Abstract

The analysis carried out in this research work, showed the inconsistencies observed in trying to identify the tube wear in heat exchangers. It is seen that the associated error could be an order of magnitude higher than the expected value, indicating the possibilities of failure at less than the predicted design life. Hence, the use of the commonly accepted concept of "work rate" as a scaling parameter is found to be questionable. Wear models available in the open literature, have always been heavy on empirical formulations, because of the complexity of the process. Therefore, it is essential to verify any proposed model experimentally, using controlled parameters. The experimental study conducted for this research work was implemented for the verification of the new theoretical wear model developed. These experiments were carried out on two independent rigs to assert the validity of the new model. The basis for this model was the extension of the delamination theory of wear for impact conditions (normal and oblique). It was essential to conduct a parametric study using a future mechanics approach, to quantify some variables for this model. The finite element method was implemented to study the crack nucleation zones and loading cycles required for the formation of a crack. It was also implemented to study the probable direction a growing crack will follow, and to estimate the crack propagation rate. Even though, all the quantities in the new wear model were determined theoretically, it was observed that the prediction of the wear volume using the trend line was close to the prediction of the wear using the experimental one. This discrepancy was adjusted using randomly selected three data points to properly calibrate the crack tip sliding displacement (CTSD). Calibrating the model using three data points or all the data points did not alter the results significantly, for most of the experiments. In order to reduce the scatter observed while verifying the wear model, it is essential to introduce a characterization parameter. The new wear model is extended to incorporate this characterization, by implementing a force and displacement category multiplier to the work rate. The verification of the characterization process was conducted using the fretting experiments which were carried out on Inconel 600 tubes and stainless steel supports. The wear rate was properly characterized using the new macro/micro contact stress dependent model, developed through a fundamental understanding of the process.