In my study, I focus on the precision forging behavior of a hypoid bevel gear driving pinion. A hypoid bevel gear is a special form of spiral bevel gear used to transmit motion between spatially crossed axes. Its driving pinion axis is offset from the driven gear axis by a fixed distance. This offset gives the hypoid bevel gear several distinctive advantages, including a larger contact ratio, better meshing performance, higher load capacity, smoother transmission, higher transmission efficiency, a more compact structure, and lower noise. For these reasons, the hypoid bevel gear is widely used in automotive drive axles and other high-performance transmission systems. However, the same offset also creates a highly asymmetric tooth surface on the pinion, which makes both design and manufacturing much more difficult than for a conventional spiral bevel gear.
I begin with the basic geometric and kinematic nature of the hypoid bevel gear. In a conventional spiral bevel gear pair, the two axes usually intersect, and the pitch surfaces can be approximated by cones. In a hypoid bevel gear pair, the axes are crossed but not intersecting, and the pitch surfaces are theoretically single-sheet hyperboloids. In practice, a pair of conical surfaces is used to replace the hyperboloids, so the hypoid bevel gear still resembles a conical gear in its external form, while its meshing behavior is fundamentally different. The relative motion between the two members of a hypoid bevel gear pair is a combination of rotation about a fixed axis and sliding along that axis. This relative motion creates a relative rotation-sliding axis. When this axis is rotated about each gear axis, a pair of hyperboloid pitch surfaces is generated. This is the origin of the name hypoid bevel gear.
The offset distance is the most important distinguishing parameter of a hypoid bevel gear. It allows the vehicle body to be lowered or raised, improves stability and off-road capability, and gives greater flexibility in the layout of the main reducer. The spiral angle of the driving pinion is larger than that of the driven gear. This increases the overlap ratio and makes the transmission smoother and quieter. A hypoid bevel gear can also realize a relatively large transmission ratio and, in some applications, can replace a worm drive. In addition, for the same number of teeth and the same normal module of the driven gear, the dimensions of the hypoid bevel gear pinion, especially the shaft diameter, can be larger than those of an equivalent spiral bevel gear pinion. This increases the stiffness, strength, and load capacity of the pinion.
At the same time, the hypoid bevel gear has a unique complexity. For a spiral bevel gear, the two sides of a tooth usually have symmetric normal tooth profiles, the same pressure angle, and the same curvature of the normal profile. For a hypoid bevel gear, the driven gear may still have a nearly symmetric normal profile, but the driving pinion has an asymmetric normal profile. The two normal profile lines of the same tooth on the pinion have different curvatures and different pressure angles. This asymmetry makes the design, calculation, and manufacturing of the hypoid bevel gear much more complicated. Furthermore, the contact between the pinion and the gear is not pure rolling; it is a combination of rolling and sliding. This sliding increases the contact pressure and reduces the surface durability and load-carrying capacity if the tooth surface is not properly manufactured.
For these reasons, I consider the hypoid bevel gear to be one of the most challenging transmission components in modern vehicle drivetrains. The manufacturing method directly determines its fatigue life, noise, strength, and cost. At present, cutting is the dominant method for producing hypoid bevel gear teeth. Cutting is mature and accurate, but it has low production efficiency, high manufacturing cost, and low material utilization. More importantly, cutting breaks the continuity of the metal fiber flow lines. This reduces the bending fatigue strength at the tooth root and the contact fatigue strength at the tooth surface. As a result, the service life of a cut hypoid bevel gear can be limited. Precision forging has a different set of advantages. A forged hypoid bevel gear can have metal fibers that follow the tooth profile continuously and reasonably, a dense microstructure, and fewer microscopic defects. This can improve impact strength, bending fatigue strength, wear resistance, corrosion resistance, and service life. It also increases production efficiency, raises material utilization, and reduces production cost.
Geometric and Performance Characteristics
I summarize the main characteristics of the specific hypoid bevel gear driving pinion studied here in Table 1. These parameters control the tooth form, the contact pattern, and the forging difficulty. The small number of teeth, the relatively large module, the large spiral angle, and the offset distance all make the tooth cavity difficult to fill. The left-hand spiral direction also affects the direction of metal flow during forging.
| Parameter | Value | Parameter | Value |
|---|---|---|---|
| Number of teeth | 6 | Spiral angle | 45.998 degrees |
| Module | 12.032 | Spiral direction | Left hand |
| Outer addendum | 15.413 mm | Offset distance | 44.45 mm |
| Outer whole depth | 19.816 mm | Cutter diameter | 304.8 mm |
| Mean pressure angle | 22.5 degrees | Backlash | 0.3 to 0.4 mm |
The transmission ratio of a hypoid bevel gear pair is controlled by the tooth numbers. If I denote the number of teeth of the pinion by \(N_p\) and the number of teeth of the gear by \(N_g\), the ratio is
$$ i = \frac{N_g}{N_p} $$
For the present pinion, \(N_p = 6\), so the ratio depends strongly on the driven gear. The offset distance \(E\) and the spiral angle \(\beta\) determine the sliding condition and the contact geometry. The relative sliding velocity along the tooth profile can be expressed conceptually as
$$ v_s = v_t \tan \beta_s $$
where \(v_t\) is the tangential velocity and \(\beta_s\) is an equivalent sliding angle related to the offset. This sliding is one reason why the hypoid bevel gear requires careful lubrication and surface integrity. When the tooth is produced by cutting, the fiber flow is interrupted, and the sliding action can more easily initiate surface damage. When the tooth is produced by precision forging, the continuous fiber flow can better resist this damage.
