High-hardness spiral bevel gears are commonly finished after heat treatment, when mechanical generating and double-side milling become less forgiving: cutting forces and tool wear rise, the curved flank is difficult to reproduce, and profile errors accumulate before the final meshing check. Electrochemical machining (ECM) shifts the problem from mechanical cutting to electrochemical shape transfer. The workpiece is the anode, the tool is the cathode, and material is removed by anodic dissolution as electrolyte flows through the small working gap. Because removal depends on current density rather than hardness, a quenched gear steel can be machined with a cathode whose working surface matches the required flank—provided the gap and flow field are controlled.
Process Principle and Tooling Requirements
ECM does not press a hard tool into the gear. The cathode is fed toward the workpiece while a DC voltage drives dissolution, and the electrolyte carries away ions and heat. The final anode surface tends to follow the cathode surface offset by the equilibrium gap, a gap set by voltage, feed rate, and electrolyte conductivity. For a spiral bevel gear, the flank is a three-dimensional curved surface with varying curvature from toe to heel and from root to tip. A simple planar cathode cannot reproduce it. The cathode head needs a conjugated three-dimensional profile and internal channels that deliver electrolyte uniformly to every part of the gap. If flow is starved in one region, the local gap changes, current density becomes nonuniform, and the tooth profile deviates.
Tooth-Surface Model and Cathode Profile
The gear model is based on a Gleason-type spiral bevel gear. A coordinate system OgXgYgZg is placed at the intersection of the end face and the reference axis. Reference points are selected on the tooth flank and offset along the local normal to form an offset point set; curves through these points describe the flank. In the design case, four curves were used, with five points per curve, giving a 4 × 5 sampling grid of 20 points. The conjugate surface equations were then solved for the cathode surface. Sequential quadratic programming and numerical optimization were used to refine the full-tooth conjugate point set, with Isight providing the global convergence and optimization environment for the cathode head model.
The point set is not the final cathode surface; it is the starting geometry from which the conjugate cathode surface is solved under the equilibrium gap. The working surface is therefore offset from the desired anode surface by that gap, rather than being a direct negative copy of the gear tooth.
| Parameter | Value |
|---|---|
| Gear system | Gleason |
| Large-end module | 2.5 mm |
| Number of teeth | 25 |
| Pressure angle | 20° |
| Clearance coefficient | 0.188 |
| Addendum coefficient | 0.85 |
| Helix angle | 35° |
The addendum coefficient of 0.85 and clearance coefficient of 0.188 are chosen to preserve meshing clearance so that thermal expansion during operation does not cause tip interference. The 35° helix angle produces the curved tooth trace that makes the flank geometrically complex and makes a point-based surface model preferable to a simple analytic sweep.
Cathode Assembly and Flow Path
The cathode is organized as four functional groups: a profiled head, a shank, an internal cavity, and a clamping interface. The head and shank are separable, so the profiled head can be replaced without rebuilding the entire tool. Six threaded holes in the shank connect the assembly to the machine tool. The internal cavity is not a simple reservoir; it is a distribution network. Electrolyte enters the main channel and fills its upper section. A flow-control surface in that section suppresses jetting and helps equalize pressure. Four feed holes on the wall lead to four branch channels, and outlet holes release the electrolyte into the working gap. The branched layout is intended to avoid dead zones across the tooth surface, where local starvation would otherwise change the gap. After machining, residual electrolyte exits through a return port in the reverse direction, allowing collection and reuse.
Sealing, Workpiece Contact, and Machine Integration
Because ECM uses a conductive electrolyte, the cathode body must be electrically isolated from parts that are not intended to be active. A transparent glass protective cover surrounds the outside of the assembly; transparency in this context is practical rather than decorative, since it allows visual inspection of leakage or sludge accumulation. Two insulating sealing rings of different diameters close the top and bottom interfaces. The workpiece side uses a square base with a locating boss and a conductive plate. The boss fixes the gear in the correct angular and axial position, while the conductive plate carries the positive potential to the workpiece. Six fastening threaded holes under the base secure the assembly to the machine table. This arrangement separates the functions of positioning, electrical contact, and sealing, which is important because a conductive chip or a damaged seal can create a stray current path and damage the flank.
Experimental Validation
The cathode was tested on a quenched 20CrMnTi gear blank. The electrolyte was a mixed solution of 25% NaNO3 and 10% NaCl, maintained at 30 ± 5 °C; the mixed electrolyte was selected for higher conductivity and machining efficiency. The gear flank was machined with the profiled cathode, and the result was inspected on a coordinate measuring machine. The profile trace was evaluated over a defined measurement length La, with ffa denoting the profile curve and H1 the profile deviation. At equal La, the ECM gear showed a smaller H1 than the conventionally generated comparison gear, indicating a tighter profile.
Surface roughness was measured at 20 points across four traces. Averaging the 20 values in each data set gives the comparison in Table 2. The ECM samples average 1.00 µm, versus 1.35 µm for the generating process; the maximum is also lower, 1.47 µm versus 1.83 µm. The comparison is a point-based roughness summary, not a full areal texture specification, but it supports the profile result: the electrochemical process produced a smoother flank under the tested condition.
| Metric | ECM cathode | Conventional generating | Difference |
|---|---|---|---|
| Measured points | 20 | 20 | — |
| Mean roughness (µm) | 1.00 | 1.35 | 0.35 µm lower (≈26%) |
| Minimum measured (µm) | 0.48 | 0.83 | — |
| Maximum measured (µm) | 1.47 | 1.83 | — |
Interpretation and Limits
The design shows that the hardness of a quenched spiral bevel gear need not be the limiting factor in finishing. Once the cathode profile is generated from a sampled tooth surface and the electrolyte is distributed through a branched cavity, ECM can copy the curved flank without the mechanical wear that affects milling cutters. The lower roughness and smaller profile deviation at equal measurement length are consistent with a stable gap and uniform flow in the tested geometry.
Those conclusions are bounded by the validation conditions. The test used one material, quenched 20CrMnTi, one Gleason-type gear geometry, and a fixed electrolyte composition and temperature window. The data do not establish long-run cathode life, gap stability over a production batch, or capability across other modules, tooth counts, and spiral angles. The roughness mean is an arithmetic summary of 20 points, not a full areal parameter. In production, filtration, temperature control, and short-circuit protection remain part of the process window because they directly affect the local gap that the cathode was designed to hold.
The cathode assembly therefore combines a point-generated working surface with a flow distribution network and an insulated workpiece interface; each addresses a separate failure mode in ECM of spiral bevel gears.