Car bodies are made from a few hundreds metal parts. In the production process, these three-dimensional parts are cut from two-dimensional sheet metal before they are press-formed and (spot- or laser) welded together by robots. The two-dimensional sheet metal is available as long rolls, better known as ‘coils’, which can have different widths, numerous material properties and variable prices and nowadays can even have different thickness along the roll. Taking the full part geometry as a starting point, the application can quickly develop the flattened blank outline and determine the optimal nesting layout in the coil corresponding to the lowest material cost. The following describes in few sentences the general workflow and available functionalities to effectively perform the job-to-be-done.

Part Geometry Definition

Part Geometry DefinitionThe part can be imported as a surface or solid model. Dedicated functionality for extracting the top and bottom surface of the solid model, as well as generation of the mid-surface are available in order to calculate correctly the flattening of the part geometry at the neutral fiber. The selected part can be assigned with required material and thickness information.

Blank Outline Definition

Blank Outline DefinitionThe bidding solution is based on ESI GROUP’s best-in-class one-step solver. In the shortest possible time it determines accurately the flattened blank outline (even if undercut areas exist), based on the full 3-D part model and few process parameters like binder force, material definition and thickness. In addition to the developed blank outline, part feasibility can also be directly and thoroughly analyzed through various contours like thinning, thickness and forming limit diagram (FLD). Direct import of a blank outline is supported as well.

Nesting

NestingThe developed blank outline can be used in the nesting process. The outline can be smoothened and can be applied with or without addendum region generation through constant offset of the outline. Multiple nesting layouts are supported like “One up”, “Two up”, “Mirror” and “Transfer Die” for optimal material utilization calculation.

Automatic Reporting

Automatic ReportingThe solution includes a fast and automated report generator for quotation of the part material. This functionality includes export of material utilization, fall off, pitch, coil width, blank area and pictures of the nesting sequence and developed blank outline.

Nowadays die face design is moving towards CAD environment. Iterations are done directly inside CAD and on native CAD data. The tool used for die face design must support the user in all phases of the process design cycle: from the very early conception phase, through feasibility studies until the final validation. An optimum with respect to performance and accuracy needs to be established for each of these separate steps. Die face design should be based on B-Spline geometry in order to keep the solution fast, powerful, flexible and compliant to the automotive industry standards. Any possible hold due to required changes in geometry must be removed. Surfaces must be accurate to guarantee an accurate description of contact surfaces and accurate simulation results – as early as possible

Customer Challenge

Automotive sheet metal forming parts come from the design department and often consider only aesthetics (outer panels) or functionality (reinforcements, beams etc.). Normally no, or very little, consideration is given to manufacturability. This is the job of the process design department or toolmaker: find ways to create the part in a robust and cost-optimized way. In most cases the time-pressure is huge (and ever increasing), but still many variants of the die face design and process need to be considered before an optimum (manufacturability and cost!) is found.

Typical Workflow

Starting from the part geometry, a first die face design for the first draw die is quickly drafted and evaluated for general feasibility like occurrences of cracks and wrinkles. As the first design will not normally fulfil all criteria, iteration loops are run to optimize the die face design and stamping process parameters. These iterations normally consist out of the following:

• Complete or partial part modifications coming from the design department
• Geometrical addendum modifications to eliminate wrinkles, cracks or to optimize the trimming conditions
• Process modifications to overcome cracks and wrinkles and improve the general robustness of the process
• Reduction of blank size for material cost optimization
• Propose part modifications in case a feasible or robust process cannot be guaranteed

After proof of concept of the general feasibility for the first drawing stage, the following operations are included in the process design (trimming, hemming, flanging, restrike…), both geometrical design and simulation validation. Next to the crack and wrinkle analysis and press force estimation, other criteria also become relevant at this stage, e.g. analysis including cosmetic defects for outer panels and springback compensation for the forming dies. Finally, the last step, after full validation of the process design, milling of the production tools will be carried out.

