Manufacturing Technology

By Peter Standring, technical secretary, Industrial Metalforming Technologies

In today’s energy expensive world, the need to only do what you need to do in accomplishing a task is termed efficiency. Digital information concerning the properties of a given metal, its strength, its initial shape, surface condition, and the machine/tooling used to work it, should be sufficient to analytically determine the energy required to change its shape. 

The developments in metal forming software make it possible to determine the total energy/force needed to accomplish the task. This can be the number of blows required in a ‘discrete’ series of operations (hammering/swaging) or if continuously applied (press/rolling). The iterative nature of the computational process will produce a conclusion provided the results of the calculations converge. The fidelity of the result will be open to conjecture. As good as this may be, it is only a ‘simulation’ of what is thought to take place in reality. However, where can you obtain the reality if the job has never been done before?

As shown in Figure One, the process to develop a forging can be considered as a linear spectrum. At one end is the totally empirical method where trial and error is used to obtain the desired goal. The opposite end is where theoretical methods of analysis are employed to glean an ‘acceptable’ predicted result. When it comes to ‘trial and error’, experience can help but it can prove expensive. Physical modelling is a low cost option and there is not interference with production, whilst with numerical modelling there is a need for management commitment and ongoing support.

Since tooling and production equipment is expensive and time-consuming to use on a ‘suck it and see’ basis, a better iterative approach is to use physical modelling to obtain an appreciation of the situation. Model materials such as plasticine, waxes, as well as low strength metals, have been used for years to gain a quick easy to obtain result. Refinements and developments of these materials have shown that different colours of plasticine have different ‘yield’ strengths. Different mixtures of waxes can be used to represent variations in strain hardening. Also, the use of such materials can realistically model forging in the hot, warm and cold conditions determined by a metal’s recrystallisation temperature.

The reason for employing this detailed level of modelling has been to obtain data that can then be used to analyse and accurately predict what the ‘real world’ forging might experience in production. Of course, to obtain such data requires instrumentation of workpiece, tooling and equipment. This low cost approach will allow process variations to be explored and analysed before committing to the real thing. Moreover, the physical modelling technique will provide data on which a sound understanding can be gained of what happens during the forging process. This can also be achieved by stopping the process at any stage during the deformation to evaluate where shape changes in the workpiece have occurred. The approach can be particularly useful in identifying lack of die fill – thus helping to refine the initial billet/workpiece shape and in checking for laps, folds or splitting produced during deformation.

These faults may in part be due to excessive or restricted material flow within the tooling and mean that suitable modifications can be made to the workpiece design to overcome them. A major problem for any forging analysis is the determination of friction at the tool/workpiece interface (this continuously changes). Friction can be assumed to vary between Stiction (zero movement) and, Friction Free (zero opposition to flow). Also, determining what happens inside a deforming workpiece has always proved a major obstacle to obtaining process control.

In physical modelling, many methods have been used to evaluate what happens inside a workpiece that is being deformed. These include sectioning a workpiece vertically, horizontally, circumferentially and/or radially prior to deformation. This naturally means the workpiece is no longer homogenous and therefore will behave differently at the split interface(s). 

Post forging examination of an actual forging using flow lines, grain size distribution, etc, can be obtained from sectioning workpieces, but these techniques only provide a snapshot of one instant in the forging history. At the analytical end of the metal forming spectrum, significant strides have been made over many years. Today’s software can provide much greater detail than can be obtained from physical modelling. It can/does deliver reliable, first approximation results if the input data provided is accurate. ‘Rubbish in – rubbish out’ as the well known phrase states but, could the user without practical experience recognise this?