# Comparison of Approaches to Determining the Coefficient of Friction in Stretch-Forming Conditions

## TL;DR

This paper compares different methods for calculating friction in metal forming, using experiments on steel sheets with modified surfaces.

## Contribution

A novel comparison of analytical models for determining the coefficient of friction in stretch-forming conditions using a self-developed testing apparatus.

## Key findings

- The coefficient of friction varied by approximately 40% across different analytical models.
- Surface modification via ion implantation increased average roughness by 19% in friction tests.
- Friction significantly reduced kurtosis values, indicating changes in surface topography.

## Abstract

Control of the friction process in stretch-forming conditions, when creating sheet metal, is essential for obtaining components of the quality required. This paper presents an approach to modelling the friction phenomenon at the rounded edges of stamping dies. The aim of the study is to compare the coefficient of friction (CoF) determined from numerous analytical models available in the literature. Experimental studies were conducted using self-developed bending under tension friction testing apparatus. The test material was low-carbon DC01 steel sheeting. Tests were conducted under lubricated conditions, using industrial oil intended for deep drawing operations. The surfaces of countersamples made of 145Cr6 substrate were modified using the ion implantation of Pb (IOPb) and electron beam melting processes. Variation in the CoF in BUT tests was related to continuous deformation-induced changes in surface topography and changes in the mechanical properties of sheet metal due to the work-hardening phenomenon. Under friction testing with a stationary countersample, the largest increase in average roughness (by 19%) was found for the DC01/IOPb friction pair. The friction process caused a significant decrease in kurtosis values. The results show that the difference between the highest and lowest CoF values, determined for the analytical models considered, was approximately 40%.