Key Capabilities

Integration of a dedicated solution for die face design into CAD environments will offer huge advantages over standard CAD usage and mesh-based die face design solutions:

• Compared to standard CAD usage, it minimizes the tool designers’ workload by implementing tool design & process knowledge, and following the natural process of tool face design.
• The integration offers a set of powerful interactive tools and functions, which provide guidance and support for part preparation, binder development and die addendum and provides quick access to important process information like trimming angle conditions and developed trimline geometry.
• It combines the convenience and speed of rapid die face design with the quality of the native CAD surfacing. Therefore highest quality simulation results can be expected – right from the beginning.
• Full CAD based design eliminates the need to regenerate a mesh model in CAD and thus doing the same work twice. The same model can be used throughout all phases of the development process, from the early (feasibility) stage up to the milling of the model
• Due to the CAD integration all native CAD functionalities can be used to reach an optimal design without need for compromises due to limitations of the mesh-based die face design software
• Easy and fast iterations due to dedicated part replace functionality: in just a few minutes, the original part geometry can be exchanged with the latest version of the part. There is no need to reconstruct manually the full die any more.
• By providing a strong dedicated link to the simulation environment, quick and easy simulation iterations can be carried out without the need for much user interaction and without loss of geometrical accuracy

Major Benefits

• Reduction of costs, by using the state-of-the-art die face development methods to deliver right the first time
• Ensuring success in prototyping and manufacturing by testing the virtual prototype first: production problems avoidance strategy
• Gain time: no need to rebuild die face designs in CAD environment based on a mesh reference
• Also in the last phase of the development process new part variants can still be easily and quickly investigated by integrated intuitive replace functionality for CAD data
• Fast learning curve for new users: with only very little training also non CAD experts can become extremely efficient in creating production ready die face designs
• The die face design product can be easily integrated into the existing host PLM structure

Support for the many iterations, which is characteristic of the early design phase of a car project, is covered through a dedicated link between both the die face design and the simulation environment. The final goal of virtual prototyping is to get the part out of the press ‘right first time’.

However, this requires beforehand a fast number of iterations within the simulation environment to come to a robust and feasible solution for the production of the component. The first simulation will usually be far from the final feasible design. In addition, the part geometry itself will frequently change during in this phase: small features can be added or removed, or parts can be completely re-designed. Historically the CAD model of the tool was imported without additional supporting information. A blank outline and drawbeads had to be defined along with their data and properties, resulting in time-consuming generation of the final tooling. This resulted in cumbersome and time-consuming iteration with a lot of manual work for the engineer, and with a huge risk for project delays and budget overspending. On top of that, also errors appeared between subsequent iterations. All of ESI-Group’s solutions for die face design are currently embedded in a CAD environment (CATIA V5 and VISUAL). They include a streamlined and efficient transition between die face design and the simulation based analysis tool, by reducing the amount of required user-interaction to an absolute minimum.

The quick link with simulation saves up to 80% of the time needed for setting up a draw die simulation. Not only are geometrical data (like punch and binder geometry) transferred, but also process data like binder forces, binder travel distances, part material definition and thickness and material offset direction and finally car and tipping system of coordinates are kept in order to enable advanced inter-disciplinary engineering.

Accurate contact permanently prohibits the nodes of the blank sheet from any penetration of the volume of the element of the tool during a calculation. The nodes are kept exactly at the surface of the element owing to the contact forces being precisely calculated.

Today, all explicit simulations in PAM-STAMP are carried out with accurate contact and high quality numerical settings, no matter if in the feasibility or validation stage. Accurate contact is applied to huge simulation models with one million or more elements. At the same time, the explicit method with many short time steps allows for the precise integration of the material history in the drawing phase. It is not possible to run accurate contact with many short time steps on large models with implicit solutions because it would be too time consuming.

