Spiral bevel gears transmit torque between intersecting shafts through teeth whose traces curve across the face width. This curved geometry produces gradual engagement, a larger contact area than straight bevel teeth, and a higher axial contact ratio, which explains their use in high-speed, heavy-duty reducers. The same geometry also makes contact nonlinear. The active contact area moves across the flank during each mesh cycle, load alternates between single- and double-tooth contact, and local edge contact can raise stress even when the nominal torque is unchanged. Classical unloaded tooth contact analysis, associated with Gleason and widely used for spiral bevel and hypoid gears, predicts where the contact pattern lies under light load, but it does not resolve load-induced deflection, mesh-stiffness variation, or dynamic stress peaks. A loaded static and dynamic finite element analysis is therefore needed to connect tooth geometry to contact-fatigue behavior.
Geometry, material, and contact model
A representative case is the power-input spiral bevel pinion of a reduction mechanism, connected directly to the motor output and operating at relatively high speed. The pair has a 90° shaft angle, 14 pinion teeth and 59 wheel teeth, a transverse module of 5 mm, a pinion pitch cone angle of 10.7131°, and a 50 mm face width. The mid-face spiral angle is 35°, the normal pressure angle is 20°, and the tooth height is an equal-clearance taper with an addendum coefficient of 0.85 and a clearance coefficient of 0.188. Radial and tangential modification are applied to the pinion. The pinion is overhung and the wheel is straddle-mounted. The accuracy grade is 7-7-6c. Both parts are 20Cr2Ni4 steel, carburized and quenched, with a surface hardness of 62–64 HRC on the pinion and 57–59 HRC on the wheel.
| Parameter | Value |
|---|---|
| Shaft angle | 90° |
| Normal pressure angle | 20° |
| Addendum coefficient | 0.85 |
| Clearance coefficient | 0.188 |
| Spiral angle at mid-face width | 35° |
| Modification | Radial and tangential |
| Tooth-height type | Equal-clearance tapered |
| Pinion / wheel teeth | 14 / 59 |
| Transverse module | 5 mm |
| Pinion pitch cone angle | 10.7131° |
| Face width | 50 mm |
| Pinion tangential modification | 0.105 |
| Pinion radial modification | 0.3760 |
| Support | Pinion overhung; wheel straddle-mounted |
| Accuracy grade | 7-7-6c |
| Material | 20Cr2Ni4, carburized and quenched |
| Surface hardness | 62–64 HRC (pinion); 57–59 HRC (wheel) |
The elastic properties used in the contact model are E = 206.75 GPa, ν = 0.3, and ρ = 7.85 × 103 kg/m3. Tooth surfaces were generated from the tooth trace and profile and assembled into a solid model. A 0.5 mm mesh was used in the contact zone. A sensitivity check from 0.5 million to 3 million elements showed that stresses at three arbitrary flank points changed by less than 5% when the element count was increased by 50–100% per step; at that level, the contact solution was effectively mesh-independent. Surface-to-surface contact was defined between the pinion flank and the wheel flank, with the pinion treated as the contact surface and the wheel as the target surface. The nonlinear contact constraints were solved with the augmented Lagrangian method, which allows the contact area to evolve while enforcing non-penetration. Static solutions were sampled at three rotation positions and for three adjacent teeth.
Static contact pattern and load transfer
At the sampled rotation positions, three adjacent teeth were simultaneously engaged in different stages of mesh. The tooth entering the contact zone showed a contact stress that increased with rotation, while the tooth leaving the zone showed a decreasing stress. The computed contact patches were elongated along the tooth trace rather than concentrated in a small ellipse. This line-like, spread-out pattern is consistent with the axial overlap of spiral bevel teeth and helps distribute load across the face width. The instantaneous contact is therefore not a single point or a narrow ellipse, but a set of overlapping zones whose size and position change with rotation.
The main loaded tooth also showed contact at the addendum edge during part of the cycle. Edge contact is undesirable because it concentrates stress at a geometric discontinuity, reduces the effective contact area, and can initiate pitting or tooth-tip fracture under high-speed, heavy-load operation. The geometric remedy indicated by the analysis is to grind the tooth flanks so that the addendum edge conforms better to the mating surface and the contact pattern does not terminate abruptly at the tip. Grinding also improves profile accuracy and surface integrity, both of which support contact-fatigue life.
Dynamic meshing and the single/double-tooth cycle
In operation, the pair does not remain in a fixed contact state. Load alternates between single-tooth and double-tooth contact, and mesh stiffness changes periodically with rotation. This stiffness variation, together with the periodic load, excites vibration and makes a purely static solution insufficient for the stress history. The dynamic model was derived from the static contact model by applying an angular velocity to the pinion and replacing the fixed connection with a revolute joint between pinion and wheel, so that the pair could rotate under contact.
The mesh cycle resolved into single-tooth contact, double-tooth meshing-in, double-tooth contact, a further double-tooth contact stage, double-tooth meshing-out, and single-tooth contact. Equivalent-stress contours at these stages showed an approximately periodic variation in maximum contact stress. The highest values occurred during the transitions, when load was transferred to or from a single tooth. When two teeth shared the load, the maximum contact stress was about 40% lower than in the single-tooth condition. This load-sharing effect is one reason spiral bevel gears can transmit high torque smoothly, but it also means that fatigue assessment must use the stress cycle rather than a single peak value. The 40% reduction is specific to the modeled pair and operating state; it follows from the contact ratio and load distribution and is not a universal constant.
Speed dependence and contact-fatigue design
At constant torque, the maximum tooth-surface contact stress increased approximately exponentially with rotational speed. At low speed, impact loading was small, stress and strain varied modestly across the mesh phases, and transmission was relatively smooth. At high speed, the stress and strain range across the mesh cycle widened. The mechanism is dynamic amplification: higher speed increases the rate of load transfer, amplifies the effect of mesh-stiffness variation, and raises the impact component of the contact force. For a gear already operating near its contact-fatigue limit, this speed penalty can reduce service life even when the static torque rating is unchanged.
Several design consequences follow. The contact pattern should be centered and kept away from the addendum edge; grinding and edge-profile control are effective because they reduce the geometric stress concentration at the tip. The contact-fatigue assessment should use the dynamic stress cycle, including the single-tooth transition peaks and the speed-dependent increase, rather than only the static contact stress at rated torque. The load-sharing benefit of double-tooth contact should be preserved by controlling tooth spacing, modification, and assembly alignment; if the contact pattern shifts, the intended sharing benefit can deteriorate. For high-speed input stages, surface hardness and finish also matter because contact stress is concentrated in a thin surface layer. Carburizing and quenching to 62–64 HRC on the pinion, as in this case, provides subsurface support, but it does not compensate for a poor contact pattern.
Scope and interpretation
The quantitative results apply to the specified 14/59 pair, the 50 mm face width, the 90° shaft angle, 20Cr2Ni4 carburized steel, and the modeled load and speed conditions. They should not be read as a general law for all spiral bevel gears. What transfers to other designs is the mechanism: curved teeth create elongated, overlapping contact zones; multiple teeth share load; the single-tooth transitions are the critical part of the cycle; and speed raises the dynamic component of contact stress. Within that scope, the analysis supports a design workflow in which loaded contact simulation, dynamic mesh analysis, and grinding or edge control are used together rather than as separate checks.