AM Parts Research Articles

Estimation of the accuracy of measurement of internal defects in X-ray Computed Tomography

Additive Manufacturing technologies found many applications in the last decade in many industries. AM parts show good mechanical properties, but the presence of porosity is still limiting the adoption of these technologies on a wider scale. X-ray computed tomography (XCT) has become a successful measurement tool to characterize AM parts, especially for detecting defects. In this work, a test object was used to simulate internal porosity and test the accuracy of the XCT measurement. Repeatability and reproducibility of the XCT system were investigated as well as the influence of the internal defects position and size on the measurement error.

Surface quality improvement using electro chemical machining process for γ-TiAl parts produced by electron beam melting

This study aims to improve the surface finish of γ-TiAl AM parts produced via Electron Beam Melting (EBM) process. The sample parts were machined by Electro Chemical Machining (ECM) by changing main variables such as electrolyte type, pulse and feed rate and their effects on the surface qualities are presented and discussed. ECM workshop trials have showed that NaCl type of electrolyte is less effective to machine the material than NaNO3. Additionally, it is also observed that it is unable to remove the sludge (machined material particles) from the workpiece with continuous voltage due to high machining areas (25x25 mm2). Pulsing during the machining process and fresh electrolyte transferred without current pass clean the surface and brighter surfaces are obtained. It is also observed that the higher feed rates above 1 mm/min causes short circuit occurrence between the workpiece and the tool therefore machining time increases, and the material removal rates decreases. Before and after ECM process, EBM part surfaces were measured via an optical profilometer. It was found that the Ra values decreased from 35.39 µm to 3.25 µm in build direction and from 26.73 µm to 2.47 µm in scan direction. The surface roughness is improved by approximately % 91 for both directions. Therefore, it is proposed that ECM is an effective method to be applicable for improving the surface finish of EBM parts surfaces as post-processing.

Hybrid Manufacturing Processes: an experimental machinability investigation of DED produced parts

As metal AM technologies have realized significant development, hybrid manufacturing, which integrates both additive and subtractive processes in a single machine tool, has emerged in order to tackle the geometrical and surface defects that are present in AM parts. Due to their unique microstructure, subtractive process planning has to be re-evaluated. This work aims at investigating the machinability of AM-produced components in the scope of hybrid manufacturing. A preliminary experimental investigation of surface quality and tool wear has been made using BOHLER K890 samples manufactured by utilizing DED process. Core milling parameters have been evaluated regarding their effect on the final result.

Deep learning based porosity segmentation in X-ray CT measurements of polymer additive manufacturing parts

X-ray computed tomography (XCT) is the only non-destructive technique able to perform a complete quality control of additive manufactured products in a single inspection. Yet, high-quality scans are associated with large acquisition times and limited to high-end AM parts. In this paper we investigate a deep learning U-net segmentation algorithm to improve the segmentation of low-quality XCT scans. A high-quality XCT scan is acquired and aligned with low-quality XCT scans to create the ground truth data. The accuracy of the segmentation is quantified with the Jaccard index and physical properties of the parts.

On the effect of laser powder-bed fusion process parameters on quasi-static and fatigue behaviour of Hastelloy X: A microstructure/defect interaction study

The mechanical performance of parts, made by laser powder-bed fusion (LPBF), under quasi-static and cyclic loadings can be altered significantly by changing the process parameters. To better understand the related changes, tensile and fatigue LPBF Hastelloy X coupons were manufactured at high and low laser scanning speeds (LSS) in the nearly full-dense processing window. The experimental results demonstrate that, although the microstructure and surface roughness of the parts varies by altering the LSS, the porosity or density of the printed samples are not changing significantly within the range studied here. High LSS samples show higher tensile strength and higher rates for strain hardening at the primary stages of the deformation. Statistical analysis of variance indicates that LSS is a significant factor affecting the fatigue life. The S-N diagram shows Basquin lines intersection at the midlife fatigue region. In the low cycle fatigue (LCF) region high LSS samples show higher cycles to failures due to hysteresis stabilization at lower strains and hence less relative pre-strain and cyclic damage. Since stress risers are important features at high cycle fatigue (HCF) due to stress concentration and associated plasticity, the low LSS samples tolerate a greater number of cycles at lower stress levels and show a higher fatigue limit obtained by the step-loading method. These findings suggest that the process parameters can be optimized to achieve the best possible performance based on the desired application in LCF or HCF for AM parts.

