Design of Experiment – Planet Gear

Increase Process Capability by adjusting
machine input parameters for surface finish

The manufacturing process included the use to CNC Mill to achieve the surface finish on a sun gear and outer ring developed by one of our customers. The experiment involved testing of surface roughness as a as per ASME Y14.36M. In order to achieve the required surface finish, the manufacturing process was proving to be inefficient on terms of tool travel time, lubrication, and tool wear. Due to the tool travel time and tool wear the component surface finish was failing to meet a CPK of 1.9. A complete machine servicing was conducted, and PM were redefined however, the output function was difficult to accomplish due to the resources used.

As per pareto analysis and time study analysis following observation were made:
1.Manufacturing cost per component demonstrated in an increasing trend.
2.High cycle time per process.
3.Low tooling life.

A surface response regression based on Design of Experiments was conducted where the tool travel path was determined based on X, Y, Z coordinates. The responses were noted as follows:
1. Tool Wear
2. Cycle Time
3. Surface Finish
4. The DOE factors included the following:
Tools: SS and Tungsten Carbide
Machine RPM: Continuous Factors
Lubricant: Synthetic, Semi-Synthetic and Organic
Tool Travel Path: Pattern A, Pattern B, Circular Pattern A, Circular Pattern B

The study developed surprising results based on tool wear and tool path. It was evident that the tool wear was a function of all other 3 factors. This resolved minimizing the cost of resources. The tool path was decided based on the Optimization Plot, Interaction Plot and Contour Plot.

The resultant CPK was at 2.1 within the controlled limits.