Precision Plastic Forming of Hypoid Bevel Gears
I classify the production of hypoid bevel gear teeth into cutting and chipless forming. Cutting includes generated milling, grinding, and lapping. Chipless forming includes precision die forging, closed-die forging, enclosed die forging, orbital forging, and hybrid processes. Among these, precision forging is the most promising route for producing a hypoid bevel gear with continuous metal fibers. The literature and industrial practice show that the driven gear of a spiral bevel gear pair can be produced by hot precision forging in some cases. However, the driving pinion of a hypoid bevel gear is far more difficult because its tooth surfaces are asymmetric, the tooth cavity is deep, the number of teeth is small, and demolding is difficult.
I see several major problems in the precision forging of a hypoid bevel gear. First, the process plan must ensure complete filling of the tooth cavity. Second, the forged pinion must be removable from the die. Third, the die structure and process parameters must be optimized. Fourth, the plastic flow and stress-strain evolution must be understood. Fifth, tooth accuracy must be controlled within the required tolerance. These problems are summarized in Table 2.
| Problem area | Main difficulty | Consequence if not controlled |
|---|---|---|
| Process plan | Metal must flow into a deep, asymmetric spiral cavity | Incomplete filling, folding, excessive load |
| Demolding | Tooth geometry can mechanically lock the forging | Die damage, production stop, scrap |
| Die and process optimization | Blank shape, flash, friction, and temperature interact | Poor accuracy, short die life, high cost |
| Forming analysis | Three-dimensional flow is non-uniform and coupled with heat | Unclear design basis, trial-and-error production |
| Tooth accuracy | Elastic deformation, thermal expansion, and oxidation | Mismatch with the mating gear |
The development trend of hypoid bevel gear manufacturing is toward net shape or near-net shape forming. Enclosed die forging can produce complex parts in one operation, reduce flash, improve tooth precision, and avoid some heat treatment and early die failure problems. Orbital forging is another promising chipless method because it can reduce the forming force compared with conventional forging. Hybrid processes such as precision forging plus orbital finishing, precision forging plus lapping, precision forging plus precision cutting, and precision forging plus cold sizing are also developing rapidly. I consider these hybrid routes to be important because they combine the fiber continuity of forging with the accuracy of finishing operations.
Rigid-Viscoplastic Finite Element Basis
To analyze the hot precision forging of a hypoid bevel gear, I use the rigid-viscoplastic finite element method. This method neglects elastic deformation, treats the material as incompressible, and is suitable for large plastic deformation and hot forming. The basic assumptions I use are as follows: elastic deformation is neglected, body forces such as gravity and inertia are neglected, the material is homogeneous and initially isotropic, volume is constant during plastic flow, the material follows the Levy-Mises flow rule, and the rigid-plastic boundary is determined by the loading condition.
The equilibrium equation is
$$ \sigma_{ij,j} = 0 $$
The geometric compatibility equation is
$$ \dot{\varepsilon}_{ij} = \frac{1}{2}\left(u_{i,j} + u_{j,i}\right) $$
The constitutive equation for a rigid-viscoplastic material is
$$ \sigma’_{ij} = \frac{2\bar{\sigma}}{3\dot{\bar{\varepsilon}}}\dot{\varepsilon}_{ij} $$
where \(\sigma’_{ij}\) is the deviatoric stress tensor, \(\bar{\sigma}\) is the equivalent stress, and \(\dot{\bar{\varepsilon}}\) is the equivalent strain rate. The equivalent strain rate and equivalent stress are defined as
$$ \dot{\bar{\varepsilon}} = \sqrt{\frac{2}{3}\dot{\varepsilon}_{ij}\dot{\varepsilon}_{ij}} $$
$$ \bar{\sigma} = \sqrt{\frac{3}{2}\sigma’_{ij}\sigma’_{ij}} $$
The von Mises yield criterion is
$$ \bar{\sigma}^2 = \frac{3}{2}\sigma’_{ij}\sigma’_{ij} $$
The volume constancy condition is
$$ \dot{\varepsilon}_V = \dot{\varepsilon}_{ii} = 0 $$
For a rigid-plastic material, the flow stress depends on strain and temperature:
$$ Y = Y(\bar{\varepsilon}, T) $$
For a rigid-viscoplastic material, the flow stress also depends on strain rate:
$$ Y = Y(\bar{\varepsilon}, \dot{\bar{\varepsilon}}, T) $$
The Markov variational principle states that among all kinematically admissible velocity fields, the true solution makes the following functional stationary:
$$ \Pi = \int_V \bar{\sigma}\dot{\bar{\varepsilon}}\,dV – \int_{S_F} F_i u_i\,dS $$
For a rigid-viscoplastic material, the plastic work function is
$$ E(\dot{\varepsilon}_{ij}) = \int_0^{\dot{\varepsilon}_{ij}} \sigma’_{ij}\,d\dot{\varepsilon}_{ij} = \int_0^{\dot{\bar{\varepsilon}}} \bar{\sigma}\,d\dot{\bar{\varepsilon}} $$
The incompressibility constraint can be introduced by the Lagrange multiplier method:
$$ \Pi_L = \int_V \bar{\sigma}\dot{\bar{\varepsilon}}\,dV + \int_V \lambda \dot{\varepsilon}_V\,dV – \int_{S_F} F_i u_i\,dS $$
or by the penalty method:
$$ \Pi_P = \int_V \bar{\sigma}\dot{\bar{\varepsilon}}\,dV + \frac{\alpha}{2}\int_V \dot{\varepsilon}_V^2\,dV – \int_{S_F} F_i u_i\,dS $$
The Lagrange multiplier \(\lambda\) is equal to the hydrostatic pressure, while in the penalty method the hydrostatic pressure is
$$ \sigma_m = \alpha \dot{\varepsilon}_V $$
Using these equations, I can calculate the velocity field, strain rate field, deviatoric stress field, hydrostatic pressure, and final stress field. This gives a complete numerical description of the forging process.