Working with accurate contact, short time steps to integrate the material law accurately and from the beginning accurate numerical settings has several advantages:

• There is no risk of finding – late and thus costly – issues like cracks or wrinkles in the validation phase, just because of different (contact or numerical) settings in the feasibility stage. Accurate contact in all simulation stages and providing an accurate representation of the tool topology and tool mesh from the beginning helps to avoid this problem, which otherwise can take a lot of time to compensate for the encountered issues.
• The accurate computation of plasticity and residual stresses after forming is guaranteed, which is the pre-condition of accurate springback.
  • This accurate springback is the best basis for predictive distortion compensation on the forming side of life or in the final assembly process (hot and cold joining)
  • Accurate springback means accurate residual stresses and plastic history – which are the required basis for predictive performance simulations
• The ironing effect can be taken into account
• The press force is predicted precisely
• Accurate contact and ironing are the basis for accurate die spotting – which can save a lot of time especially in the development of a press hardening die

Accurate contact is relatively computational intensive. However, with the new triple speed mode in PAM-STAMP (referred to later on in this document), simulations with accurate numerical settings are now also possible even in the earliest stage of the project. Consequently, any numerical compromises in the feasibility stage can now be discarded.

Geometrical drawbeads are well managed with accurate contact. This further contributes to the high quality of computed results. The new triple speed mode provides short response time without trade-off in result quality. Simplifications are not necessary any more.

Residual stresses inside the sheet metal, after stamping just before tool’s removal, cause ‘Springback’. Consequently, accurate springback prediction requires accurate stress prediction during stamping. However, conventional material models like isotropic hardening cannot predict the stress accurately.

Yoshida-Uemori (Y-U) Kinematic Hardening Model

To describe the material behavior under cyclic deformation accurately, the Bauschinger effect must be taken into account in the material model. For that purpose, a kinematic hardening (KH) model is employed instead of an isotropic hardening model. The Yoshida-Uemori model is the best one for sheet metal forming among existing KH models. This is because the Y-U model involves only seven parameters of cyclic plasticity, and each parameter has a physical definition. There are no artificial mathematical parameters. In addition, a Young’s modulus depending on plastic strain is introduced in the model to describe stress-strain response more accurately after stress reversal.

Material Parameter Identification by ‘MatPara’

If the material test and parameters identification are difficult, it is difficult to use a material model in production, even though the material model itself can give results that are more accurate. Thanks to ‘MatPara’ developed by Prof. Yoshida and distributed by ESI Group, material parameters for the Y-U model can be easily identified. The following figure shows ‘MatPara’, where Y-U parameters are calculated from the material tests that include cyclic tension-compression and tensile test until rupture. Thanks to a material database inside ‘MatPara’ and a strong optimization technology behind the software, Y-U parameters can be estimated even when only tensile test results are known.

I am text block. Click edit button to change this text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

Ironing is used to control springback – in particular with high strength steel – and a natural phenomenon in processes like coining.

Precision forming of processes where ironing or coining occurs requires the use of a particular Finite Element (FE) formulation, which takes into account through thickness stress.

The normal shell formulation is not sufficient to describe this phenomena, as the stress and strains in the element normal are not described good enough. Therefore, PAM-STAMP features a particular element to simulate ironing: TTS – Through Thickness Stress element.

This element takes into account thinning, normal stress and 3D plasticity induced by bilateral contacts. It is activated automatically when ironing phenomena appears. It is compatible with Yoshida kinematic hardening model, means it can be used for advanced springback simulations.

A new solver option in PAM-STAMP allows analysis speed up in the same manner as conventional numerical tuning, but without a loss in quality. Specifically, a 3-4X analysis speed up in the drawing phase, with no associated loss of quality, allows these high quality feasibility studies to be completed within the same time constraints as conventional feasibility studies.