Fatigue performance of additive manufactured metals under variable amplitude service loading conditions including multiaxial stresses and notch effects: Experiments and modelling

Additive manufacturing technology has gained significant attention in recent years. However, due to lack of sufficient understanding of their fatigue behavior under service loading conditions, design of critical load carrying parts using this technique is still at early stages. These conditions include multiaxial stress states, notches and stress concentrations, and variable amplitude load cycles. Some of the distinguishing features of AM metals as compared to the conventionally fabricated metals include defects, surface roughness, anisotropy and build orientation effects, and residual stresses due to the fast solidification during the fabrication process. These factors resulting from AM processes may significantly affect the fatigue performance of AM parts under multiaxial variable amplitude loading. In this work, unnotched and notched fatigue behavior of two commonly used AM metals with different surface roughness conditions were studied under variable amplitude multiaxial loads. Both crack initiation approach using critical plane-based model and fracture mechanics approach using crack growth from the rough surface or surface defects at the notch were used to estimate the fatigue lives. Fatigue estimations are then compared to the experimental results.

Surface finishing of additively manufactured stainless steel surgical instruments

PurposeAdditive manufacturing (AM) has the potential to revolutionise the fabrication of complex surgical instruments. However, AM parts typically have a higher surface roughness compared to machined or fine cast parts. High surface roughness has important implications for surgical instruments, particularly in terms of cleanliness and aesthetic considerations. In this study, bulk surface finishing methods are described to produce end-use selective laser melting parts.Design/methodology/approachThe aim was to achieve a surface finish as close as possible to machined parts (Ra = 0.9 µm, Wa = 0.2 µm, Pv = 7.3 µm). A sample coupon was designed to systematically evaluate different finishing techniques. Processes included bulk finishing, blasting and centrifugal finishing methods on individual parts, as well as heat treatment before and after surface finishing.FindingsAbrasive blasting or centrifugal finishing alone was not adequate to achieve an end-use surface finish. White oxide vapour blasting at high water pressure was the most effective of the abrasive blasting processes. For centrifugal finishing, a 4 h runtime resulted in an acceptable reduction in surface roughness (Ra = 2.9 µm, Wa = 2.0 µm, Pv = 34.6 µm: inclined surface [30°]) while not significantly increasing part radii. The combination of finishing methods resulting in the smoothest surfaces was white oxide blasting followed by 4 h of centrifugal finishing and a final glass bead blast (Ra = 0.6 µm, Wa = 0.9 µm, Pv = 6.9 µm: inclined surface [30°]). The order of these methods was important because white oxide blasting was significantly less effective when applied after the centrifugal finishing.Originality/valueCollectively, these results describe the development of a practical bulk finishing method for stainless steel surgical instruments produced by AM.

Mapping value clusters of additive manufacturing on design strategies to support part identification and selection

PurposeAdditive manufacturing (AM) allows companies to create additional value in the processes of new product development and order fulfillment. One of the challenges for engineers is to identify suitable parts and applications for additive manufacturing. The purpose of this paper is to investigate the relation between value creation and the design process. The implications of this relation provide an orientation on the methods for identifying parts and applications for additive manufacturing.Design/methodology/approachMapping the value clusters of AM on design strategies allows determining the expected degree of change in design. A classification into major and minor design changes is introduced to describe the predictability of the impact of AM on past performance and business model. The ability to predict the future properties of an AM part determines the suitability of identification and selection methods from literature. The mapping is validated by an identification process that creates a shortlist of potential AM parts based on the strategic decision for a value cluster. Shortlisted parts are then evaluated based on the criteria technology readiness, required post-processing, customer benefit and manufacturer benefit.FindingsThe mapping of value clusters on expected design changes determines the type of selection process. For minor design changes, automated part identification serves as a powerful tool while major design changes require the judgment of skilled engineers.Research limitations/implicationsThe mapping of value clusters to design strategies and degree of change in design is based on empirical observations and conclusions. The mapping has been validated in an industrial context in different identification and selection processes. Nevertheless the versatility of AM and industrial environments impede a universal validity of high-level concepts.Practical implicationsThis value-driven process of identification and selection was applied in technology transfer projects and proved to be useful for AM novices and experts. The mapping supports the identification and selection process, as well as the general product development process by providing an indication of the design effort for implementing AM.Originality/valueThe novel mapping links the economic domain of value creation to the engineering domain of design strategies to provide guidance in the selection of economically and technically suitable parts for additive manufacturing.

