The SPE Library contains thousands of papers, presentations, journal briefs and recorded webinars from the best minds in the Plastics Industry. Spanning almost two decades, this collection of published research and development work in polymer science and plastics technology is a wealth of knowledge and information for anyone involved in plastics.
The SPE Library is just one of the great benefits of being an SPE member! Are you taking advantage of all of your SPE Benefits?
Plastic manufacturing can be unpredictable. Deviations in material batches, moisture content, machine calibration, among other variables, lead to issues in manufacturing quality and final part properties. This webinar will introduce how dielectric analysis (DEA) sensors be used to directly measure material behavior in-mold. New technology has been developed to combine dielectric analysis with machine learning and material models, allowing for dynamic adjustments to machine settings, removing uncertainty from your process, and optimizing cycle times. The material covered will include:
To address growing supply chain pressures, manufacturers are turning to Additive Manufacturing (AM) to create quality, cost-efficient products faster. Plastic thermoforming companies like Duo Form have discovered how to leverage large-format extrusion 3D printing using low-cost plastic pellets to gain a competitive edge. They are producing medium-to-large-sized thermoforming molds in less than half the time, and at a fraction of the cost compared to traditional mold-making methods.
Join engineering and business experts from 3D Systems and Duo Form as we dive deep into the integration that has made pellet-extrusion AM so beneficial for Duo Form, and how you can reap the same benefits in your own thermoforming processes.
In this webinar, you will learn about:
Co-injection molding has been introduced into industrial application for several decades. However, due to the formation of the interface between skin and core materials is very difficult to be observed, and controlled, a good quality of co-injection product can not be obtained effectively. The reason is that the formation of that interface in co-injection molding is very sensitive to various factors. In this study, the formation of the interfacial morphology and its physical mechanism in coinjection molding have been studied based on the ASTM D638 TYPE V system by using both numerical simulation and experimental observation. Results showed that the critical skin/core material ratio to generate the skin breakthrough is identified. The reason to cause the breakthrough is due to the flow front of core material catches up with the melt front of skin, and the skin is stop at a fixed distance. This mechanism is similar with that of literature. However, when the higher core material ratio is selected, the mechanism of the interfacial morphology is different. Specifically, after core melt front catches the skin melt front, the broken skin material can move forward with the inner core material to generate special core-skin-core structure. It could be due to different forces balance inside the skin and core melts, but needs to do more study in the future.
Precise predictive models are required for the use of machine learning methods for quality control in injection molding. Thermal images offer the advantage of containing information in the data that is not available in machine and process data. Currently, convolutional neural networks (CNN) have numerous applications in image recognition. Therefore, the objective of this work was to investigate the application of convolutional neural networks to thermal images of injection molded parts. For this purpose, 751 injection molding cycles from a central composite design were used. The goal was to predict the weight, height, and width of the injection molded part. The results were also compared with classical machine learning methods. Depending on the quality parameters, the networks were able to achieve an R² of up to 0.91 and were thus among the three best methods.
Nonlinear warpage analysis which considers different kinds of nonlinearity effects has attracted more and more attention recently, especially in the automotive industry. This study is mainly aimed at using the new functions in Moldex3D, “Nonlinear warp analysis” and “Buckling analysis”, to predict the warpage of the products. These new solvers cooperate with the temperature distribution and the residual stress caused by the phase change from the manufacturing process and predict the deformation of the product considering the geometric characteristic and process conditions.
Injection molding is one of the most popular techniques for global plastic production. With this automation technique, the plastic products can be manufactured at low cost with a complex geometrical shape. A manufacturing process with high productivity of an injection molding machine depends on optimized injection molding parameters. Injection molding pressure and temperature inside the mold cavity are the most critical parameters. However, cavity pressure transfer is not used due to cost and maintenance issues. During this research, an experimental procedure is performed to determine a process monitoring system using asynchronous data acquisition, through the incorporation of a wired piezo-ceramic sensor to acquire pressure of the injection molding system. This piezoelectric sensor is designed in such a way that, a Bluetooth device can be connected with a sensor and can take live data reading of parameters from the running molding machine.
Standard magnetic levitation (MagLev) device consists of two identical square permanent magnets with like poles facing each other. Limited by the size of the permanent magnet, standard MagLev device cannot levitate samples with large size. This paper proposed a novel MagLev device using magnet arrays, which can accommodate large-scale samples. Unlike magnet arrays in previous studies, all magnets employed herein face the same direction. The magnetic field generated by the magnet arrays is similar to that of the standard magnet. Within the magnetic field induced by the magnet arrays, the polymer can be levitated to an equilibrium position in a paramagnetic solution and the levitation height is related to its density. The equation correlating density and levitation height can be obtained according to the simulation results. Solutions of different concentrations were used to measure densities of a variety of polymers with an accuracy of ±0.0003 g/cm3. The non-destructive testing could also be used for plastic parts based on its posture (orientation) within the paramagnetic solution. The use of magnet arrays circumvents the trouble of manufacturing large magnets, realizes testing of polymers/parts with large sizes, and facilitates industrialization of magnetic levitation detection.
