4. Discussion
4.1. Effect of material properties on rollover
The reduction in rollover depth for the light alloys is not surprising as the ductility and formability of both these materials is noticeably lower than the drawing steels (Table 1). It is expected that the rollover of both drawing steels will be higher as more material will be drawn into the deformation region. There do not appear to be, however, any relationships developed in the literature between the rollover and tensile properties of the work-piece. Intuitively, ductility and work hardening should play an important part in the different behavior of each of the metals. Oudin and Ravalard have shown than an increase in work hardening will increase the amount of rollover on the part by comparing high and low work hardening materials in bar cropping experiments [11]. Atkins has also suggested that rollover has a positive dependence on the work hardening using a simple shear model [4].
4.2. Mechanism of part burr formation—steel
Trimming experiments at 5% and 28% clearance were interrupted at a range of partial punch penetrations to examine the development of the fracture surface and mechanism of part burr formation. The two clearances were chosen as they represent the two distinct regions of burr height versus clearance behavior observed for the two steel grades. It was assumed that there was no difference between the fracture surface of the partially penetrated sample and the instantaneous fracture surface for that specific penetration of a sample currently in the process of being sheared.
At 5% clearance, complete part and scrap separation for both steel grades occurred before the punch had fully penetrated the thickness of the sheet. The primary crack formed in the sample adjacent to the punch tip (Fig. 8) and propagated directly through the sheet to the die tip (Fig. 9a). The crack initiation penetration depths for HA3P and CA5 were 42% and 55% of the sheet thickness, respectively, while the penetration depths at complete part and scrap separation were approximately 55% for HA3P and 80% for CA5.