Thermal Coupling
Hot forging of a hypoid bevel gear involves heat transfer between the blank and the dies, between the blank and the environment, and between the dies and the environment. Friction work and plastic deformation work are also converted into heat. Temperature changes affect flow stress, microstructure evolution, friction, and die stress. Therefore, I use a coupled thermal-mechanical analysis. The heat conduction equation is
$$ \rho c \frac{\partial T}{\partial t} = \lambda \left( \frac{\partial^2 T}{\partial x^2} + \frac{\partial^2 T}{\partial y^2} + \frac{\partial^2 T}{\partial z^2} \right) + \dot{q} $$
where \(\rho\) is density, \(c\) is specific heat, \(\lambda\) is thermal conductivity, \(T\) is temperature, and \(\dot{q}\) is the internal heat generation rate. The coupling procedure is iterative. First, I solve the velocity field for an initial temperature field. Then I solve the temperature rate field and update the nodal coordinates, equivalent strain, and related variables. Next I solve the temperature field and then solve the corresponding velocity field again. This process is repeated until convergence. In this way, the interaction between deformation and temperature is captured for the hypoid bevel gear forging process.
Process Parameter Selection and Blank Design
I selected the gear material, forging temperature, die material, die preheating temperature, lubricant, forging method, and equipment before building the finite element models. The material for the hypoid bevel gear pinion is a low-carbon alloy carburizing steel, 20CrMnTi. It is commonly used for automotive drive axle gears because it can provide high surface hardness, good core toughness, and high bending fatigue strength after carburizing and quenching. The mechanical properties at room temperature are listed in Table 3.
| Property | Value | Property | Value |
|---|---|---|---|
| Tensile strength | 1080 MPa | Impact energy | 55 J |
| Yield strength | 835 MPa | Impact toughness | 69 J/cm² |
| Elongation | 10% | Hardness | 217 HB |
| Reduction of area | 45% | Specimen size | 15 mm |
At room temperature, the elongation is only about 10%, so cold precision forging of the complete hypoid bevel gear tooth form would be very difficult. I therefore use hot precision forging. The forging temperature range of 20CrMnTi is generally 800 to 1200 °C, but there are brittle regions around 600 to 975 °C and near the melting point. In the range of 1000 to 1350 °C, dynamic recrystallization occurs, and the reduction of area can exceed 75%. This indicates good plasticity. To avoid oxidation, I select a forging temperature between 1000 and 1200 °C with a controlled heating atmosphere. The flow stress decreases as temperature increases, as shown in Table 4.
| Temperature (°C) | 1200 | 1100 | 1000 | 900 | 800 |
|---|---|---|---|---|---|
| Tensile strength (MPa) | 28 | 44 | 80 | 97 | 140 |
For the die material, I choose a hot-work die steel equivalent to 4Cr5MoSiV1 or AISI H13. It has good high-temperature strength, hardness, impact toughness, wear resistance, and thermal fatigue resistance. The allowable stress is about 1500 MPa. I preheat the dies to about 350 °C. Preheating reduces the temperature difference between the blank and the die, prevents excessively rapid cooling of the blank, and reduces thermal stress in the die. The lubrication requirement for hot precision forging of a hypoid bevel gear is demanding. The lubricant must have sufficient viscosity and activity, good lubrication in the forging temperature range, good release performance, good high-temperature wettability, and good thermal stability and cooling capacity. I select water-based graphite as the lubricant. The friction condition is represented by the shear friction model:
$$ \tau_f = m k $$
where \(m\) is the friction factor and \(k\) is the shear yield strength. In hot forging, the maximum friction condition corresponds to \(m = 1\), and the corrected friction coefficient is
$$ \mu’ = \frac{m}{\sqrt{3}} \approx 0.577 $$
In my simulations, I use a friction coefficient around 0.3 as a representative value for lubricated hot forging and later study the influence of friction coefficient from 0.1 to 0.3.