Effect of the thermal barrier coating (TBC), deposited on the mold for plastic injection molding was investigated. The mold cavities were coated by yttrium stabilized (YSZ) and phosphorous doped (PDZ) zirconium dioxide as multilayer film using chemical vapor deposition (CVD) method. It was found that films deposited at higher temperatures have better thermo-insulating properties than films deposited at lower temperatures. Growth rate and film porosity increase as deposition temperature increases. It was observed that the TBC slightly affects the flow length of the plastic melt but improves the filling ability of poorly vented molded part areas.
Microporous ultra-high molecular weight polyethylene (UHMWPE) parts were produced by microcellular injection molding (MIM) technology, which enabled higher production efficiency and lower part cost compared to the traditional powder sintering method. The microstructure could be tuned by adjusting the shot size to produce either sandwiched solid-skin – porous-core – solid-skin parts or open porous parts. The pore morphology, average pore size, pore size distribution, and pore density were characterized, and the water contact angle (WCA) and degree of oil-water separation were determined. The part weight reduction of open-porous UHMWPE and sandwiched UHMWPE parts were 16.5 wt% and 11.8 wt%, respectively. The WCA results showed that the porous surface transformed molded UHMWPE samples from being hydrophilic (34.5°) to hydrophobic (124.6°). Furthermore, the open-porous structure exhibited good oil water separation capacity. Tensile tests were carried out to study the effect of morphology on the mechanical performances of the molded UHMWPE parts. The characterization shows that a possible application for the sandwiched UHMWPE parts could be as a bone replacement material because of its high mechanical performance, and an application for the open-porous UHMWPE is as a functional filter material due to the fine pore size and high pore density.
The importance of utilizing recycled materials to manufacture plastic products has been a topic of great interest due to the environmental repercussions. Processing issues arise from the usage of these resins due to the variation in their molecular weight and rheology. In this work, pressure-controlled injection molding is evaluated and compared against conventional velocity-controlled injection molding. The effects of injection velocity, mold temperature, and pressure on part shrinkage and mechanical properties of injection molded parts fabricated with post-consumer film-grade polyethylene were evaluated. The experimental results show that the different processing techniques significantly affect the mechanical properties and part shrinkage for both materials. Additionally, different levels of injection pressure and velocity significantly affect the shrinkage of the plastic parts. Moreover, it was seen that parts fabricated using pressure-controlled injection molding had preferable overall quality.
This conference paper presents the investigations, results and findings from the research project "Tool-integrated assistance system for production control of highly complex and demanding component specifications" (acronym in German WASABI). The project investigates the possible use of sensor technology in combination with machine learning methods for the prediction of quality-determining component features on large-format plastic products. Furthermore, the information obtained will be used to propose target-oriented recommendations for action based on the predicted feature characteristics. An outer skin component (bumper) from the automotive sector was defined as the reference product for the investigations into the prediction possibilities of demanding component specifications. The injection molding tool required for production was designed as part of the project work and equipped with a variety of different sensor types (including pressure, melt contact, displacement measurement). The recording of the measurement signals is realized by a self-developed hardware system concept. The aim of the research is to predict various quality-determining characteristics from the fields of geometry (including total length) and surface (including sink marks). In the course of the project, extensive tests were carried out to generate a meaningful database. Through analysis and evaluation, it was possible to define the positions and number of sensors that provide a high level of information. Ultimately, three different approaches of machine learning methods could be learned for the prediction of component qualities and the prediction of corrective actions. These structures could be verified in laboratory environment by appropriate test data sets.
In flexible packaging, film thickness transitions can be problematic regions to seal due to their propensity for leaking, as well as the high seal pressure required to create a continuous seal over the transition. A compliant anvil can be used to decrease the required seal pressure, as the hot tool will be able to contact both the thick and thin regions of the packaging, with compression of the compliant anvil. However, a compliant anvil cannot be used in a double-sided heating process. Therefore, in a double-sided heating process, high seal pressures must be utilized in order to reduce the film thickness in the thick region, to facilitate tool to film contact in the thin region. In this study, the required seal pressure needed to create continuous (non-leaking) seals over a 4-film to 8-film thickness transition was explored, with both a rigid and conformable anvil. With a rigid anvil 3.25 MPa was required to consistently create continuous seals. With a conformable anvil 0.87 MPa was required to consistently create continuous seals.
Vibration welding flash occurs when molten polymer flows under pressure from the weld interface. This study examines the formation of small hair-like fibrils during vibration welding. Polypropylene and nylon 6 plates were butt-welded and the assemblies were assessed using a high-speed camera and digital microscopy. A mechanism has been proposed whereby initial asperities at the weld interface first melt to form a polymer pool. Thermal expansion of this pool allows polymer to be extruded laterally towards the edge of the weld interface. The extrudate is rolled up to form fibrils that can eventually grow to several millimeters in length.
