Computational generation of wear maps for H13 tool steel in open die hot forging

  • James Marashi

Student thesis: Doctoral Thesis


Forging tools are associated with as much as 40% of the costs any forging operation and consequently understanding the causes of their failure is key to improving productivity. The literature suggests that wear is responsible for 70%of failures in hot forging with abrasive and adhesive wear being the main failure modes in open die forging.Understanding the modes and mechanisms with which wear occurs on worn surfaces and contact faces is essential to minimise or eliminate the product defects and improve the quality. Tool wear phenomena can be understood and represented by wear maps for different materials (generated using a lab-based pin on disc).Such maps illustrate the wear mechanisms and wear progress from mild to transition and severe.However, some researchers prefer to explain wear behavior using analytical methods instead of empirical wear map. This thesis argues that, firstly, the pin on disc method, often used to generate wear maps, is not reliable, and produces many errors and is not representative of the industrial process. Secondly, wear maps created from a mathematical model alone (i.e. without physical trial cannot truly capture the wear characteristic of the material. As an alternative, this thesispresents a series of abrasive and adhesive wear maps created using modifiedArchard mathematical model that is validated with a series of physical forging trials.;The Archard mathematical model subroutine was embedded in the DEFORM FE simulation software. A series of FE simulations implemented a full factorial design of experiment with furnace temperature and energy in the screw press as the main variables. The FEA results were validated with a series of physical trials. A similar experiment was repeated after nitriding the tool by 0.1mm case depth. The comparison between the FE simulations and physical trials showed a good correlation of 80-90% for abrasive wear on un-nitrided tools and 70-84% on nitrided tools. While the correlation of 80-85% for adhesive wear on un-nitrided tools and 70-85% on nitrided tools was achieved. These comparisons then were used to produce a series of wear maps.This novel method of wear map generation can be used to optimise H13 tool steel performance and make the manufacturing process more cost effective. The optimised and predicted wear conditions help to minimise tool wear and improve the quality of obtained parts. Designer by having all the information using FE simulation alongside the wear maps can make the right decision in design and material selection. Potentially this methodology could also be used to compare different die materials, lubricants, and coatings.
Date of Award12 Apr 2019
Original languageEnglish
Awarding Institution
  • University Of Strathclyde
SponsorsUniversity of Strathclyde & EPSRC (Engineering and Physical Sciences Research Council)
SupervisorPaul Xirouchakis (Supervisor) & Remi Christophe Zante (Supervisor)

Cite this