SPE Library

Computer aided engineering-based design methodologies have been utilized throughout the Automotive Composites Consortium Focal Project 4 to assess the vehicle level structural stiffness and impact performance of the composite underbody design proposals and to estimate the potential mass reduction for several candidate material scenarios. To increase confidence in the vehicle level CAE model predictions and to better understand the effect of material and manufacturing variability prototype molded underbody components were fabricated and subsequently built into underbody assemblies to assess their structural performance. Non-destructive component and assembly tests were devised to assess the general static and modal performance of the underbody component and a quasi-static destructive test of a built-up underbody assembly was developed to simulate the deformation and loading observed in the worst case vehicle impact design load case. The paper will discuss the preparation and fabrication of the built-up test assemblies the structural stiffness and modal performance testing of trimmed underbody molded components and assemblies and the destructive testing of assemblies. The predicted performance was investigated for two composite thickness assumptions to account for the additional thickness observed in the prototype components. Predictions were then compared to the measured test results to understand the status of correlation between the response of idealized components and the as-molded prototype test components. A comparison of the non-destructive stiffness and modal test results to the predictions indicated that the stiffness and modal response were reasonable. The destructive underbody test was developed to better represent the physical composite and metallic components. The destructive underbody test was limited by buckling of the longitudinal rail. The results correlated well with predictions up until rail buckling occurred after which significant local damage was

Folgar Tucker model has been in use in commercial software for predicting fiber orientation for fiber/polymer suspensions. One of the major challenges in modeling injection molding processes is the complex flow in the frontal region. However the standard method of using the model with Hele Shaw approximation limits its capability as a prediction tool especially near the advancing front region and in the outer layers of the molded part. In this work the effects of the fountain flow region were assessed by including a simplified semi-circular cap to the finite element mesh. Simulations were performed with a fixed mesh and a full 2-D velocity field was solved using Navier Stokes equation for steady state and the orientation equations were decoupled from momentum equations. We looked at combinations of inlet conditions for orientation and the model parameters to determine which are most compatible with the geometrical simplification used to describe the front. All combinations of model parameters and initial conditions considered in this work qualitatively reproduce the measured orientation profile. However large discrepancies between predicted and experimental orientation near the walls suggest the need for a robust approach to handle the effects of the advancing front on fiber orientation.

The Automotive Composites Consortium (ACC) is conducting a multi-year project to develop a better understanding of the root causes of the visual surface distortion effect known as bond-line read-through (BLRT). Initial studies using a finite-element analysis (FEA) based approach showed good agreement with experimental observations and highlighted the importance of accounting for viscoelastic adhesive material properties. A parametric FEA-based study of a small laboratory scale coupon was conducted to examine the effect of the adhesive joint cross-section geometry and adhesive type on the predicted peak curvature resulting from an elevated temperature adhesive cure. The parameters evaluated in this study were uniform and non-uniform adhesive thickness SMC substrate thickness adhesive bead width and adhesive type.

Hydrogen fuel cell-driven electric cars continue on a slow but steady progression toward commercial viability. Dismissed by many as being too expensive fuel cells are within range of the cost of other vehicle propulsion systems due to advancements in design and manufacturing that have taken place in recent years. Composites have been an integral part of the success of proton exchange membrane (PEM) fuel cells. Bipolar plates made from conductive bulk molding compound have proven to be effective durable and low cost in comparison to other materials. This presentation documents properties recent developments and successful commercialization of thermoset bulk molding compound for transportation fuel cell applications.

Automakers have developed successful computer simulation processes to meet the most stringent crash noise/vibration/ harshness (NVH) and aerodynamic and vehicle dynamics requirements making computer-aided engineering (CAE) an established component in today vehicle-design process. Engineers and management are comfortable with CAE deliverables for traditional metal-based vehicle design and now require reliable simulation technologies and methods to integrate engineered plastic such as carbon fiber laminates in their standardized and automated simulation procedures. This presentation will discuss the challenges of composite material calibration how CAE simulation can be used to aid material characterization the unique modeling and visualization requirement for composites and how optimization simplifies the design of laminate composite structures tailoring the material itself to the loading requirements and avoiding overdesign of part.

The primary aims of the project were to determine the suitability of ESI PAMFORM with regards to modelling a multiaxial fabric and to assess how a manual forming process could be simulated. Material models for multiaxial fabrics were developed through physical testing. Four different simulation methods were investigated and compared in terms of ease of set up processing time and results. An optimisation process was developed using batched input files in order to examine the optimum fabric property for a given component. This process is now used to assist in the selection of a fabric in the early stages of any new component design. A UK government-funded Knowledge Transfer Partnership (KTP) with Nottingham University was started in January 2013 with the aim to create a validated materials database for use with this simulation.

