October 23, 2003
Material selection is a very important aspect of design flexibility when striving to fulfill part functionality requirements. Choosing the correct material is fundamental to making the part effectively and efficiently.
The way a material is formed and the conditions it needs to withstand for successful pressure sequence hydroforming (PSH) and high-pressure hydroforming (HPH) differ greatly from one material to another. It is important to understand how a material behaves during each process to select the best possible material for the job.
PSH re-forms a tube's cross section by bending the tube wall from a normally round starting tube into the desired shape along the length of the part. The wall is bent from a smooth, relatively gentle curve to sharper cross-sectional corners and flat portions, common in structural parts (see Figure 1). This reshaping places the outside surface of a corner in tension and the inside surface in compression. In the flat areas of the part, tension is on the inside and compression on the outside, since it is straightened. Much of this deformation is plastic.
Exactly how tension and compression forces manifest themselves in the final part depends on how previous operations, such as flat sheet production, tube making, bending, and preforming, have affected the material. All PSH prototype and production applications use material and tube made to common commercial standards with no special requirements or development to serve the needs of the hydroforming process.
Material testing shows a 5 to 10 percent increase in yield strength from a PSH-produced part. The PSH process will be described in detail in a later article in this series. In the final analysis, the most concise and relevant way to assess the net effect of residual stress patterns is to consider the part's dimensional stability.
Normally, PSH produces no wall thinning (See Figure 2). However, previous operations such as sheet and tubemaking, bending, and performing introduce thickness variation, which is still evident in the finished part.
In HPH the tube typically is designed to be 5 to 10 percent smaller than the die cavity periphery. The starting round tube must stretch into the final cross-sectional shape after the die is closed. The stretching pattern varies depending on the part shape; its position within the final form cavity; material type and properties; as well as lubrication, surface finish, and several other factors that can reduce wall thinning concentration.
Generally, the pattern appears as in Figure 3, with little or no stretching in flat regions of a cross section. This occurs because the areas destined to flatten usually are the first to contact the cavity surface. Being pressed against the cavity surface by higher forming pressure causes friction that is too great to allow sliding. Wall thinning increases substantially closer to the corners, increasing the likelihood of rupture. Avoiding rupture while expanding the cross section is the fundamental reason behind extensive efforts to develop hydroforming steels, as well as forming simulation and lubrication.
A tubular blank placed in the die will have some deformation from previous processes. In most cross sections, most of the material is stretched by HPH, but to varying degrees, as described previously. In HPH more of the material will have a net tensile residual stress pattern than in PSH. A high degree of dimensional stability is attributed to HPH, but supporting data has not been publicized.
As material is stretched it work-hardens and its yield strength [YS] increases. Experiments have shown that for mechanically stretched mild steel, the increase is roughly proportional to the percentage of strain the material experiences. Hydraulically stretched material behaves similarly. On average, around the cross section, YS gain is the same as the percentage of expansion (5 to 10 percent), but locally it can vary from near zero in flat areas to 30 to 40 percent in corner regions. Work-hardening rates differ for other alloys, but mild steel is a common reference point.
Assertions that mild steel can be work-hardened to be equivalent to high-strength, low alloy (HSLA) should be scrutinized carefully. Mild steel is noted for its relatively wide variation in YS, elongation, and n value, which is one of the reasons it is so often the economical choice. HSLA benefits compared to mild steel include tighter control of alloying elements, reduced mechanical property variability, increased YS, as well as increased minimum YS after annealing [such as results from welding].
Stretching material by expansion will increase YS, but it is unclear how the other benefits can be achieved reliably. Additionally, stretching, and therefore the YS increase, are quite variable around a typical cross section. Ranges of 1 to 2 percent in flat areas and up to 20 percent in corners are common. Because of this fluctuation, the average increase in mild steel may not be as high as expected.
The two main metals used in the automotive structural industry are steel and aluminum. Various alloys from each group serve a number of purposes.
One of the most important differences between PSH and HPH is the effect of their forming mechanisms on material. Their ability to work with the characteristics of a material and capitalize on its advantages also differ. Selecting the right material often is fundamental to successful production, best part economy, and function. Consequently, using the process best able to form the material to the part requirements is equally important. Choosing a material to suit the process might not suit the part function.
PSH's cross-section reshaping by bending has several benefits that widen the range of materials feasible for prototype and production. Lower-elongation material can be used. N values normally produced when making commercial-quality, electric resistance-welded mechanical tubing are satisfactory.
PSH makes low demands on the material but still is able to form complex shapes. In all production applications, the customer specifies the material. Mild steel commonly is chosen because it provides the most rigidity for the least cost. Special material is needed only when it is dictated by the part function, not the production process requirements. However, PSH can be used on special materials developed for other processes, because these materials typically feature enhanced formability, elongation, and n value.
HPH places higher formability demands on material, because of expansion, thus requiring materials with higher elongation and n values to form a given part. Even if special materials are called for, HPH can be a more economical forming method than other options discussed later. As formability increases, YS tends to decrease. This is a process benefit because it reduces the pressure needed to form the part, but the end result may not be consistent with the part's functional needs.
Hydroforming simulation by finite element analysis (FEA) has become very important to provide some insight into what material properties are needed for a reasonably robust process, particularly when expansion occurs.
The forming characteristics of aluminum can be challenging because they make it less susceptible to stretching. Forming aluminum with HPH requires a cautious approach because of the stretching required to avoid pinching and possible cross-section expansion.
