Sheet Metal Simulation Software

[Cortney Ruic]

Inmetal forming simulation, the forming of sheet metal is simulated on the computer with the help of special software. Simulation makes it possible to detect errors and problems, such as wrinkles or splits in parts, on the computer at an early stage in forming. In this way, it is not necessary to produce real tools to run practical tests. Forming simulation has become established in the automotive industry since it is used to develop and optimize every sheet metal part.

To illustrate the metal forming process, there must be a model of the real process. This is calculated in the software using the finite element method based on implicit or explicit incremental techniques. The parameters of the model must describe the real process as accurately as possible so that the results of the simulation are realistic.

The typical parameters for forming simulation are, for example, part and tool geometry, material properties, press forces and friction. The simulation calculates stresses and strains during the forming process. In addition, simulations allow for the recognition of errors and problems (e.g. wrinkles or splits) as well as results (e.g. strength and material thinning). Even springback, the elastic behavior of material after forming, can be predicted in advance. Forming simulation also provides valuable information about the influence of process variations on stamping robustness.

Forming simulations are used throughout the entire process chain of sheet metal forming. The simulation allows a part designer to estimate the formability of a sheet metal part already during the design phase, which results in the design of a part which is easy to produce. A process engineer can already assess the process during the planning phase and optimize various alternatives using the simulation, which can subsequently reduce the fine tuning of a forming tool. Finally, regarding the fine tuning of a forming tool, simulation can provide useful information on how an existing, not yet fully functioning tool must be adjusted. It is also possible to see how the process parameters must be adjusted in order to guarantee optimal drawing results.

Metal forming simulation enables the fast review of several alternative concepts for quality and cost improvements, which results in huge cost and time savings. Furthermore, simulating the forming process improves development and planning reliability. The number of tool tryouts is reduced and tryout time is shortened. Metal forming simulation leads to the highest quality in part and tool design as well as maximum reliability in production.

Inspire Form is a complete stamping simulation environment that can effectively be used by product designers and process engineers to optimize designs, simulate robust manufacturing and reduce material costs.

With the fast and easy feasibility module, users can analyze parts in seconds to predict formability early in the product development cycle. The automated blank nesting proposes an efficient layout of the flattened blank on the sheet coil to maximize material utilization.

The tryout module includes a highly scalable incremental solver, helping users to iterate and simulate multi-stage forming, trimming and springback in a modern and intuitive user interface, reducing complexity and making the production of high quality parts more economical.

Inspire Form enables users to quickly and reliably check the formability of apart early in the product design cycle. With Inspire Form, users can visualize potential defects such as splits or wrinkles, and then modify to eliminate defects and improve overall design.

Inspire Form has a simple and highly intuitive workflow that is easy to learn and apply. Standard training sessions last only 4-6 hours, although most users can learn Inspire Form applications with no formal training at all.

I have a simplified assembly of a sheet metal cabinet door measuring approx 800mm-w x 1350mm-h x 70mm-d. The door is made from 1.5mm galvinised sheet metal, with two folds @ 90 degrees on each of the 4 edges to effectively create the door shape. Each of the corners is subsequently welded. We have several of these in the field, and they are twisting.

Having run a static stress simulation, I can generate a similar displacement profile, however the amount of the displacement is no where near actuality. e.g. simulated displacement of 0.8mm ... reality is more like 5 to 10mm.

However, I further read that using parabolic solid mesh elements means we only require 2 nodes through the thickness of the part to demonstrate accurate results (wish I could find the article to cite ...). True ?

And finally - in attempting to eradicate my twisting issue, even if I ignore the absolute values, can I assume that an x% improvement in the simulation will translate into an x% improvement in reality ? ie. 50% simulated improvement would result in 50% improvement in reality ?

There are only 2 parts in the sim; the door and a horizontal rail. (I acknowledge the screen cast showed a vertical rail ... this is an alternative setup from my baseline testing that I didn't realise I was showing).

