Search results

This paper aims to explore the fluid flow in the stereolithography process during the recoating step. The understanding of the flow dynamics can be used as an input for an active control of the resin surface height map. The recoating over a rectangular cavity has been considered to investigate the influence of the cavity depth on the resin surface height map.

Two-dimensional numerical simulations have been used to obtain the flow characteristics as function of the cavity depth. An experimental setup, which mimics the recoating process in the stereolithography process, was used to verify the results of simulations and to test the suitability of the 2D model. The surface height profile along the centreline was measured by a confocal chromatic distance sensor and compared to the 2D numerical results.

By means of computational fluid dynamics (CFD) simulation, the flow in the cavity and the free-surface behaviour of the resin was explained for different cavity depths and confirmed by experiments.

The study is focused only on the cavity depth variation to show feasibility and suitability of the presented CFD model and the proposed analytical expression to estimate the layer thickness.

The proposed approach can serve as a tool for designing the closed-loop control for the recoating system in the next generation of stereolithography equipment.

In the present work, the fluid flow behaviour, a source of significant imperfection in the recoating process, has been investigated during the recoating step over a rectangular cavity.

Most stereolithography systems use a blade to accomplish the recoating of the part being built with a new layer of resin. States the problems associated with this technique and describes experiments conducted to determine how recoating parameters should be controlled. Differentiates between recoating over an entirely solid substrate and over one consisting of solid and liquid, i.e. the “trapped volume” condition. Discusses parameter control for both of these conditions. Concludes that recoating is an important part of the stereolithography process which must be optimized to ensure accuracy of prototype parts.

This study aims to examine the feasibility of feedforward actuation of the recoater blade position to alleviate the resin surface non-uniformity while moving over deep-to-shallow transitions of submerged (already cured) geometric features.

A two-dimensional computational fluid dynamics (CFD) model has been used to determine optimized blade actuation protocols to minimize the resin surface non-uniformity. An experimental setup has been designed to validate the feasibility of the proposed protocol in practice.

A developed protocol for the blade height actuation is applied to a rectangular stair-like configuration of the underlying part geometry. The evaluation of the actuation protocol revealed the importance of two physical length scales, the capillary length and the size of the flow recirculation cell below in the liquid resin layer below the blade. They determine, together with the length scales defining the topography (horizontal extent and depth), the optimal blade trajectory. This protocol has also shown its efficiency for application to more complicated shapes (and, potentially, for any arbitrary geometry).

This study shows that incorporation of a feedforward control scheme in the recoating system might significantly reduce (by up to 80%) the surface unevenness. Moreover, this improvement of performances does not require major modifications of the existing architecture.

The results presented in this work demonstrate the benefits of the integration of the feedforward control to minimize the leading edge bulges over underlying part geometries in stereolithography.

To satisfy the demands for rapid prototyped small‐size objects with intricate microstructures, a high‐resolution stereolithography (SL) system is developed.

This novel SL system consists of a single mode He‐Cd laser, an improved optical scanning system, a novel recoating system and a control system. The improved optical system consists of a beam expander, an acoustic‐optic modulator, a galvanometric scanner and an F‐θ lens; the recoating system consists of roller pump, resins vat with an integrated high‐resolution translation stage and part building platform and a scraper. Experimental studies were performed to investigate the influences of building parameters on the cured line width and depth.

With the SL system, a laser light spot with a diameter of 12.89 μm on the focal plane and resin layers with a thickness of 20 μm have been obtained. The experimental results indicate that cured depth and width increase with the ratio of laser power to scanning speed, and cured line with a width of 12 μm and a depth of 28 μm was built, which showed the capability building microstructures with this new SL system.

The building area limited to 65 × 65 mm, is smaller than that of current SL system.

Small objects with intricate microstructures can be fabricated with the SL system.

The high‐resolution SL system provides a solution to the problem that has hampered the progress of SL process into a high resolution ranges below 75 μm.

Conventional machining methods for fabricating piezoelectric components such as ultrasound transducer arrays are time-consuming and limited to relatively simple geometries. The purpose of this paper is to develop an additive manufacturing process based on the projection-based stereolithography process for the fabrication of functional piezoelectric devices including ultrasound transducers.

To overcome the challenges in fabricating viscous and low-photosensitive piezocomposite slurry, the authors developed a projection-based stereolithography process by integrating slurry tape-casting and a sliding motion design. Both green-part fabrication and post-processing processes were studied. A prototype system based on the new manufacturing process was developed for the fabrication of green-parts with complex shapes and small features. The challenges in the sintering process to achieve desired functionality were also discussed.

The presented additive manufacturing process can achieve relatively dense piezoelectric components (approximately 95 per cent). The related property testing results, including X-ray diffraction, scanning electron microscope, dielectric and ferroelectric properties as well as pulse-echo testing, show that the fabricated piezo-components have good potentials to be used in ultrasound transducers and other sensors/actuators.

A novel bottom-up projection system integrated with tape casting is presented to address the challenges in the piezo-composite fabrication, including small curing depth and viscous ceramic slurry recoating. Compared with other additive manufacturing processes, this method can achieve a thin recoating layer (as small as 10 μm) of piezo-composite slurry and can fabricate green parts using slurries with significantly higher solid loadings. After post processing, the fabricated piezoelectric components become dense and functional.