Currently the automotive industry is making a major push toward vehicle weight reduction. While traditional SMC provides several advantages over other materials for use on Class “A” body panels weight reduction is not necessarily one of them. The invention of a lower density Class “A” SMC allows the material to maintain its traditional advantages while also competing with other lightweight alternatives. Unreinforced panels (e.g. fenders roof panels etc.) molded with the material can reduce weight by up to 20%. Closure panels (e.g. hoods decklids etc.) when bonded to low-density inner panels can provide up to 30% weight savings over a traditional SMC assembly. This paper will summarize the development of the material as well as present manufacturing trial and part performance data. Initial evaluations at OEM facilities will also be discussed.

Fiber-reinforced polymer composites are finding new applications in aerospace high-performance as well as medium build-volume alternate powertrain automobiles civil infrastructure sports equipment and emerging alternate energy industries due to their high stiffness-to-weight ratio. Laminated structures are among the most common forms of structural fiber-reinforced polymer composites. Fiber orientation in each lamina and the stacking sequence of the laminated structures can be chosen to tune the desired strength and stiffness. For enhancing the predictive modeling capability of composite structural performance an accurate computation of the effective material properties of composite materials is of special interest to engineers. This paper discusses the prediction of the effective mechanical properties of glass fiber-reinforced epoxy composites (fabricated using an infusion process) utilizing both classical laminate theory as well as a finite element-based micromechanics approach and compares the results against experimental findings. The results from the physical tests exhibit good correlation with the predicted mechanical properties.

The experimental orientation of long semi-flexible glass fibers has been evaluated in complex 3-dimensional flow. Preliminary experimental values of long-fiber orientation were obtained within injection-molded end-gated plaques at multiple percentages of plaque length and width including in the areas of complex flow near the mold side walls. Additionally experimental values of orientation were obtained within the sprue and gate region of the injection molded parts. Modification of the experimental method for measuring fiber orientation in these regions due to the increased length and flexibility of long fibers is included.

It is well known that retained fiber length in random fiber composite materials relates directly to the mechanical properties. Longer fibers lead to higher aspect ratios that increase stiffness and strength as well as enhance the creep and fatigue properties. Direct compounding of thermoplastics promotes fiber length retention by the use of continuous glass fiber in the compounding process. In the same way pre-compounded long fiber pellets provide increased fiber length relative to traditional short fiber injection molding compounds but perhaps not to the extent of direct compounded methods. Despite the known fiber length retention characteristics of these various materials and processes via qualitative analysis and examination of resultant mechanical performance a rapid and robust quantitative fiber length characterization method seems to have eluded the industry to date. Time consuming counting of individual fibers randomly selected from samples seems to be the norm. Based upon these limitations and needs a method to rapidly characterize the fiber length distribution in random fiber composites was investigated. The experimental procedure is discussed and the results to date are presented.

Short-fiber-reinforced thermoplastics are a feasible alternative to develop lightweight materials for semi-structural applications. These materials present a layered structure showing a complex fiber orientation distribution along the molding. The details of fiber orientation in a center-gated disk with diameters of 1.38 and 2.05 mm were obtained in several regions including the gate and advancing front. Several modifications were introduced in the method of ellipses to obtain unambiguous orientation measured over small sampling area. Two fiber suspensions (30 % short glass-fiber PBT and PP) with different rheological characteristics were used in these experiments. The results showed an asymmetric distribution of fiber orientation that gradually washs out as the flow progress. In addition the initial orientation measured at the gate presented a fiber distribution different from the random orientation that is assumed in literature for a center-gated disk.

Two micro-mechanically based composite fatigue models are introduced in this presentation. The focus is on the high- cycle fatigue model implemented specifically for chopped- fiber-reinforced plastics. Its application for a Toyota Motor Europe automotive oil-cooler bracket made of a nylon 6/6 material reinforced by short-glass fibers will be presented. Through this case study the presentation aims to show how the use of proper fatigue-modeling tools developed specifically for composites can increase the accuracy of simulation in the field of durability and pave the way for new simulation standards towards the desired lightweight reductions.

This presentation details how polyurethane spray sandwich technology originally developed for sunshades has been improved for use in more demanding applications such as load floors and parcel shelves. Polyurethane sandwich construction combines the low weight of a honeycomb core with the high strength of a fiber-reinforced polyurethane skin to produce load-bearing parts with very-high flexural stiffness and excellent thermal properties making it an attractive lighter weight alternative to ABS polypropylene sheet-molding compound (SMC) and wood products. Information on the deflection performance of different constructions with different systems including some with natural and some with glass mats will be given to guide manufacturers on the best ways to hit specific targets such as cost thickness or weight. Newer formulations enable productivity improvements including longer open times and shorter demolding times which facilitate production of larger parts and reduced scrap as well as feature higher bio-renewable content than previous versions.