For the forging method, I use enclosed die forging, also called no-flash die forging. In this method, the upper and lower dies close first, and then one or two punches press the blank inside the closed cavity. The blank flows in multiple directions, and a complex shape can be formed in one operation. The total forging load is the sum of the die-closing force and the punch extrusion force. This method is suitable for the hypoid bevel gear because the tooth cavity is complex and the metal flow must be controlled. A schematic principle is described by the relationship
$$ P_{\text{total}} = P_{\text{closing}} + P_{\text{punch}} $$
For the forging equipment, I evaluate friction presses, crank presses, hydraulic presses, and hammers. A friction press has no fixed bottom dead center, so its load depends on the deformation energy. It is suitable for forgings with large deformation energy, large deformation stroke, or high load with small stroke. It also has good process adaptability, simple die installation, and a short dwell time. Therefore, I select a friction press. The forging force can be estimated by two common formulas. The first is
$$ P = \alpha \left(1 + 0.1\sqrt{\frac{V}{F}}\right)\sigma_b F $$
where \(\alpha\) is 4 for open-die forging and 5 for closed-die forging, \(F\) is the projected area including the flash land, \(V\) is the forging volume, and \(\sigma_b\) is the flow stress at the final forging temperature. The second formula is
$$ P = K A $$
where \(A\) is the projected area in cm² and \(K\) is an empirical coefficient. For a complex forging with high ribs and thin walls, \(K\) can be 120 to 150 kN/cm². For a clearer outline, \(K\) can be 80 kN/cm², and for a forging with generous fillets, \(K\) can be 50 kN/cm². The required press capacity is then
$$ P’ = \frac{P}{q} $$
where \(q\) is a deformation coefficient. For large deformation, \(q\) is about 0.9 to 1.1. My calculations are summarized in Table 5.
| Method | Condition | Estimated force | Equivalent tonnage |
|---|---|---|---|
| Formula 1 | 1100 °C | 4542 kN | about 454 T |
| Formula 1 | 1000 °C | 8259 kN | about 826 T |
| Formula 2 | K = 80 kN/cm² | 7599 kN | about 760 T |
| Formula 2 | K = 50 kN/cm² | 4750 kN | about 475 T |
From these estimates, I conclude that a press capacity above about 800 T is needed. Later finite element results refine this estimate. The forging drawing design for the hypoid bevel gear includes the parting line, machining allowance, draft, fillet radii, and shrinkage. I place the parting line at the largest horizontal projection, which is the large end section of the tooth form. For hot precision forging, the achievable tolerance is about \(\pm 0.2\) to \(\pm 0.4\) mm. This is not sufficient for the final hypoid bevel gear tooth accuracy, which may require about \(\pm 0.01\) to \(\pm 0.05\) mm. Therefore, I use hot precision forging followed by precision cutting or cold sizing. In this study, I simulate only the hot precision forging stage. I leave a small machining allowance of about 0.75 to 1.2 mm per side on the tooth form. The shrinkage is accounted for by
$$ L = (1 + \delta)l $$
where \(L\) is the hot forging dimension, \(l\) is the cold forging dimension, and \(\delta\) is the shrinkage. For steel, \(\delta\) is about 1.2% to 1.5%, and I use 1.5%. The fillet radii are designed as
$$ r = \text{machining allowance} + a $$
$$ R = (2 \text{ to } 3)r $$
where \(r\) is the external fillet radius, \(R\) is the internal fillet radius, and \(a\) is the corresponding corner radius or chamfer of the part. These fillets are important for the hypoid bevel gear because they allow smooth metal flow and continuous fiber lines, and they prevent incomplete filling at the intersection of the face cone and the back cone.
Blank Design and Forming Schemes
Blank design is critical for the hypoid bevel gear precision forging. The blank must be easy to locate in the die, must allow metal to fill the tooth cavity, and must produce a short flow path. The tooth form should be filled mainly by radial flow that is driven by axial compression of the two ends of the tooth body. This rule means that the blank must be contained within the geometric body bounded by the tooth root cone and the two cylindrical surfaces defined by the small-end and large-end root circles. If the blank is too large, it cannot be placed in the closed die before the dies close. If it is too small, the tooth cavity will not fill. The blank diameter is usually taken as 0.6 to 1.0 times the maximum section of the forging, and the height-to-diameter ratio should be about 0.5 to 3. The volume constancy condition is
$$ V_{\text{blank}} = V_{\text{forging}} $$
I designed several blank shapes and forming schemes. In one group, the punch acts from the small end. In another group, the punch acts from the large end. In a third group, two punches act from both ends. After evaluating the feasibility of each blank, I selected five representative forming schemes. These are listed in Table 6.
| Scheme | Forming method | Punch action | Blank type |
|---|---|---|---|
| 1 | Single-action punch | From small end | Cylinder at small-end root diameter |
| 2 | Single-action punch | From large end | Large-end diameter close to large-end root circle |
| 3 | Single-action punch | From large end | Cylinder larger than small end but smaller than large end |
| 4 | Single-action punch | From large end | Large-end diameter slightly larger than shaft diameter |
| 5 | Double-action punches | From both ends | Tapered blank |
Finite Element Modeling and Scheme Optimization
I built three-dimensional finite element models for the five schemes using a commercial forming simulation platform. I treated the dies as rigid bodies because die deformation is not the focus of this study. I treated the blank as a rigid-viscoplastic body because hot forging involves large plastic deformation. I meshed the blank and dies with tetrahedral elements. The blank had about 80,000 elements. I measured the minimum element edge length and set the simulation step to no more than one third of that length. I used a step size of 0.6 mm and saved data every ten steps. The material model for 20CrMnTi was entered from hot compression test data at different temperatures and strain rates. The die material was H13. The environment temperature was 20 °C, the initial blank temperature was 1100 °C, the die preheating temperature was 350 °C, the shear friction coefficient was 0.3, the heat transfer coefficient between the blank and the environment was 0.02, the heat transfer coefficient between the blank and the die was 11, and the fraction of plastic work converted to heat was 0.9. For schemes 1 to 4, the punch speed was 200 mm/s. For scheme 5, the large-end punch speed was 200 mm/s and the small-end punch speed was 225 mm/s. I also applied volume compensation during remeshing to preserve the blank volume. The strain rate was estimated by
$$ \dot{\varepsilon} = \frac{1}{L_0}\frac{dL}{dt} $$
For an initial length of 221 mm and a length change of 131 mm in about 2 s, the average strain rate is about 0.296 s⁻¹. This lies within the range of the material data and confirms that the punch speed is reasonable. The main simulation parameters are summarized in Table 7.
