Abstract
Due to the continuously growing demands from high-added-value sectors such as aerospace, e-mobility or biomedical bound-abrasive technologies are the key to achieving extreme requirements. During grinding, energy is rapidly dissipated as heat, generating thermal fields on the ground part which are characterized by high temperatures and very steep gradients. The consequences on the ground part are broadly known as grinding burn. Therefore, the measurement of workpiece temperature during grinding has become a critical issue. Many techniques have been used for temperature measurement in grinding, amongst which, the so-called grindable thermocouples exhibit great potential and have been successfully used in creep-feed grinding operations, in which table speed is low, and therefore, temperature gradients are not very steep. However, in conventional grinding operations with faster table speeds, as most industrial operations are, the delay in the response of the thermocouple results in large errors in the maximum measured value. In this paper, the need for accurate calibration of the response of grindable thermocouples is studied as a prior step for signal integration to correct thermal inertia. The results show that, if the raw signal is directly used from the thermocouples, the deviation in the maximum temperature with respect to the theoretical model is over 200 K. After integration using the calibration constants obtained for the ground junction, the error can be reduced to 93 K even for feed speeds as high as 40 m/min and below 20 K for lower feed speeds. The main conclusion is that, following the proposed procedure, maximum grinding temperatures can be effectively measured using grindable thermocouples even at high values of table speed.