Long fiber-reinforced thermoplastic (FRT) composites in automotive industrial fabrication are of critical requirement -- more so than short FRTs. The FRT products’ mechanical properties and warpage are dominated by fiber orientation within the part. This presentation will discuss a recently proposed new fiber orientation model for improving the prior developed models with regard to interaction and diffusion of the fibers immersed in a matrix namely iARD- RPR (Improved Anisotropic Rotary Diffusion model combined with Retarding Principal Rate model). The iARD-RPR model has been demonstrated to describe changes in fiber orientations well whether treating short fibers or long fibers. In this study 40 wt% glass-fiber immersed in polypropylene matrix was injection molded in a center gated disk and then predicted fiber orientation distribution pass the thickness was compared with measured results. Good agreement with experimental observations was achieved.

A methodology utilizing a non-linear topology-optimization technique was applied to develop designs of mass- efficient composite beam structures. The traditional linear optimization technique is shown as suitable to develop designs that are maximized only for part stiffness. Non- linear effects like plasticity and material failure are not taken into consideration using linear techniques and hence the suitability of the linear-optimization technique can prove to be inadequate for applications that require energy management. Non-linear topology optimization using the software tool LS-Tasc from LSTC uses fully non- linear LS-Dyna simulations to arrive at the optimized design shape. Plasticity material damage and failure and load path variation on account of contact are taken into consideration as is typical with non-linear LS-Dyna simulations. The optimization process tracks the contribution of each element in the finite-element model of the design space to the stated objective and performance constraints to determine the ideal load path and hence the part shape. Development of the beam structure designs using this methodology results in design shapes that can be optimized for energy management rather than stiffness.

Most industrial composite parts that are large and complex in geometry are manufactured by the hand layup method. The resin transfer moulding (RTM) process is a better substitute but is not used readily due to the lack of proper manufacturing technology. Development of a proper RTM manufacturing process for a specific application requires a proper mould design. In addition the difficulty in the tooling design and mould fabrication cost increases with size and complexity of the component. The scale down strategy of full scale product avoids bigger size mould requirements prototype production for product testing and quality check at the starting phase of product development. Moreover the scale down strategy can be used to validate the process and the product with less capital input. In this work we propose a methodology to develop a scaled down prototype for a large and complex composite structures based on virtual simulation technique keeping the mold fill time and mold fill pattern unchanged. The methodology has been demonstrated taking a composite cab front that is currently used by the hand layup technique as case study. From the simulations and actual experiments it was found that the injection pressure at the full scale model has to be reduced to the times of reciprocal of square of geometrical mould scale down factor to meet the same mould fill time and mould fill pattern keeping the injection strategy the same.

A method to identify root causes of manufacturing quality defects has been developed that allows for the implementation of process and material improvements via a databased analysis system known as “The SMC Consistency Method.” In 2006 a statistical method that ties SMC molding parameters to process and raw material parameters was introduced. The following year an SMC viscosity improvement effort using this method was announced. The current paper presents additional examples that have identified root causes of material and process variations that have resulted in sporadic defects in the molded product. The case study will show how defect data from the molding plant was successfully used to identify key molding compounding and raw material factors. The SMC consistency method utilizes actual production data as opposed to the use of data generated by conducting special DOEs.

Carbon fiber-reinforced polymer (CFRP) is considered to be a good candidate for energy absorption due to its high specific energy absorption (SAE) as the ratio of energy absorbed by the tube mass. However composite damage distribution in the components should be carefully designed to confine damage progression in the load application region and prevent any premature catastrophic failure. This presentation addresses different damage and failure modes triggered in composite crush tubes that have different ply orientation angles. The modeling strategy is validated by experimental quasi-static crush tube experiments and the study contains a comprehensive damage-mode tracking in each ply to identify the effectiveness of the candidates. The correlation between the damage propagation is compared with the overall crash response in terms of crush load vs. crush displacement.

This paper presents a methodology for the selection of thermoplastic materials in order to achieve the most cost effective manufacturing solution. Unlike conventional materials selection methods—which rely almost exclusively on quantitative performance data—this method relies on a comprehensive evaluation of cost, including material costs, processing costs, and the cost of secondary operations.

Dry food packaging typically contains a combination of polyethylene resins to provide toughness and barrier with a sealant layer to provide a specific shelf life. Development of the materials and resulting structures requires fundamental knowledge of structure / property relationships as well as the ability to tailor properties for in-use performance. Dow has been developing both resins and novel testing methodology to help expedite the development process, focusing on the consumer needs and benefits.