In combination with multi-core processors, very short simulation times can be achieved. With the new solver option and a four core CPU, a total speed up of 12-15 X is possible compared to a one core simulation without the new solver option. Simulation times are now short enough for formability stage precision to be used in the early feasibility phase – all with precision forming quality provided by PAM-STAMP. Consequently, it is no longer necessary to tune explicitly simulations by numerical means i.e. mass scaling, numerical settings, mesh coarsening etc. – to achieve feasibility run time and consequently pay with a loss of quality. Moderate baseline settings can be used in the feasibility stage. Because the new solver development is not at the expense of quality, it is also applicable in the validation / formability phase. Consequently, it is possible to achieve, with standard baseline settings for conservative engineering, the same simulation time as with tuned baseline settings, or even less. With the new solver, it is also no longer necessary to use coarser meshes or any other kind of model simplifications to achieve less simulation time. This allows for high-quality results with a reasonable simulation time at the earliest stage possible. Without loss of quality, calculations with triple speed mode are faster by a factor of 3.7 compared to the normal mode.

Especially in a hot forming process, wrinkles can come and go. In case not flattened out at the end they could even destroy a die. PAM-STAMP simulates wrinkles with no compromises – as they would occur in reality, including folding. Only an explicit simulation scheme can do that – thanks to the many small time steps and the ability to manage significant geometrical non-linearity’s without slowing down or even stop the simulation.

Optimisation of stamping process parameters, such as forces, and drawbeads, and subsequent checking of the robustness of the process to natural variations in material properties and process variables is slowly being adopted, as computer hardware and software technology evolve in synergy to reduce calculation times, optimisation and robustness assessments become a logical extension of forming simulation.

An automatic optimization module runs iterations until a good result is found. Optimized blank shape and trim line helps saving material and reducing trimming operations.

Modern requirements in automotive and aerospace industries mean higher strength with reduced weight. This means increased use of high strength steel and aluminium parts in modern vehicle structures. The problem of springback has come to the forefront of the die engineering process. The degree of springback experienced with the latest generation materials is so high, and the materials so strong, that it is not possible to ‘correct’ the springback in the prototyping, it becomes mandatory to compensate for springback as part of the draw die design.

PAM-STAMP 2G has in the recent years seen several improvements to help the engineer master the springback:

• Automatic die compensation module
• Yoshida-Uemori material model well suited for simulation of springback in HSS
• Accurate contact formulations
• Buckling detection
• Bottoming detection
• Through thickness stress element for simulation of bottoming and ironing effects

As well as for the evaluation of cosmetic defects, accurate modeling and a good process for the draw die is the precondition for successful compensation.

A part with 20mm springback – as an example – should not be compensated. Compensation can be applied to the drawing operation, or across multiple operations, with various strategies.

In the early days of simulation, it was about simulating the forming stage with the objective to eliminate cracks and wrinkles. Then simulation technique moved on to springback and the compensation thereof. With PAM-STAMP it is possible to do a full virtual prototyping of the whole stamping chain, including:

• Forming and springback
• Restrike and springback
• Flanging and springback
• Hemming and springback

This allows the engineer to have full control over the whole stamping process, ensuring highest quality and no “surprises” during physical prototyping.

The apparition of cosmetic defects is deeply linked to springback phenomenon. The detection of such small defects is thus only possible with a very precise simulation of springback, which has been proved to be possible with PAM-STAMP. It features new dedicated defect contours based on stoning or sensor for detection of the defects and their quantification. With such contour the user can easily localize the defects as it is done in workshop and measure their depth and area.

As a consequence of all the new developments, trade-offs between simulation time and result quality can now be discarded, and realistic virtual prototyping is now possible.

The tools for die face design from ESI produce high quality die faces based on B-Spline geometry, which represent an accurate description of the contact surfaces. An automatic transfer of data to PAM-STAMP and set up of the drawing simulation minimizes work time. Iterations on part geometry, die face, binder, draw beads and any other process related parameters are completed in no time. Today, the PAM-STAMP solver always works with accurate contact in the drawing operation, no trade-offs in numerical settings and, if needed, with geometrical draw beads and any advanced material model for the yield surface and hardening. The new triple speed mode in conjunction with four or eight core parallel processing delivers simulation results in a breath-taking short response time, even on low cost computers. Simulation times are amazingly short on computers with eight cores, a solid-state drive, and a more recent processor. The speed up with an eight-core configuration and triple speed mode can be up to 20 against a single core configuration with no triple speed option. This allows constant high quality results from early feasibility to high-end formability, and consequently the minimization of the overall engineering cost.