Anisotropy effect of additively manufactured Ti6Al4V titanium alloy on surface quality after milling

The microstructural and mechanical anisotropy of metallic additive manufactured parts is well-established as the nature itself of the process induces the formation of columnar grains along the build-up direction. However, the effect of this anisotropy on the alloy machinability is rarely considered and a deeper understanding is needed. In the current study, Ti6Al4V prisms were fabricated by laser powder bed fusion in both vertical and horizontal orientation. After heat treatment, the prisms were slot milled at different cutting parameters to assess the influence of the build-up orientation on the surface quality, focusing on the surface topography. The results show that the skewness parameter can distinguish the machined surface of differently oriented AM parts and that the horizontal orientation allows for better machinability when considering the burrs height and the chip morphology.

Sustainable additive manufacturing: predicting binder jettability of water‐soluble, biodegradable and recyclable polymers

AbstractAdditive manufacturing potentially generates customized polymeric objects with tailored geometry while reducing waste, employing biodegradable polymers and striving to use recyclable polymer feedstocks for a more sustainable future. Binder jetting additive manufacturing (BJ AM) creates 3D objects upon depositing liquid adhesive onto a powder bed, while a roller spreads powder over a build stage to form subsequent layers. Many factors affect jettability and printability including printhead design, binder material, binder solvent and powder characteristics. Material design for higher strength BJ AM parts using metals, ceramics or polymers exploits binders or particles that transform into thermosets or permanently shaped matrices through thermal or UV curing and sintering. While many BJ AM examples utilize biodegradable, water‐soluble and recyclable polymers for biomedical and industrial applications, a wider array of opportunities exist to degrade, dissolve and reuse printed materials. Systematically shifting to these polymers will facilitate reuse and recycling of both binders and powders, but more studies to improve part strength are required. This review describes an overview of polymer jettability and a perspective into opportunities currently underutilized in transitioning BJ AM into a more sustainable AM platform. © 2020 Society of Industrial Chemistry

Design for additive manufacturing process for a lightweight hydraulic manifold

This article examines the process of redesigning a conventionally manufactured hydraulic manifold to one designed for additive manufacturing (DfAM). In this case, the factors to be optimized included minimizing the support material, and post-processing required to make the part, as well as minimizing the products’ weight and print time. Improvement in product functionality would also be a desirable outcome of the project. The paper also compares the new and old manifold in terms of material usage, weight, costs, Manhattan distances, and residual stress.The DfAM process used in this article includes the steps of removing all material that doesn’t serve a specific engineering function, exploring the effects of different print orientations, techniques for eliminating support material and post-processing. These techniques are demonstrated in the redesign of the hydraulic manifold and present the resulting printed product. The design steps used in the case study are also described in a generic manner to make them applicable to other AM parts printed in either metal or polymer.

Selective electron beam melting of high strength Al2024 alloy; microstructural characterization and mechanical properties

Abstract To date, the most commonly used aluminum alloys for additive manufacturing are eutectic/near-eutectic alloys. However, fabricating high strength aluminum parts has not yet been fully investigated due to solidification cracking and porosity formation, which make these alloys very hard to be processed. Electron beam selective melting (EBSM) of aluminum alloys has some advantages over selective laser melting. The process is not affected by reflectivity of aluminum powders, and parts exhibit less induced thermal stresses. In addition, oxidation is reduced due to vacuum condition. Therefore, the process provides high potential for the fabrication of aluminum alloys, specifically, Al2024 alloy which is widely used in different industries such as automotive and airline industries. Hence, these industries are highly benefitted from being able to build defect-free AM parts. This study is significant because Al2024 alloy is highly susceptible to solidification cracking and porosity formation, and crack-free samples with high relative densities have not been built by EBM process previously. The baseplate preheating temperature was set to 350 °C, and samples were fabricated by applying a proper powder bed preheating configuration for each layer. Different sets of processing parameters were used to investigate their effect on the performance of the parts. Archimedes technique was used to measure the relative density of samples, ranging from 95% to full-dense. The results showed that higher relative densities could be achieved by increasing the input energy to an optimum value; however, by further increasing the energy, the relative density decreases, which can be attributed to porosity formation. SEM and EBSD results showed equiaxed grains in a crack-free microstructure. Grain structure consisted of high angle grain boundaries having different crystallographic orientations with a few sub-grains. Additionally, AlCuMnFe, AlCuMnFeSi, and eutectic Al2Cu phases were uniformly distributed in the α phase matrix. Microhardness results showed an almost uniform change of hardness values in both horizontal and vertical sections, ranging from 100 HV to 110 HV for as-built samples. Moreover, tensile and yield strengths of as-built samples reached 314 MPa and 191 MPa, respectively, which can be further increased by a proper post-processing treatment.