| Parameter | Value |
|---|---|
| Blank material | 20CrMnTi |
| Die material | AISI H13 |
| Blank temperature | 1100 °C |
| Die preheating temperature | 350 °C |
| Environment temperature | 20 °C |
| Shear friction coefficient | 0.3 |
| Blank-die heat transfer coefficient | 11 |
| Plastic work to heat fraction | 0.9 |
| Mesh type | Tetrahedral |
| Number of elements | About 80,000 |
| Step size | 0.6 mm |
I evaluated the schemes by tooth filling, equivalent stress, equivalent strain, and load-stroke curves. The results are summarized in Table 8. Scheme 1 gives the best overall performance. The tooth cavity is filled completely, the outline is clear, and metal flow is smooth and continuous. There is no folding or severe corner defect in the working tooth region. In schemes 2 and 3, the punch acts from the large end, and the deformation resembles upsetting. The middle of the blank bulges first, then the large end and small end fill. The flow path is long, and the deformation is non-uniform. Both ends show incomplete filling. In schemes 4 and 5, the large-end diameter differs from the large-end root circle, so a step or corner is created. Metal flows axially into the cavity and then radially, but near the step the flow reverses. This produces an obvious folding defect in the large-end tooth region. Scheme 1 avoids this problem because the blank is a cylinder with a diameter close to the small-end root circle. The flow is mainly axial at the small end and then radial toward the tooth cavity, so the path is short and direct.
| Scheme | Filling quality | Maximum equivalent stress | Maximum equivalent strain | Maximum load | Main defect |
|---|---|---|---|---|---|
| 1 | Complete tooth form, clear outline | 377 MPa | 8.54 | 198 T | Minor corner underfill at large end |
| 2 | Partial, both ends underfilled | 4880 MPa | 4.62 | Higher than scheme 1 | Large-end corner collapse |
| 3 | Partial, small end underfilled | 4300 MPa | 3.49 | Higher than scheme 1 | Non-uniform flow |
| 4 | Partial, large end folded | 2140 MPa | 8.53 | Higher than scheme 1 | Folding at diameter step |
| 5 | Partial, both ends folded | 1140 MPa | 4.95 | Higher than scheme 1 | Folding near both punches |
Scheme 1 has the lowest maximum equivalent stress, a high equivalent strain, and the lowest forming load. The stress value is 377 MPa, which is much lower than the values in the other schemes. This means the deformation resistance is lower and the metal fills the cavity more easily. The equivalent strain is 8.54, which indicates that the material undergoes sufficient cumulative deformation and the tooth form is well filled. The maximum load is 198 T, which is the lowest among the five schemes. This reduces the required press capacity, reduces energy consumption, and reduces die load. Therefore, I select scheme 1 as the optimal forming scheme for the hypoid bevel gear driving pinion.
Influence of Process Parameters
I also studied the influence of friction coefficient and forging temperature on the precision forging of the hypoid bevel gear. The friction coefficient depends on the lubricant, temperature, surface condition, and deformation speed. For water-based graphite, the friction coefficient can be in the range of about 0.05 to 0.19, but a representative value of 0.1 to 0.3 is often used in simulation. I performed simulations at friction coefficients of 0.1, 0.2, and 0.3 while keeping the forging temperature at 1050 °C. The results are summarized in Table 9. As friction increases, the maximum equivalent stress increases from 366 MPa to 441 MPa, and the forming load also increases. Higher friction increases the resistance to metal flow, makes filling more difficult, and requires higher die strength. Lower friction improves flow, reduces load, reduces die wear, and lengthens die life.
| Friction coefficient | Maximum equivalent stress | Forming load trend | Flow condition |
|---|---|---|---|
| 0.1 | 366 MPa | Lowest | Smooth and uniform |
| 0.2 | 428 MPa | Moderate | Acceptable |
| 0.3 | 441 MPa | Highest | More resistance and less uniform |
The friction effect can be expressed by the shear friction relation
$$ \tau_f = m k $$
where \(m\) increases with friction. I conclude that friction is a major factor in the hypoid bevel gear forging process. Good lubrication with a low friction coefficient is necessary to reduce load, improve filling, and extend die life.
