Tolerance Analysis Framework for Cutting Tool Interface Design

Licentiate thesis, 2019

Tolerance analysis of cutting tool interface designs is a field with many opportunities for further development of existing methodologies. Cutting tool interface designs require multiple contacting surfaces, allocating the stress generated from the cutting forces, to avoid excessive deformation of the interface. With numerous contact areas, the insert will be overdetermined in the interface. The positioning of the insert in the tool body is dependent on the positioning of the contacting surfaces and of the magnitude and direction of the cutting force. Erroneous or varying positioning of
the insert can result in reduced productivity. This research project aims at creating a framework to handle tolerance allocation of cutting tool interface designs. The main issues found within current tolerance analysis methodologies are their inability to incorporate overdetermined surface-to-surface contacts and nonlinear material behaviours.

The backbone of the framework follows a typical empirical research model: set the design space, simulate, build a meta-model, optimize and visualize. The first iteration of the framework relies on current methodologies to gain a holistic view of the research field and to identify areas that need improvements. In the second iteration, a reliability-based optimization routine with a genetic algorithm is used to accommodate the stochastic nature of overdetermined assemblies. The framework in its current state allows the practitioner to set predefined contact zones to define the
positioning of the insert in the cutting tool body. The optimization routine finds a nominal set of input parameters that fulfils a predetermined criterion limiting the variation. The proposed framework allows for the practitioner to apply and analyse tolerances in cutting tool interface designs. The conducted research contributes to filling the scientific gaps regarding the positioning of surface-to-surface contacts in assemblies.


An approach to incorporating nonlinear material behaviour in variation simulation of sheet metal parts has been proposed using Taylor’s expansion of the primary variable in a finite element analysis. The approach has shown potential in reducing computational time with limited effect on the accuracy of the simulation. The method has not yet been implemented in the framework and needs further work before being considered for implementation.


Current limitations of the framework involve computationally heavy simulations, which grow exponentially with added input parameters. Further research needs to investigate how computational time can be reduced to increase the applicability of the framework in early design phases.