Abstract
Ring Rolling is widely considered as a near-net manufacturing process, due to the rough and the relatively imprecise final products. In most relevant industrial productions, a Ring Rolling cycle is followed by multiple, and sometimes extensive, machining cycles to achieve the required dimensional accuracy. Given that metallic ring products are widely used in some crucial applications, producing more dimensionally accurate products at less time through Ring Rolling can significantly increase the production quantities and reduce the production cost per piece. In the current dissertation, multiple different precision increasing techniques and methods of a typical (flat), hot Ring Rolling process were thoroughly investigated. The proposed methodologies corresponded to several practices of a Ring Rolling manufacturing line and thus they covered all production stages – from billet to final product. Moreover, the time required for these additional corrective practices was significantly lower ...
Ring Rolling is widely considered as a near-net manufacturing process, due to the rough and the relatively imprecise final products. In most relevant industrial productions, a Ring Rolling cycle is followed by multiple, and sometimes extensive, machining cycles to achieve the required dimensional accuracy. Given that metallic ring products are widely used in some crucial applications, producing more dimensionally accurate products at less time through Ring Rolling can significantly increase the production quantities and reduce the production cost per piece. In the current dissertation, multiple different precision increasing techniques and methods of a typical (flat), hot Ring Rolling process were thoroughly investigated. The proposed methodologies corresponded to several practices of a Ring Rolling manufacturing line and thus they covered all production stages – from billet to final product. Moreover, the time required for these additional corrective practices was significantly lower than running lengthy post-process machining cycle, since the former would be performed using the same Ring Rolling mill, without a need for re-positioning or intermediate storage of the workpiece. The entire analysis of the current dissertation was performed numerically, since the necessary equipment for an experimental research was not available. The conducted numerical models were prepared using the commercial FEA software ANSYS/LS-DYNA, which offered the necessary tools and algorithms for these simulations. After a fully validated model was established, it was used as a basis for every subsequent analysis performed. In that way, the conducted simulations could act as a strong indicator towards the feasibility of the proposed methodologies and practices. Initially in the first chapter of the current dissertation, a thorough literature review of the relevant experimental, analytical, and numerical research performed on Ring Rolling was conducted. Additionally, the principles of Ring Rolling and the necessary equipment required were presented. Finally, based on all of the above, the research questions that would drive the current dissertation were set. In the second chapter, a step-by-step development of a typical, hot Ring Rolling simulation of an IN718 ring was performed. The developed numerical model was validated with corresponding experimental data found in literature. The choices that were made during the development of the aforementioned model were based on thorough literature and/or trial-and-error analyses, so that the physical phenomena that occur during the process would be properly simulated. In the end, the numerical model simulated the actual experiment very realistically, while several points of interest were detected and further analyzed. In the third chapter of the current dissertation, three different process attributes taking place before or during Ring Rolling and that affect the dimensional accuracy of the process were investigated. These attributes were: (a) the precise volume estimation of the initial workpiece billet, (b) the effects of the thermo-elastic tool deformations, and (c) the effects of the support tool movement algorithm and its relationship to the material of the ring. For the precise billet volume estimation, a novel semi-analytical methodology was developed, which facilitates a combination of analytical equations and numerical models of every preceding process to Ring Rolling, in order to calculate the required billet volume for a final ring with specific dimensions. The proposed methodology was validated via a series of simulations, with the divergencies of the final product being far less than 1 %. In the case of the effects of the thermo-elastic tool deformations on the dimensional accuracy of Ring Rolling, three different numerical models were developed. In each model, the deformability of the tools varied (rigid, only elastic, coupled thermo-elastic) and the final results from the three models were compared to one another. From the aforementioned comparisons, it was clarified that even small deformations of the tools could lead to greater dimensional imprecisions on the product (especially in cases of high-precision products). Lastly for the analysis of the support roll movement algorithm, an AA5754 product Ring Rolling simulation was performed and compared to the IN718 simulation from the second chapter. The two simulations were then repeated, but with higher order polynomials describing the movement of support rolls. The final results revealed a dependence of the final ring dimensions from the material of the workpiece, while the higher order polynomials affected defect generation, as well as the growth rate of the ring.In the fourth chapter, a newly proposed process for the "correction" of dimensional imprecisions right after a Ring Rolling process was introduced. This novel process was named "Reverse Ring Rolling". From its conceptualization, Reverse Ring Rolling involves a specific movement pattern of the tools, which leads to the reduction of the ring’s diameters (both internal and external), and a subsequent increase in its height. The whole process could be distinguished into two separate steps: (a) one in which both diameters are reduced, and (b)a second in which only the external diameter is reduced. The feasibility of this process was validated via numerical modeling and the calculated results revealed a high-precision final product. Additionally, several important process parameters were further analyzed, with the number of support rolls and the velocity of all tools proven to be the most affecting ones. Finally in the fifth chapter of the dissertation, a novel approach for the production of metallic polygonal products using a typical Ring Rolling mill was introduced. This practice was presented mainly as a proof of concept, since a more in depth analysis would diverge significantly from the main scope of the current dissertation. However, since the core mechanics of this process are explained to some extent, any future research on the subject will not have to start from scratch. The basic idea of this approach lies with the simultaneous movement of support rolls in the same direction, which led to the formation of straight edges on the product. Through a carefully calculated sequence of tool movements and velocities, the manufacturing of a polygonal product was simulated as a proof of concept, while the most crucial attributes of the aforementioned process were briefly discussed. Overall, the proposed process was proven to be feasible, although further research should be conducted.
show more