Analysis of the effect of surface roughness on fatigue performance of powder bed fusion additive manufactured metals

An important advantage of AM is to create net-shaped components with no need for further surface machining since it is usually very difficult for complex geometries. The effect of surface roughness on fatigue performance of powder bed fusion additively manufactured metals was evaluated in this work. The intrinsic rough surface of metal AM parts that could act like existing surface cracks and significantly affect fatigue performance is investigated. The effect of processing and post-processing conditions on surface roughness of AM parts are reviewed and the effect of surface roughness on fatigue performance is analyzed. The synergistic effect of internal defects and surface roughness on fatigue performance is also evaluated. Arithmetic mean surface roughness parameter Ra is used here as a representation of surface roughness, since this parameter has been presented in most of the studies dealing with metal AM surface roughness. Surface roughness effect on fatigue performance is shown to be dominant even in the presence of relatively large internal defects and various microstructures. Ra decreases with an increase in power. This consequently increases fatigue limit, although scatter is observed due to the secondary effect of internal defects and microstructure. For similar microstructures, considering synergistic effect of Ra along with average critical internal defect size in combination with the stress amplitude, resulted in improved correlation of the S-N fatigue data.

Reference standards for XCT measurements of additively manufactured parts

An increasing number of industrial sectors are considering the potential of additive manufacturing as an asset to improve their production. Indeed, additive manufacturing enables the fabrication of very complex geometries and inner cavities that cannot be manufactured with conventional techniques. However, in critical sectors such as aerospace, defence and medical, the parts need to be certified, which requires parts to be non-destructively characterised in terms of flaws, geometry and dimensional accuracy. X-ray computed tomography is the only current 3D volumetric technique, which is suited for the non-destructive analysis of internal flaws, geometry and measurements of internal dimensions and roughness. However, regardless of its huge potential, Xray computed tomography is not as mature a technology for dimensional metrology as compared to conventional tactile coordinate measuring machines. In most cases there is no traceability to SI units in the dimensional domain. Recently, numerous reference standards (i.e. physical artefacts) addressing X-ray computed tomography dimensional accuracy have been published, but they do not necessarily address the calibration of XCT system in connection with AM parts. In this work, a new and improved standard in three different materials has been designed with a dual purpose: Fully calibrating X-ray computed tomography for dimensional measurements while being representative of additively manufactured parts in terms of flaws and material, meeting the needs of the industry. These standards will be used to metrologically validate X-ray computed tomography for the inspection of additively manufactured parts.

Post-processing and testing-oriented design for additive manufacturing – A general framework for the development of hybrid AM parts

Abstract Additive Manufacturing is frequently depicted as a promising manufacturing technology for various applications. Due to its ability to produce highly integrated parts economically in low quantities, industry and academia are exploring the new range of opportunities. Therewith, however, requirements concerning quality assurance, qualification and life cycle related monitoring are consistently rising. This impedes the adoption in a broader industrial range and favors aversion against AM technologies. To address these issues, this contribution presents a framework for a post-processing, testing and life cycle monitoring oriented development of AM products based on comprehensive requirements analyses, inter-processual correlation and provision of expert knowledge.