I also studied forging temperature at 1000 °C, 1050 °C, and 1100 °C while keeping the friction coefficient at 0.3. The results are summarized in Table 10. As the forging temperature increases, the maximum equivalent stress decreases from 446 MPa to 377 MPa, and the forming load decreases. The reason is that higher temperature increases atomic mobility, reduces the resistance between atoms, promotes dynamic recovery and recrystallization, and increases grain boundary sliding and diffusion. These effects reduce flow stress and improve plasticity. Therefore, within the safe forging temperature range, a higher forging temperature is beneficial for the hypoid bevel gear. It improves metal flow, reduces forming load, and helps fill the tooth cavity.
| Forging temperature | Maximum equivalent stress | Forming load trend | Plasticity |
|---|---|---|---|
| 1000 °C | 446 MPa | Highest | Lower |
| 1050 °C | 441 MPa | Moderate | Moderate |
| 1100 °C | 377 MPa | Lowest | Higher |
Detailed Numerical Simulation of the Selected Scheme
After selecting scheme 1, I carried out a detailed numerical simulation of the hypoid bevel gear precision forging process. I analyzed the mesh evolution, equivalent stress field, equivalent strain field, velocity field, temperature field, and load-stroke curve. The finite element model used the same assumptions and parameters as in the optimization study. The minimum element edge length was smaller, and the step size was 0.6 mm. The simulation saved data every ten steps. I now describe the forming characteristics and laws obtained from this detailed analysis.
Mesh Evolution
At the beginning, the blank mesh is uniform. As the punch moves downward, metal enters the tooth cavity from the small end. The surface mesh near the small end becomes denser than the interior mesh. This occurs because the surface metal contacts the die cavity, deforms severely, and is remeshed. As deformation continues, the dense surface mesh extends from the small end toward the large end. When the tooth form is completely filled, the mesh density in the tooth region is clearly higher than that in the interior and in the shaft region. This confirms that filling proceeds from the small end to the large end. The metal flow is smooth, the tooth outline is clear, and a complete tooth form is obtained.
Equivalent Stress
The equivalent stress distribution reflects the deformation resistance and stress concentration. In the initial stage, the stress action zone appears near the small end. The stress value decreases from the surface toward the interior. The cylindrical part of the blank must pass through this zone to enter the cavity, so the deformation is intense and the stress increases rapidly. As the punch moves further, the small end fills first. Because the tooth root is the first part of the tooth form to contact the die cavity, the stress at the tooth root is higher than that at the tooth tip. After the small end is filled, metal flows toward the large end. The stress action zones near the small end and near the large-end step move toward the center of the blank. The lowest stress occurs in the central region. When the tooth form is nearly full, the last regions to fill are the tooth tips and the corner regions at both ends. The maximum stress occurs at the corner of the tooth form and reaches about 279 MPa. The shaft region does not deform, so its stress remains low. The stress stages are summarized in Table 11.
| Stage | Stress location | Stress trend | Interpretation |
|---|---|---|---|
| Initial | Near small end | Increases from inside to outside | Small-end metal enters cavity first |
| Small-end filling | Tooth root | Root higher than tip | Root contacts die first |
| Large-end filling | Root and step | Stress zones move inward | Flow continues toward large end |
| Final filling | Corners and tips | Maximum at corners | Last unfilled regions require high pressure |
Equivalent Strain
The equivalent strain distribution reflects the severity and accumulation of deformation. In the initial stage, the small-end region has the highest strain because it deforms first. The strain decreases from the small end toward the large end. The tooth root has a higher strain than the tooth tip. The interior of the tooth body has a lower strain than the surface. As filling proceeds, the small end accumulates more strain than the large end. The maximum equivalent strain occurs at the tooth root near the small end and reaches about 4.58. The shaft region does not deform, so its strain is almost zero. The strain stages are summarized in Table 12.
| Stage | High-strain region | Low-strain region | Maximum strain |
|---|---|---|---|
| Initial | Small-end root | Interior and large end | Increasing |
| Small-end filling | Small-end root and tip | Shaft and center | About 4.0 |
| Large-end filling | Small end | Large end | About 4.5 |
| Final filling | Small-end root | Shaft | 4.58 |
Velocity Field
The velocity field directly shows the metal flow pattern. At the beginning, the cylindrical part of the blank moves axially downward. The tooth part flows radially and downward from the small end toward the large end. Because the small end fills first, the velocity decreases from the small end toward the large end. During filling, metal flows radially toward the tooth root and tooth tip and also tangentially because of the spiral tooth form. The small end fills first, and then metal flows through the small-end root and tip toward the large end in a radiating spiral pattern. The flow direction is clear, and the maximum and minimum velocities do not change greatly. There is no backflow and almost no velocity interweaving. This indicates that the velocity distribution is reasonable and uniform. The shaft region does not deform, so its velocity is nearly zero. Because the pinion tooth surfaces are asymmetric, the tangential motion is stronger when the tooth tip is filled. This can cause incomplete filling at the intersection of the face cone and the back cone at the large end. Therefore, I intentionally design a fillet at this location in the forging drawing.