PSH can be more effective for forming aluminum. The severity and range of features can be greater because more formability can be devoted to making them. In several instances, aluminum has been substituted for steel in production hydroforming dies with no change in tools or process conditions. Several alloys in the 5,000 and 6,000 series were used. This demonstrates that aluminum can be production hydroformed quite easily. Figure 4 shows an instrument panel beam made with aluminum [top], mild steel , and steel with 80-KSI minimum YS [bottom].
Many materials can be hydroformed as long as they possess some ductility. A production 310-MPa (45,000-PSI)-minimum-YS HSLA is shown in Figure 5. Ultra-HSLA and stainless steel prototypes have been made in production tools, demonstrating similar forming ease. Other lower-elongation materials also can be formed, but bending determines what can be done. One production die normally running mild steel was used to make a number of aluminum and ultra-HSLA steel parts with no process change. Figure 6 shows prototypes of a production part made with 80-KSI-min. YS and 140-KSI UTS steel.
Some material properties in tube hydroforming, such as yield stress or strength, elongation, n value, and, perhaps, r value, are important in varying degrees.
YS is the point at which the material starts to deform plastically and eventually form, as shown in Figures 2 and 3. YS changes have no effect on the process conditions for PSH. The required internal pressure remains the same when YS doubles.
For HPH, YS increases as the part progresses toward full formation by stretching, which is concentrated in cross-section corners. This increase requires that internal pressure be elevated accordingly to continue forming to completion.
With steel, as YS increases, elongation decreases, and the amount of deformation that the material can withstand before necking and rupturing is reduced substantially.
Elongation is the amount that the material stretches linearly prior to necking. PSH uses a small amount of elongation to reshape the part cross section, even in its most severe case. As a result, lower-elongation materials can be used, such as HSLA steel, ultra-HSLA steel, and aluminum. In addition, where expansion is required, the amount that can be achieved is greater than with HPH.
HPH requires more elongation for successful basic forming, particularly in the cross-section corners. Where the linear part of the stress-strain curve is exceeded, necking can lead to rupture. Designing to prevent this is a key consideration leading to one or more of several methods to spread out or reduce wall thinning.
It can be difficult to detect where the curve is exceeded and rupture is possible. There will be little elongation left in the affected region. Stresses may cause premature fatigue cracking farther along the part. Consequences and costs can be formidable; designers must consider where rupture may occur and prevent it.
N value is the work-hardening exponent that expresses material behavior as it is plastically deformed. Higher n values indicate that as stretching commences, a particular element stretches to the point where it gets too strong. Any continuing strain is shared with elements surrounding it that are not as strong. Spreading strain more effectively allows larger expansion, and thus this property describes a material's ability to balloon. In lower-n-value material, stretching concentrates locally and the material bursts more easily.
N-value always is important in HPH because all parts are expanded, whether to prevent pinching, to vary the periphery, or to form larger expansions. It is a concern for PSH only when hydroforming die expansion is a component of the part's design.
N value also plays a role in how quickly a part can be formed. Higher values indicate that faster forming is possible, while lower values dictate a longer cycle time to avoid rupture.
R value quantifies material drawability and is applicable to large expanded sections, normally achievable only at the part end where end feeding is most effective. Attention to maintaining a specified range of values is necessary for some applications, particularly for HPH.
Tube manufacturers normally buy steel with the required YS of the final part and an elongation property sufficient to survive the operations to be performed. The finished tube should allow for process robustness, part toughness, and reasonable fatigue life. Normally, the n or r values aren't tracked during processing, but they can be at a cost.
Tube manufacturers do not plan or depend on work hardening of the material during processing to bring YS up to a final specification value, such as a minimum YS. Ensuring a minimum strength level may be needed from a part design perspective, but it is difficult and risky to guarantee, considering the variability of each step in the progression of a hydroformed part.
The start tube wall thickness affects HPH and PSH differently, a fact that must be considered to achieve the desired final thickness.
Equation 1 shows that thickness has a direct effect on HPH process pressure, with thinner material allowing less pressure to be used. As shown in Figure 3, the material thins in the corners. Depending on the amount of expansion and available elongation, stretching may exceed the linearly plastic region of the stress–strain curve to the onset of necking. Exceeding this point will cause rupture. Even approaching it creates a strong possibility of dramatically reducing part fatigue life.
Another consideration is the minimum material thickness specification required of most components. Wall thickness consistent with the part specification will be too thin in the cross-sectional corners because of thinning. For the finished part to meet the minimum thickness requirement, the starting wall thickness must increase accordingly. This can result in a significant weight penalty, for example, 20 percent. This conflicts with assertions that buying a tube that is smaller, 5 to 10 percent, and expanding it makes a lighter part. When combined with the near-rupture situation discussed previously, it is wise to scrutinize this area carefully.
It is common to weld stamped brackets to structural parts. Variable wall thickness affects gas tungsten arc welding stability. A good-quality weld in a thicker area may burn through in a thinner area, or a good-quality weld in a thin area may give a cold weld in a thicker area. Also, any YS gains planned on from processing the part will be lost where stress normally is highest, because of the local annealing effect of welding.
Consequently, effective process simulation is important to predict problems during part development before prototypes are made. This is especially true for HPH, predominantly because of the process's wall-thinning effect. The difference between a good part and rupture can be small; variables, such as material elongation, n value, and consistency (properties and thickness) can be that difference. Other variables include the blank and tool surface finish, lubrication, and process equipment parameters. All need to be considered in designing the hydroformed part.