For simulations, suitable material data is often the biggest unknown. While finite element models today are created with increasing resolution and detail, the associated material data is often surprisingly sparse. In many cases, only a few values from a technical datasheet are available, e.g. tensile modulus, tensile strength, and elongation at break. At first glance, this seems hardly sufficient to derive a complete material card. MatScape can be used to fill in these and other gaps.
A key goal of the software is to generate robust material models even from incomplete, noisy, messy or sparse data. Models that can be used in both linear and nonlinear simulations. The starting point in MatScape is the definition of a so-called “grade,” which represents the material behavior of a specific commercial plastic raw material. This includes the polymer family, any fillers (e.g., glass fibers or mineral additives), the supplier, and aging conditions. This information can either be taken from the extensive integrated database or entered manually by the user if needed.
It becomes especially interesting when a complete stress-strain curve is not available. MatScape offers several ways to generate a usable curve nonetheless. The simplest approach is based on so-called Single Point Data (see Figure 1). Here, it is sufficient if, for example, the modulus of elasticity, tensile strength, and elongation at break are known from a datasheet. Using these values and analytical assumptions, the software constructs a plausible stress-strain curve that corresponds to the typical material behavior of similar materials. This can be visualized in the curve editor and compared with existing reference curves.
MatScape becomes even more powerful when at least some curve data is available—such as from tensile tests or material cards from a simulation program. These data can be imported as so-called engineering or true stress-strain values, either via CSV file or directly from the clipboard. Upon importing, the software checks the plausibility of the curve: Is the first point defined at the origin? Do stress and strain values increase monotonically? Is there a clear turning point or even any physically unrealistic behavior in the plastic region (e.g. negative plastic strain)? The software reliably detects such issues and provides feedback on invalid data points, which can then be selectively deleted or corrected (see Figure 2).
For noisy or messy data, MatScape additionally offers the option to generate smoothed curves using analytical functions. Particularly noteworthy is the 3-parameter function, which has proven sufficiently flexible in many practical cases to reproduce realistic stress-strain behavior (see Figure 3). The parameters (initial modulus P1 and two curvature factors P2, P3) can be adjusted interactively or estimated automatically. This results in a smooth curve with controlled behavior, even from few or poor datasets. For long-term data, mathematical approaches based on the Maxwell model to describe viscoelastic behavior are available. These are especially suitable for approximating isochronous stress-strain curves.
3-Parameter-Approach [1]:
(1)
Maxwell Model [2],[3]:
(2)
If no curve is available for a specific temperature range, MatScape can also generate interpolated curves. These interpolated curves can be visually inspected and further edited if necessary. This allows for targeted completion of gaps in the temperature range without the need for additional measurements.
Another advantage of the curve management in MatScape is the clear separation between raw data and processed curves. Every import is stored in the background and can be restored at any time. This makes it easy to perform an initial rough edit while still being able to revert to the original data later, for example, when conditions change or for comparison purposes.
If the curves are consistently validated and sufficiently available over the relevant temperature range, material cards can be generated from them—either isotropic or anisotropic, depending on the material and application. Isotropic modeling is largely automated and is based directly on the provided curves (see Figure 4). The user can decide which temperatures should be included in the material card, whether additional parameters, such as coefficients of thermal expansion, are needed, and whether the curves should be scaled in stress and/or strain for the material card.
For anisotropic materials, such as short-fiber-reinforced plastics, MatScape offers a multi-scale modeling approach that systematically considers micromechanical parameters like matrix modulus, fiber geometry, and orientation. Here, too, the software guides the user step-by-step, even with limited data—from elastic to plastic calibration, up to the export function for common FE solver formats (see Figure 5).
If no own material data is available, e.g., in early project phases or when searching for a substitute material, the “Compare Curves” function can be used to quickly find a solid starting point. The integrated database already provides a wide range of stress-strain curves for various materials that can be compared with each other (see Figure 6).
With the comparison function, materials can be filtered purposefully and their mechanical behavior visually compared. Filters such as polymer type, filler content, temperature, or flow direction are available. The selected curves are displayed in a shared diagram, allowing direct comparison of typical stiffness and strength behaviors. This visual compariosn enables a quick understanding of the behavior of similar materials—even without own test data.
As a result, a suitable reference material can be identified, which can serve as a starting point for a material card or even be used directly for simulation. Values can be adopted, adjusted, or used as a template for creating custom curves. This makes the function a valuable tool when data is missing but decisions need to be made.
Thus, the strength of MatScape lies not only in managing complete datasets but especially in its intelligent handling of incomplete information. The system offers a seamless framework to create simulation-ready material models even with limited data.
[1] Ernst Schmachtenberg. “ The Mechanical Properties of Nonlinear Viscoelastic Materials ”. German. Dissertation. Aachen: RWTH Aachen, 1985.
[2] F. Brinson and L. Catherine Brinson. Polymer Engineering Science and Viscoelasticity - An Introduction. Boston, MA, USA: Springer-Verlag US, 2008. ISBN: 978-0-387-73860-4. URL: https://doi.org/10.1007/978-0-387-73861-1.
[3] Hal F. Brinson and DasGupta. “The strain-rate behavior of ductile polymers”. In: Experimental Mechanics 15. December (1975), pp. 458–463. DOI: https://doi.org/10.1007/BF02318360.
Author: Sascha Pazour, Sales- & CAE-Engineer, PART Engineering GmbH, Bergisch Gladbach, Germany
