_a)$$ where $\epsilon_v$ is the volumetric strain, $\beta$ is a material constant, $V_m$ is the volume of martensite, and $V_a$ is the volume of austenite. Lower cooling rates reduce the rate of transformation, thereby decreasing $\epsilon_v$ and the associated stress.
In addition to these theoretical aspects, we performed regression analysis on the distortion data to derive predictive models for internal gears. For example, the distortion after carburizing (D) can be correlated with process parameters such as carburizing temperature (T_c), quenching temperature (T_q), and stress relief time (t_s). A simplified linear model might be: $$D = k_0 + k_1 T_c + k_2 T_q + k_3 t_s$$ where k_0, k_1, k_2, and k_3 are coefficients determined from experimental data. For our internal gears, the data from Scheme 2 and 3 suggest that reducing T_c and T_q while increasing t_s leads to a decrease in D, confirming the effectiveness of our modifications.
Furthermore, the microhardness gradient can be described using a diffusion-based model for carbon concentration, which influences the case depth. The carbon profile C(x) as a function of depth x can be approximated by Fick’s second law: $$\frac{\partial C}{\partial t} = D \frac{\partial^2 C}{\partial x^2}$$ where D is the diffusion coefficient. Solving this for our conditions, we obtain a case depth that aligns with the measured values, ensuring that the internal gears achieve the desired mechanical properties without excessive distortion.
In conclusion, as an internal gear manufacturer, we have demonstrated that through systematic process improvements—including stress relief annealing, controlled heating rates, reduced carburizing and quenching temperatures, and moderated cooling intensity—the distortion of heavy-duty internal gears made from 18CrNiMo7-6 steel can be significantly reduced. These strategies not only enhance dimensional stability but also maintain the required microstructure and hardness, ensuring that internal gears perform reliably in demanding applications. The repeated emphasis on “internal gear manufacturer” and “internal gears” throughout this discussion underscores our commitment to advancing the manufacturing techniques for these critical components. By adopting these optimized processes, we can minimize scrap rates, reduce grinding allowances, and deliver high-quality internal gears that meet stringent technical standards. Future work may involve finite element analysis to simulate distortion patterns and further refine the process parameters for even greater precision in internal gear production.