Temperature Field
Temperature is an important factor in hot forging of a hypoid bevel gear. Heat transfer occurs between the blank and dies, the blank and the environment, and the dies and the environment. Friction work and plastic deformation work also generate heat. The temperature field analysis shows that the outer surface of the blank cools gradually and the cooling front moves inward. This occurs because the outer surface contacts the die and loses heat more easily. In the initial stage, the cylindrical surface loses heat because it contacts the die wall, but the tooth region still has a gap and cools slowly. The temperature remains above 1000 °C, which is within the preferred hot forging range. As the punch moves, the small-end root contacts the die and loses heat, but deformation work and friction work compensate for part of the loss, so the temperature remains about 1000 to 1050 °C. The large-end root also loses heat, but the deformation there has just begun, so the temperature is about 950 to 1000 °C. The tooth tips do not contact the die cavity at this stage and remain at about 1050 to 1100 °C. After the small end fills, the small-end root continues to lose heat but is compensated by deformation work, so it remains at about 950 to 1000 °C. The small-end tip loses heat with less compensation and drops to about 900 to 950 °C. The large-end root remains at about 950 to 1000 °C. The large-end tip is still about 1050 to 1100 °C. When the tooth form is nearly complete, the small-end tip continues to cool to about 800 to 900 °C, but it is already formed and further cooling has little effect on filling. The large-end tip is the last to form and remains at about 1050 to 1100 °C. All tooth roots remain at about 950 to 1000 °C because of deformation heating. The highest temperature is at the large-end tooth tip. Overall, the temperature difference in the blank is less than about 200 °C. Most of the tooth surface remains between 900 and 1000 °C. This uniform temperature distribution helps avoid large residual thermal stresses, improves forging quality, and prolongs die life. Only a few regions fall below 900 °C because they contact the die early and deform little in the final stage. The temperature stages are summarized in Table 13.
| Stage | Coolest region | Hottest region | Typical range |
|---|---|---|---|
| Initial | Cylindrical surface | Tooth region | Above 1000 °C |
| Small-end filling | Upper cylindrical surface | Small-end root | 1000 to 1050 °C |
| Large-end filling | Small-end tip | Large-end tip | 900 to 1100 °C |
| Final filling | Small-end tip | Large-end tip | 800 to 1100 °C |
Load-Stroke Curve
The load-stroke curve of the punch can be divided into three stages: initial stage, filling stage, and final stage. In the initial stage, the blank has not yet entered the tooth cavity. The deformation is similar to upsetting. The load increases linearly. In the filling stage, metal flows into the tooth cavity from the small end to the large end. This stage lasts the longest. The load increases steadily, and the curvature of the curve increases gradually. This means deformation becomes more difficult as more of the cavity is filled. In the final stage, the tooth form is almost full. Only a small amount of corner and tip volume remains. The free surface is very small. To fill these last regions, the metal must overcome the hydrostatic pressure of the already filled material. Therefore, the load rises sharply to a maximum of about 153 T, or 1530 kN. Near the end, the curve shows a sudden drop and rise because the mesh becomes severely distorted and is remeshed. The load stages are summarized in Table 14.
| Stage | Approximate step range | Load behavior | Physical meaning |
|---|---|---|---|
| Initial | Start to step 10 | Linear increase | Upsetting before cavity contact |
| Filling | Step 10 to step 175 | Steady increase, curvature grows | Tooth cavity fills from small end to large end |
| Final | Step 175 to step 196 | Sharp increase | Last corners and tips fill under high pressure |
Forming Characteristics and Laws
From the detailed simulation, I summarize the main forming characteristics of the hypoid bevel gear precision forging as follows. First, the surface mesh in the tooth region becomes denser than the interior mesh, which reflects severe surface deformation and confirms that the tooth form is filled from the small end toward the large end. Second, the equivalent stress and equivalent strain increase throughout the process. The highest stress and strain appear first at the small-end tooth root. During filling, stress and strain decrease from the small end to the large end and from the tooth root to the tooth tip. Third, the velocity field shows a radiating spiral flow from the small end toward the large end. The flow direction is clear, and there is no backflow or velocity interweaving in the selected scheme. Fourth, the temperature difference in the blank is small, and most of the tooth surface remains within the safe forging temperature range. Fifth, the load-stroke curve has three stages, and the final stage requires a sharp increase in load to fill the last corner regions. These characteristics are important for die design, press selection, and process control of the hypoid bevel gear.
Equipment Capacity and Experimental Context
In the selected scheme, the lower die carries both the die-closing force and the punch extrusion force. The simulated load on the lower die reaches about 702 T. To ensure that the forging is accurate and the dies are pressed tightly together, a press with a slightly larger capacity is needed. Therefore, an 800 T friction press is a reasonable choice. This agrees with the theoretical estimates from the forging force formulas and with the simulation results. The agreement between the analytical calculation and the numerical simulation increases my confidence in the finite element model for the hypoid bevel gear. In earlier trial production of a hypoid bevel gear, the die structure and demolding method were not fully successful, and the desired pinion was not obtained. The present numerical study provides a better understanding of metal flow, load evolution, and process parameters. It can guide the redesign of the die structure and the selection of the press for future experiments on the hypoid bevel gear.
Discussion of Process Optimization
The optimization of the hypoid bevel gear precision forging process depends on several coupled factors. The blank shape must be compatible with the die cavity and the punch motion. The friction coefficient must be low enough to allow filling but not so low that the lubricant fails at high temperature. The forging temperature must be high enough to reduce flow stress but not so high that oxidation or grain growth becomes excessive. The die preheating temperature must reduce thermal shock but not soften the die. The punch speed must be compatible with the press and the material strain-rate sensitivity. In my simulations, I found that scheme 1 with a small-end punch gives the most stable flow pattern. The metal moves axially at the small end, then radially and tangentially into the spiral tooth cavity, and finally fills the large end. This flow path is shorter than the flow path in schemes 2 to 5. A shorter flow path reduces the chance of folding, lowers the forming load, and improves the final tooth quality.