Analysis of strain and stress concentrations in micro-lattice structures manufactured by SLM

Purpose Additive manufacturing (AM) enables the production of lightweight parts with complex shapes and small dimensions. Recent improvements in AM techniques have allowed a significant growth of AM for industrial applications. In particular, AM is suitable for the production of materials shaped in lattice, which are very attractive for their lightweight design and their multi-functional properties. AM parts are often characterised by geometrical imperfections, residual porosity, high surface roughness which typically lead to stress/strain localisations and decreasing the resistance of the structure. This paper aims to focus on the study of the effects of geometrical irregularities and stress concentrations derived from them. Design/methodology/approach In this paper, several technique were combined: 3D tomography, experimental tests, digital image correlation and finite elements (FE) models based on both the as-designed and the as-manufactured geometries of lattice materials. The Digital Image Correlation technique allowed to measure local deformations in the specimen during the experimental test. The micro-computed tomography allowed to reconstruct the as-manufactured geometries of the specimens, from which the geometrical quality of the micro-structure is evaluated to run FE analyses. Findings Experimental and numerical results were compared by means of a stress concentration factor. This factor was calculated in three different specimens obtained from three-different printing processes to compare and understand their mechanical properties. Considering the as-designed geometry, it is not possible to model geometrical imperfections, and a FE model based on an as-manufactured geometry is needed. The results show that the mechanical properties of the printed samples are directly related to the statistical distribution of the stress concentration factor. Originality/value In this work, several techniques were combined to study the mechanical behaviour of lattice micro-structures. Lattice materials obtained by different selective laser melting printing parameters show different mechanical behaviours. A stress concentration factor can be assumed as a measure of the quality of these mechanical properties.

Ultrasonic non-destructive characterization of hybrid additively manufactured 420 stainless steel made with directed energy deposition

Additive manufacturing is remarkably suitable for creating high complexity performance parts. An added benefit of the Directed Energy Deposition (DED) process is the ability to make functionally graded materials. This expansion of the design and manufacturing spaces presents a challenge for the non-destructive evaluation of AM parts. Hybrid AM parts are an example of functionally graded materials for which the variation in microstructure and material properties across the bulk of the parts is created through a combination of manufacturing processes. In this work, a hybrid 420 stainless steel coupon was created using the DED method, and a surface treatment was applied between added layers. Characterization using destructive and non-destructive methods was performed. The microstructure, hardness profile, phase velocity, attenuation, and backscatter results from the hybrid coupon are compared with those from an AM coupon and a wrought coupon. Agreement between destructive and non-destructive measurements is studied. Furthermore, ultrasonic non-destructive methods are shown to be effective for identifying the gradient in the material properties of the hybrid coupon. The work hereby presented can further inform non-destructive evaluation decisions for hybrid AM parts.Additive manufacturing is remarkably suitable for creating high complexity performance parts. An added benefit of the Directed Energy Deposition (DED) process is the ability to make functionally graded materials. This expansion of the design and manufacturing spaces presents a challenge for the non-destructive evaluation of AM parts. Hybrid AM parts are an example of functionally graded materials for which the variation in microstructure and material properties across the bulk of the parts is created through a combination of manufacturing processes. In this work, a hybrid 420 stainless steel coupon was created using the DED method, and a surface treatment was applied between added layers. Characterization using destructive and non-destructive methods was performed. The microstructure, hardness profile, phase velocity, attenuation, and backscatter results from the hybrid coupon are compared with those from an AM coupon and a wrought coupon. Agreement between destructive and non-destructive measurements is ...

Development of an Electropolishing Process for Additively Manufactured 316L Parts Under Direct and Pulsed Potential Control

