Search Results

Traditional warm forming of aluminum refers to sheet forming in the temperature range of 200°C to 350°C using heated, matched die sets similar to conventional stamping. While the benefits of this process can include design freedom, improved dimensional capability and potentially reduced cycle times, the process is complex and requires expensive, heated dies. The objective of this work was to develop a warm forming process that both retains the benefits of traditional warm forming while allowing for the use of lower-cost tooling. Enhanced formability characteristics of aluminum sheet have been observed when there is a prescribed temperature difference between the die and the sheet; often referred to as a non-isothermal condition. This work, which was supported by the USCAR-AMD initiative, demonstrated the benefits of the non-isothermal warm forming approach on a full-scale door inner panel. Finite element analysis was used to guide the design of the die face and blank shape.

The influence of elevated temperature forming on local mechanical properties of AZ31B magnesium (Mg) sheet material was investigated. The Mg sheet was formed into a closure component with high temperature gas pressure at 485°C. Miniature tensile testing specimens were cut from selected areas of the component where different levels of thinning occurred. The specimens were strained in tension to fracture using a miniature tensile stage. The two-dimensional strain distribution in the necking region along with true stress-true strain curves were computed using a digital image correlation technique to assess the influence of the forming-induced thinning on tensile strength and percent elongation at fracture.

This study details preliminary results of hot rolling trials of AZ31 alloy sheet using a pilot-scale rolling mill. The aim is to design and optimize the hot rolling schedule for AZ31 in order to produce sheet with a fine and homogeneous microstructure. The study examined three different hot rolling temperatures, 350, 400 and 450°C and two rolling speeds, 20 and 50 RPM. A total thickness reduction of 67% was obtained using multiple passes with reductions of either 15% or 30% per pass. The entry temperature of each rolling schedule was kept constant, by reheating the strip between passes. It was found that the microstructure of the AZ31 alloy was sensitive to the rolling temperature, the reduction (i.e. strain) per pass and the rolling speed (i.e. strain rate). A combination of a rolling temperature of 400°C, reduction per pass of 15%, and rolling speed of 50 RPM produced the finest and most homogeneous microstructure.

The deformation characteristics of a commercial AZ31 magnesium sheet alloy were investigated at elevated temperatures. Tensile experiments were conducted at temperatures 300°C, 400°C and 450°C and at strain rates, 0.001s-1, 0.01s-1 and 0.1s-1. Depending on the test temperature, fracture analysis of failed specimens revealed three different types of failure: (1) by moderate necking, (2) by interlinkage cavity, (3) by strong necking. Plastic strain ratios, r-values were derived from the strain ratios of width and thickness of the fractured tensile specimens. The r-value increased with increasing temperature and strain rate.

Experimental quick plastic forming (QPF) of commercially available magnesium alloy AZ31B sheet into Cadillac STS decklid inner panels was done successfully with existing QPF tools and processes developed for forming QPF-grade AA5083 aluminum sheet. This demonstrates that QPF parts designed for aluminum can be made with magnesium. The post-formed properties of the formed panel were investigated. Thinning of the magnesium alloy sheet in the successfully formed panel was limited to just under 50%, which is normally considered acceptable in QPF aluminum panels. The basal crystallographic texture of the sheet material was essentially maintained through the forming process. Tensile properties of samples cut from the formed panels exceed the specified minimums for the O-temper AZ31 sheet. Significant reduction in cycle time is expected based on the results of this work.