Hot forming is growing rapidly, and is a fascinating manufacturing technique, where the good formability of the warm blank is combined with exceptional strength of the end part due to the quenching in the tools. No traditional available material that is formable is even close to the strength of the hot formed steel. This makes it a natural choice for the crash relevant parts in the car. Today all major OEMs work with hot formed parts in the cars as crash re-enforcements. It allows building even small sized cars with an outstanding crash performance – enabling the traditionally weaker group A cars also to get the 5 stars in the EURO-NCAP crash tests (e.g. Fiat 500).

This means that looking only on the formability of the part during stamping doesn’t really make sense. The whole chain has to be kept in mind – this from early design stage on. To get the properties of the end part right is crucial to achieve the crash performance. This means that the crash engineers have to rely on the stamping department to manufacture the parts with the right properties.

Hot forming itself is a manufacturing technique where different fields will play together to make it work or not. The stamping department has to build knowledge now also in metallurgy, heat transfer, cooling and fluid dynamics – areas where normally several specialists are involved.

So to summarize, with this new process, the stamp engineer is suddenly confronted with several new areas where he has to have a high level of knowledge to get the process right.

Even for the most talented engineers, it will be to ask for too much to become an expert in all these fields. This is typical area where the virtual manufacturing can play an important role in getting the new processes running. The part manufacturing with all its different aspects can be tested virtually before the expensive hot forming process is started. Also the part performance in the final crash can be tested virtually. This is again a step towards the end to end virtual manufacturing – even if the challenge to simulate all the different aspects still remains.

The complete value chain is nowadays available, allowing analyzing the complete press hardening process from initial part cost to distortion after quenching, cooling channel analysis and virtual reality check.

In the previous section, the fabrication of components including the physics of materials has been tackled. In an automotive manufacturing process, the components are then assembled in the body shop. Again, shape, material properties, and stiffness are modified in this process. Product and process architects would benefit from this information, if available at the right time. To provide this information at the right time is the goal of the Virtual Body Shop. In technical terms, distortion due to the assembly process and related assembly problems, the true point in the stress-strain space, stiffness properties, residual stresses, and remaining hardening capabilities are provided to product and process architects – at the right time.

The virtual assembly process is described hereafter and illustrated with a few milestones.

• Structural steel sheet metal parts are formed by stamping simulation,
• Parts are positioned on fixtures,
• Contact interference is detected between stamped geometries,
• Clamping tools are closed,
• Contact and distortion is updated,
• Effects of pin locators etc. are taken into account,
• Spot welds join components,
• Effects of gap closing and thermal shrinkage is taken into account,
• Contact and distortion is updated,
• Assembly is unclamped and final distortion / residual stresses / true point in the stress-strain space are computed and made available for any virtual investigation.

The benefits for product and process architects lie on hand:

• Uncover positioning problems,
• Study effects of pin locators, fixtures, clamps, welds, gaps between parts,
• Find assembly problems and apply counter measures,
• Save cost for prototyping and apply costly engineering only where necessary,
• Improve virtual performance evaluations in the field of durability and crash behavior.

In concurrent engineering, a workflow still exists, but iterations are taken into account. The main difference from the conventional approach of product design is that all disciplines are now involved in the earliest stages of product design; they progress concurrently, so that the iterations result in less wasted effort and lost time. A key to this approach is a well-recognized importance of communication among and within disciplines. A powerful and effective tool used in the process of planning the manufacture and performance of the product is computer simulation. The next step in the production process is to make and test a prototype, that is, an original working model of the product.

Now, that next step is done in the process of virtually planning the manufacture of the product, simulating the physics of the material for each manufacturing step, and making this complex technology available for engineers and designers with process automation. Virtual reality from ESI GROUP is available to push the process of virtual prototyping to the next level.