I also note that the asymmetric tooth surfaces of the hypoid bevel gear pinion cause a non-uniform tangential velocity during the final filling of the tooth tip. This is a direct result of the different curvatures and pressure angles on the two sides of the pinion tooth. Therefore, the die design must allow enough clearance and fillet radius at the large-end face cone and back cone intersection. If the fillet is too small, the corner will not fill. If it is too large, the machining allowance and material loss will increase. A balanced fillet design is necessary. The shrinkage must also be compensated because the hypoid bevel gear is forged at high temperature and then cooled. The shrinkage equation
$$ L = (1 + \delta)l $$
is used to scale the die cavity. For steel, a shrinkage of about 1.5% is appropriate. The die cavity must also account for the machining allowance and the subsequent finishing operation. Because hot precision forging cannot directly achieve the final tooth accuracy of a hypoid bevel gear, I recommend hot precision forging followed by precision cutting or cold sizing. This hybrid route combines the fiber continuity of forging with the accuracy of finishing.
Comparative Summary of Schemes
To make the optimization decision clear, I present a comparative summary of the five schemes in Table 15. The table includes the filling quality, defect type, maximum stress, maximum strain, and maximum load. Scheme 1 is superior in all critical respects. It produces a complete tooth form, avoids severe folding, has the lowest stress, has a high strain, and requires the lowest load. Schemes 2 and 3 suffer from long flow paths and non-uniform deformation. Schemes 4 and 5 suffer from folding caused by diameter steps and reverse flow. Therefore, I select scheme 1 for the hypoid bevel gear driving pinion.
| Scheme | Flow path | Filling | Folding risk | Stress | Strain | Load | Decision |
|---|---|---|---|---|---|---|---|
| 1 | Short, small end to large end | Complete | Low | Lowest | High | Lowest | Selected |
| 2 | Long, large end to both ends | Incomplete | Moderate | Very high | Moderate | High | Rejected |
| 3 | Long, large end to small end | Incomplete | Moderate | High | Low | High | Rejected |
| 4 | Reverse near step | Partial | High | High | High | High | Rejected |
| 5 | Reverse near both punches | Partial | High | Moderate | Moderate | High | Rejected |
Key Equations Used in My Analysis
For clarity, I collect the main equations used in my study. The equilibrium, compatibility, constitutive, yield, and volume constancy equations are
$$ \sigma_{ij,j} = 0 $$
$$ \dot{\varepsilon}_{ij} = \frac{1}{2}\left(u_{i,j} + u_{j,i}\right) $$
$$ \sigma’_{ij} = \frac{2\bar{\sigma}}{3\dot{\bar{\varepsilon}}}\dot{\varepsilon}_{ij} $$
$$ \bar{\sigma} = \sqrt{\frac{3}{2}\sigma’_{ij}\sigma’_{ij}} $$
$$ \dot{\bar{\varepsilon}} = \sqrt{\frac{2}{3}\dot{\varepsilon}_{ij}\dot{\varepsilon}_{ij}} $$
$$ \dot{\varepsilon}_{ii} = 0 $$
The variational functional is
$$ \Pi = \int_V \bar{\sigma}\dot{\bar{\varepsilon}}\,dV – \int_{S_F} F_i u_i\,dS $$
The heat conduction equation is
$$ \rho c \frac{\partial T}{\partial t} = \lambda \nabla^2 T + \dot{q} $$
The friction relation is
$$ \tau_f = m k $$
The forging force estimates are
$$ P = \alpha \left(1 + 0.1\sqrt{\frac{V}{F}}\right)\sigma_b F $$
$$ P = K A $$
$$ P’ = \frac{P}{q} $$
The shrinkage and fillet relations are
$$ L = (1 + \delta)l $$
$$ r = \text{machining allowance} + a $$
$$ R = (2 \text{ to } 3)r $$
The strain rate estimate is
$$ \dot{\varepsilon} = \frac{1}{L_0}\frac{dL}{dt} $$
These equations form the basis for the thermal-mechanical finite element simulation of the hypoid bevel gear precision forging process.
Final Assessment
In my assessment, the precision forging of a hypoid bevel gear driving pinion is feasible when the blank shape, punch direction, friction condition, and forging temperature are properly controlled. The best scheme uses a single-action punch from the small end and a cylindrical blank whose diameter is slightly smaller than the small-end root circle. This scheme produces a complete tooth form with clear outlines, low stress, high strain, and a relatively low forming load. The metal flow is continuous and follows the spiral tooth cavity without severe folding. The temperature field remains within the safe forging range, and the final load is compatible with an 800 T friction press. The numerical results also show that lower friction and higher forging temperature reduce the forming load and improve filling. These findings provide a useful basis for die design and process planning for the hypoid bevel gear.
I conclude that precision forging can improve the performance of a hypoid bevel gear by producing continuous metal fibers along the tooth profile, increasing tooth root bending fatigue strength, and improving tooth surface contact fatigue strength. Compared with cutting, precision forging can also raise material utilization, reduce production cost, and increase productivity. However, the asymmetric tooth surface, the offset geometry, and the deep spiral cavity of the hypoid bevel gear require careful process design. The die must be designed with appropriate fillets, shrinkage compensation, and demolding clearance. The process must be controlled with good lubrication and a suitable forging temperature. Numerical simulation is an effective tool for understanding the complex flow and for optimizing the process before physical trials. I believe that the combination of precision forging, numerical simulation, and hybrid finishing operations will play an increasingly important role in the future manufacturing of hypoid bevel gears.