Electromagnetic interference (EMI) is a common problem encountered by electronic devices, especially in electric vehicles. External electromagnetic (EM) waves affect the operation of an electronic device by interfering with the internal EM signals. To provide EMI shielding, various materials were studied, and the measured electromagnetic shielding effectiveness (SE) data are presented in this study. The main factors affecting EMI SE are quantified statistically – filler loading, shield thickness, and base polymer resin matrix. Long steel fiber thermoplastics provide the highest EMI SE, at over 60 dB at frequencies ranging from 30 MHz to 20 GHz, and at thickness as low as 1.6 mm. It is also demonstrated that carbon fiber filled thermoplastics can provide EMI shielding at levels greater than 50 dB.
The effects of the processing parameters on the curing of continuous carbon fiber composite made from Hexcel AS4/8552 prepreg tape are studied. A commercial process simulation finite element method, that takes in account the residual stresses due to chemical, thermal, and mechanical shrinkages, is utilized. This method solves the curing process sequentially. In the first step, the distribution of temperature and degree of cure in the composite is computed. In the second step, the information from the previous step is used to calculate the stress evolution during cure. At the end of the second step, the composite deformation due to tool removal is also calculated. The impact of three different process parameters on the final degree of cure and the residual stresses are studied in detail.
Ultrasonic joining is a novel friction-based joining technique to produce through-the-thickness reinforced hybrid joints between surface-structured metals and unreinforced or fiber-reinforced thermoplastics. The reinforcements’ presence is responsible for improving the out-of-plane strength of the parts, enhancing their damage tolerance. The process feasibility has been successfully demonstrated to join additively manufactured (AM) metal and polymer parts. However, further investigation of its main advantages and the joining process of subcomponents to support the technique’s further development is still missing. This paper aims to demonstrate the application of U-Joining to fabricate AM 316L and PEEK hybrid structures produced via laser powder bed fusion and fused filament fabrication, respectively. Firstly, the quasi-static single lap shear performance of coupon specimens produced with optimized joining parameters was assessed. The results indicate an improvement of 2.7 times in the ultimate lap shear force and 5.9 times in the displacement – when compared to non-reinforced flat samples. Fracture surface analyses of tested samples exhibited a mixture of cohesive and adhesive failure. Further microstructural analyses at the metal-polymer interface showed micromechanical interlocking between the parts. As observed, the PEEK was able to flow and penetrate the cavities at the metallic specimen’s rough surface due to the joining friction heat input. Finally, a selected skin-stringerbracket case study was analyzed, showing the potential of AM and U-Joining to drastically reduce the structure’s weight by about 64%. To validate this idea, a scaled-down skin-stringer-bracket technology demonstrator was successfully fabricated.
The use of in-mold melt-front detecting switches were used to control the velocity-to-pressure (v/p) transfer during injection and/or to monitor the injection in a 2-cavity, hot runner valve-gated mold. The switches were connected to a data acquisition/control system either independently, in series or in parallel. When the switches were not used for v/p transfer, screw position was used. It was found that using the in-mold switches for monitoring was more effective than either peak injection pressure or cushion monitoring to sort suspect parts and alert of changes in cavity balance. When the switches were either hooked up in parallel or independently, using the first switch closed for v/p transfer, overpacking of the mold was prevented when the heater in the drop/gate of one cavity was turned off.
A machine learning approach based on artificial neural network is presented and applied to injection molding process. Fill time, maximum fill pressure and transient cavity pressure profiles are predicted with the input process conditions of injection speed, melt temperature and mold temperature. The physics based model using Autodesk Moldflow is evaluated by comparing it with experimental fill pressure profiles for various process conditions, and it is used to generate enough data to train and validate the machine learning model. With the present machine learning model using 400 data samples, not only the fill time but also the transient pressure profiles are accurately predicted with less than 4.7% error. Further, a new machine learning model is trained with 200 data samples, instead of 400 samples, to check the dependence of the model accuracy on the sample size, and the error in prediction of transient pressure profiles increases only to 6.7%.
Injection molding is the process of injection molten plastic into a mold to form desired shape of part and it’s widely used process for mass production of plastics over the world. This process is not complete without the mold as it is the most critical part of the process. The cost of producing mold is huge due to manufacturing process and technique, tool material and cost of labor. The more effective the mold, the more efficient the process and the more profitable to the business. A critical factor is the cooling time, and a well-designed mold can achieve even cooling in the shortest period, which leads to increased productivity and higher quality of molded parts. In this research, an alternative core design was employed, to achieve these goals during the molding process. The core has 2 parts: the core and core insert. The core insert was produced using SLA technology to achieve the conformal cooling while the core was machined, and the deflection was studied using finite element analysis.
Any article that is cited in another manuscript or other work is required to use the correct reference style. Below is an example of the reference style for SPE articles:
Brown, H. L. and Jones, D. H. 2016, May.
"Insert title of paper here in quotes,"
ANTEC 2016 - Indianapolis, Indiana, USA May 23-25, 2016. [On-line].
Society of Plastics Engineers
Note: if there are more than three authors you may use the first author's name and et al. EG Brown, H. L. et al.
If you need help with citations, visit www.citationmachine.net