This paper focuses on the determination of the heat transfer coefficient (HTC) between blanks and dies in the hot stamping of high-strength steel sheets. Experiments to record the temperature evolution inside the dies were carried out on custom-built test equip- ment, where process conditions typical of hot stamping could be reproduced, but avoiding any deformation of the metal sheet. The HTC was determined by using an inverse analysis procedure, based on the comparison between temperature measured inside the dies and temperatures calculated by means of a numerical model of the test. HTCs between the steel blank and dies made of different materials were identified, showing that the use of ceramic inserts can modify the cooling rate inside the blank in order to avoid a complete martensitic transformation. Moreover, the influence of the contact pressure was investigated, proving that HTC dependence on this factor is not negligible

The positioning accuracy of large boring and milling machines (with axes travel larger than 5 m) is severely affected by structural deformations. Heat induced deformations, long-period deformation of foundations, and the machining process itself, cause time-dependent structural deformations of the machine body, which are difficult to model and to predict. In order to overcome these difficulties and to enhance the positioning accuracy, a composite sensor has been designed and tested, which allows direct and continuous (up to 250 Hz) measurement of geometrical deformations on machine structural elements. The present paper i) presents the operating principles of the proposed composite sensor, which is based on an array of Fiber-optics Bragg Gratings (FBG), ii) discusses requisites and performances of the sensor as well as the algorithm used to calculate the deformed shape as a function of the sensor output, iii) illustrates the results of a finite elements virtual model aimed to demonstrate the feasibility and to evaluate the expected performance of the sensor, and iv) validates the model by showing the results obtained by a sensor prototype giving a real-time measurement of the deformed shape of a structural beam.

Intelligent cutting operations capable to adjust in-line their process parameters in order to accommodate unexpected changes of working conditions are more and more desirable in order to increase the current automation level. In order to do that, a general model of the cutting process under investigation needs to be set up in advance, and then, during cutting, continuously compared with the output of process sensor signals. Among the different phenomena that contribute to the general cutting model, the evaluation of the tool wear is a matter of primary importance. The in-line prediction of tool wear critical levels would indeed prevent the production of components of unacceptable quality. The paper presents the development of a tool condition monitoring system and its preliminary results applied to interrupted cutting operations. In the paper first part, the process sensors implemented in the system for in-line acquisitions are described and their distinctive but complementary function in the system evidenced. Later on, the FE-based numerical model of the reference process is presented together with the approach followed for its calibration. It is shown that the numerical model is well suited to reproduce process mechanics, giving chance to understand how wear advancement affects process parameters and component quality. This knowledge can be exploited in building up a general model for wear evolution. Finally, data acquired by process sensors are further elaborated into process control charts that can help in monitoring the tool state as its wear evolves during cutting.