To operate a multiple material stereolithography (MMSL) system, a material build schedule is required. The purpose of this paper is to describe a scheduling and process‐planning software system developed for MMSL and designed to minimize the number of material changeovers by using low‐viscosity resins that do not require sweeping.

This paper employs the concept of using low‐viscosity resins that do not require sweeping to minimize the number of material changeovers required in MMSL fabrication. A scheduling and process‐planning software system specific to MMSL is introduced that implements four simple rules. Two rules are used to select the material to be built in the current layer, and two rules are used to determine at which layer a material changeover is required. The schedule for the material to be built depends on the material properties stored in a user‐defined materials library. The developed algorithm produces sliced loop data for each material using the predetermined layer thickness from an input CAD model, and the four rules are applied at each layer. The algorithm then determines the build order for each material, the material‐specific number of layers to be built, and whether or not sweeping is required. Output data from the program are the scheduling and process‐planning report and the partitioned computer‐aided design models to be built before changing a material according to the process planning. Two examples of the algorithm applied to multiple material parts are provided.

The MMSL scheduling and process‐planning software system is developed using Microsoft Visual C++7.0. For verification, a simple demonstration is conducted on a two material part where the process plan could be easily determined through intuition. A more complex multiple material part is also tested that consisted of four subparts. Several cases of resin assignment are tested showing that the ultimate scheduling and process planning vary significantly depending on the material combinations and specifications. These examples demonstrate that the strategy, method, and software developed in this paper can be successfully applied to prepare for MMSL fabrication.

Although the software system is demonstrated on two multiple material parts, more extensive work will be performed in the future on fabricating multiple material parts using the MMSL machine. It is expected that additional rules will be developed as additional limitations of MMSL are identified. It is also anticipated that particular emphasis will be placed on building without sweeping as well as development of advanced non‐contact recoating processes.

As designs incorporating multiple materials increase in the future and additive manufacturing (AM) technologies advance in both building out of multiple materials and fabricating production parts, the scheduling and process‐planning concepts presented here can be applied to virtually any AM technology.

Optoforming is meant to be a potential substitution for the established Stereolithography (SLA) process. Its potential is that different ceramic‐filled photo‐curable epoxy‐resins can theoretically be used to manufacture highly loadable parts and tools. The stiffness as well as the thermal and chemical resistance of the material used (an epoxy resin named Tooling B) are higher than those of established SLA‐materials, such as SOMOS 7120. Automotive applications, in fields where the parts are directly used parts, such as lighting housings for prototype purposes, as well as tools for the veneering process in small batch production, were successfully tested. In order to enable precise and cost‐effective fabrication, optoforming has to be developed further in the field of secondary processes, such as inline‐filtration of the material and its feeding, as well as the machine software. Currently, another competitor offers a more mature process based on upgraded SLA machines, which use a ceramic‐filled epoxy‐resin also.

Stereolithography apparatus (SLA) is capable of in situ fabrication of complex parts, as well as mechanisms and complex devices with embedded components. In this paper, a series of example devices are presented to illustrate the power of building around embedded components (inserts). The problem formulation, solution approach, and specific rules and procedures are presented using these examples and experimental results. A case study approach is used for presentation. These procedures and results lend insight into promising new applications of SLA technology, as well as novel methods of implementing additional functionality into SLA and other rapid prototyping technologies.

This paper aims to present parametric models to estimate the environmental footprint of the selective laser sintering (SLS)’ production phase, covering energy and resource consumption as well as process emissions. Additive manufacturing processes such as (SLS) are often considered to be more sustainable then conventional manufacturing methods. However, quantitative analyses of the environmental impact of these processes are still limited and mainly focus on energy consumption.

The required Life Cycle Inventory data are collected using the CO2PE! – Methodology, including time, power, consumables and emission studies. Multiple linear regression analyses have been applied to investigate the interrelationships between product design features on the one hand and production time (energy and resource consumption) on the other hand.

The proposed parametric process models provide accurate estimations of the environmental footprint of SLS processes based on two design features, build height and volume, and help to identify and quantify measures for significant impact reduction of both involved products and the supporting machine tools.

The gained environmental insight can be used as input for ecodesign activities, as well as environmental comparison of alternative manufacturing process plans.

This article aims to overcome the current lack of environmental impact models, covering energy and resource consumption as well as process emissions for SLS processes.

Accurate build‐time prediction for making stereolithography parts not only benefits the service industry with information necessary for correct pricing and effective job scheduling, it also provides researchers with valuable information for various build parameter studies. Instead of the conventional methods of predicting build time based on the part’s volume and surface, the present predictor uses the detailed scan and recoat information from the actual build files by incorporating the algorithms derived from a detailed study of the laser scan mechanism of the stereolithography machine. Finds that the scan velocity generated from the stereolithography machine depends primarily on the system’s laser power, beam diameter, materials properties and the user’s specification of cure depth. Proves that this velocity is independent of the direction the laser travels, and does not depend on the total number of segments of the scan path. In addition, the time required for the laser to jump from one spot to another without scan is linearly proportional to the total jump distance, and can be calculated by a proposed constant velocity. Most profoundly, the present investigation concludes that the machine uses a velocity factor which is only 68.5 per cent of the theoretical calculation. This much slower velocity results in an undesired amount of additional cure and proves to be the main cause of the Z dimensional inaccuracy. The present build‐time predictor was developed by taking into account all the factors stated above, and its accuracy was further verified by comparing the actual build‐time observed for many jobs over a six month period.