The initial purpose of additive manufacturing was mainly rapid prototyping, but all industrial sectors have adopted this technique that has increased in profitability and allows innovative shapes in almost all alloys. Advances in technology have led to AM components whose mechanical and structural properties are equivalent to machined parts. However, a major impediment to the extension of the process is due to the bad surface finish of the as built parts, characterized by a high roughness (Ra up to 40 µm) and the presence of un-melted particles and weld beads. These defects are incompatible with the planned applications (like inner channels for fluid flow), and also induce a degradation of the material functional properties such as an increased sensitivity to fatigue and corrosion. Electropolishing, an electrochemical method for planarization of metals based on their anodic dissolution in an appropriate electrolyte, is a solution to consider. During this work a specific electropolishing process dedicated to 316 L AM parts was developed with the originality of being under potentiostatic control [1,2]. First, a preliminary study done at laboratory scale in a three electrodes electrochemical cell, allowed us to determine the process parameters including the working potential range. Then, the same type of arrangement was subsequently up-scaled in a 3 L reactor in order to electrochemically polish bigger parts ( ≈40 cm2). The potential control is made possible thanks to the introduction of a reference electrode in the pilot and by the use of a high current/high voltage potentiostat able to monitor both direct and pulse potentials. When transposing the process at pilot scale, new difficulties are emerging, due to the large initial roughness of the parts. If a satisfactory roughness decrease is easily achievable in less than one hour, it goes with an inhomogeneous loss of dimension. This geometric deformation of the parts needs to be considered as least as much as roughness parameters. In this context, pulsed potential are emerging as an attractive way to performed electropolishing, by controlling the dissolution while getting interesting final surface properties and limited deformation [3]. Samples present a very bright finishing and low roughness values (Fig.1) In this study a special attention was paid to the quantification of the deformation on especially designed grooved test samples by 3D optical microscopy and laser scanning [4]. Acknowledgments : the authors would like to thank the IRT-M2P program AFTER ALM for its financial support. [1] Rotty C., Mandroyan A., Doche M.-L., Hihn J. Y., Surf. Coat. Technol., 307 Part A, 125–135 (2016) [2] Rotty C, Doche M.-L, Mandroyan A.,Hihn J-Y., Montavon G., Comparison of electropolishing behaviours of TSC, ALM and cast 316L stainless steel in H3PO4/H2SO4, Surfaces and Interfaces, 6, 170-176 (2017) [3] Taylor EJ., Sequential electromachining and electropolishing of metals and the like using modulated electric fields, US Patent 6,558,231 (2003) [4] C.M. Sulyma, P.C Goonetilleke, D.Roy, Journal of materials processing technology, 1189-1198 (2009) Figure 1

Mitigation of scan strategy effects and material anisotropy through supersolvus annealing in LPBF IN718

The nickel-based superalloy Inconel 718 (IN718) is an excellent candidate among the existing aerospace alloys for laser powder bed fusion (LPBF) manufacturing. LPBF IN718 has a preference for (001) growth, resulting in a non-uniform, anisotropic microstructure which translates into orthotropic mechanical behavior. The most common heat treatment applied to IN718 is AMS 5662. This treatment was developed 60 years ago for wrought and cast metal forming processes. The small grains and columnar grain structure of LPBF IN718 are not affected by treatments per AMS 5662. Recrystallization and the removal of scan strategy effects have big implications for the acceptance of AM parts. If parts can be heat treated to remove any OEM-related microstructural differences, then parts fabricated on different machines can be printed in any orientation and possess the same properties. This research studies the microstructure of LPBF IN718 as it evolves under an annealing treatment of 1160 ∘C for up to 8 hours. It was hypothesized and later confirmed that this higher-temperature annealing would mitigate the scan strategy effects and anisotropy resulting from the LPBF process. The grain size, shape, and recrystallization are compared throughout the evolution. Additionally, the X–Y and X–Z planes are compared to find a point at which the annealing process results in equiaxed, isotropic grains and the scan strategy effects are mitigated. An equiaxed microstructure was successfully achieved through recrystallization and grain growth, resulting in isotropic microstructure for each scan strategy that was considered. Isotropic mechanical properties were achieved following a modified annealing treatment at 1160 ∘C for 4 hours and validated via nanoindentation and tensile testing.

X-Ray CT Reconstruction of Additively Manufactured Parts using 2.5D Deep Learning MBIR

In this paper, we present a deep learning algorithm to rapidly obtain high quality CT reconstructions for AM parts. In particular, we propose to use CAD models of the parts that are to be manufactured, introduce typical defects and simulate XCT measurements. These simulated measurements were processed using FBP (computationally simple but result in noisy images) and the MBIR technique. We then train a 2.5D deep convolutional neural network [4], deemed 2.5D Deep Learning MBIR (2.5D DL-MBIR), on these pairs of noisy and high-quality 3D volumes to learn a fast, non-linear mapping function. The 2.5D DL-MBIR reconstructs a 3D volume in a 2.5D scheme where each slice is reconstructed from multiple inputs slices of the FBP input. Given this trained system, we can take a small set of measurements on an actual part, process it using a combination of FBP followed by 2.5D DL-MBIR. Both steps can be rapidly performed using GPUs, resulting in a real-time algorithm that achieves the high-quality of MBIR as fast as standard techniques. Intuitively, since CAD models are typically available for parts to be manufactured, this provides a strong constraint "prior" which can be leveraged to improve the reconstruction.