I do have some welds modelled (created as LOFTS between ARC WELD style sheet metal bend reliefs). These are left in the simulation model - and set with BONDED contact. The welds are modelled as steel.

The top hinge hole was modelled as tangential only constraint. In essence, that hole rides on a simple rod, and in the real world is not constrained. I think I might agree that a tangential constraint is debatable; but key in my thought process was not to constrain the top hole axially along the rod so the door was free to sag and consequently slide up and/or down the rod (which would resemble reality I hope ... that was the intent anyway). Granted I did not have an axial constraint.

As I continued to simulate, the model started exhibiting some singularities in the corner WELD bodies, so they were removed with adjacent faces set to CONTACT=BONDED (they were only representative for our sheet metal fabricator).

The one thing missing from your description is the load. I assume that you have "equal and opposite" loads to create the twisting. The problem with this approach (in load case 3) is that the door is free to pivot around the hinges. (The constraints are correct to model the action of the hinges by the way.) But the "equal forces" will never be equal in the analysis to the precision of 16 decimal points, so there is a net moment acting on the door applied over an infinite length of time, and the door can rotate any angle that it wants to. The answer depends on the round off in the calculations.

What you want is one force and one constraint in "equal and opposite directions". The constraint provides the reaction force that counters the applied force, but more importantly, the constraint prevents the door from pivoting around the hinges an arbitrary angle. In other words, the constraint makes the model statically stable so that there is a unique solution.

The tray assembly has been replaced with 10 individual shelves, each attaching to the two vertical rails (the 3rd rail carries very little load in reality, but is left in the simulation as it will add rigidity to the door overall). The horizontal rail from the prior setup has been removed.

I assume that the fixed Z constraint on the bottom surface of the door is only on a small area of the door, not the entire bottom face. Assuming the hinge does not hold the door in the vertical direction, the constraint on a small area simulates where the door rests on the frame. I agree with that approach.

Today the metal forming industry is making increasing use of simulation to evaluate the performing of dies, processes and blanks prior to building try-out tooling. Finite element analysis (FEA) is the most common method of simulating sheet metal forming operations to determine whether a proposed design will produce parts free of defects such as fracture or wrinkling.[1]

The most painful and most frequent defects are wrinkles, thinning, springback and splits or cracks. Few methods are being used around the industry to cope with the main defects, based on the experience of the technicians. However, the correct process is the most vital, since it involves the correct geometry followed by number of steps to reach at final geometry. Which demands for specific experience or higher number of iterations.[2]

Deformation of the blank is typically limited by buckling, wrinkling, tearing, and other negative characteristics which makes it impossible to meet quality requirements or makes it necessary to run at a slower than desirable rate.

Wrinkling in a draw are series of ridges form radially in the drawn wall due to compressive buckling. Practically these are duo to low blank holder pressure due to which material slips and wrinkles formed. The optimum blank holding pressure is the key, however in certain cases it doesn't work. Then draw beads are the solutions, the location and shape of draw bead is the challenge, which can be analysed with FEA during design stage prior to tool manufacturing.[2]

Crack in the vertical wall due to high tensile stresses, some small radius block the material flow and results in excessive thinning at that point usually more than 40% of the sheet thk. result in cracks. In some cases it may happen due to excessive blank holder pressure, which restrict the metal flow. Somewhere it might be due to wrong process design, like try to make a more deep draws in a single stage, which otherwise feasible only in two stages.[2]

Thinning is a Excessive Stretching in the vertical wall due to high tensile stresses cause thickness reduction specifically on the small radius in the metal parts, however up to 20% thinning is allowed due to process limitations.[2]

Springback is a particularly critical aspect of sheet metal forming. Even relatively small amounts of springback in structures that are formed to a significant depth may cause the blank to distort to the point that tolerances cannot be held. New materials such as high strength steel, aluminum and magnesium are particularly prone to springback.[3]

3a8082e126