Novel sensorized additive manufacturing-based enlighted tooling concepts for aeronautical parts

Abstract

This paper presents lightweight tooling concepts based on additive manufacturing, with the aim of developing advanced tooling systems as well as installing sensors for real-time monitoring and control during the anchoring and manufacturing of aeronautical parts. Leveraging additive manufacturing techniques in the production of tooling yields benefits in manufacturing flexibility and material usage. These concepts transform traditional tooling systems into active, intelligent tools, improving the manufacturing process and part quality. Integrated sensors measure variables such as displacement, humidity and temperature allowing data analysis and correlation with process quality variables such as accuracy errors, tolerances achieved and surface finish. In addition to sensor integration, additive manufacturing by directed energy arc and wire deposition (DED-arc) has been selected for part manufacturing. The research includes the mechanical characterisation of the material and the microstructure of the material once manufactured by DED-arc. Design for additive manufacturing" principles guide the design process to effectively exploit the capabilities of DED-arc. These turrets, equipped with sensors, allow real-time monitoring and control of turret deformation during clamping and manufacturing of aeronautical parts. As a first step, deformation monitoring is carried out within the defined tolerance of ± 0.15, which allows a control point to be established in the turret. Future analysis of the sensor data will allow correlations with process quality variables to be established. Remarkably, the optimised version of the turret after applying DED technology weighed only 2.2 kg, significantly lighter than the original 6 kg version. Additive manufacturing and the use of lightweight structures for fixture fabrication, followed by the addition of sensors, provide valuable information and control, improving process efficiency and part quality. This research contributes to the development of intelligent and efficient tool systems for aeronautical applications.

Introduction

Aeronautical parts are becoming increasingly sophisticated, particularly in terms of geometric complexity^1,2. This growing complexity poses challenges in the proper tooling of these parts, as conventional manufacturing methods often have limitations in fabricating the necessary components. As a result, manufacturers are compelled to devise production processes that demand meticulous control over every detail. The integration of innovative manufacturing technologies, such as additive manufacturing (AM)³, offers a solution by enabling the customization of critical tooling components that directly interact with the surfaces of the parts being fabricated. The design freedom and manufacturing capabilities of AM empower the development of highly complex tooling, opening new possibilities for tooling orders that were previously unattainable^4,5.

The increasing complexity of aeronautical parts also brings about higher requirements for precision, surface integrity, and tolerance. Ensuring accurate positioning and clamping of parts becomes crucial in mitigating quality risks. By incorporating additive manufacturing into the tooling fabrication process, more precise part positioning and clamping can be achieved. The geometries of the clamping components can be designed to match the final part's geometry, maximizing the adaptability of the tooling to the part⁶.

In the aeronautical sector, Original Equipment Manufacturers (OEMs) place greater emphasis on reducing tooling delivery times, often prioritizing this factor over price when selecting suppliers⁷. Consequently, there is a need to streamline the entire tooling development chain, from design to final validation, encompassing calculations and fabrication stages. The utilization of additive manufacturing capabilities enables the manufacturing of complex tooling components in-house, eliminating the need for time-consuming machining operation outsourcing⁸. Furthermore, the utilization of advanced calculation tools and expertise in their application contributes to reducing delivery time by minimizing design errors that could result in validation failures, necessitating redesigns and fabrication delays⁹.

Aeronautical OEMs are also demanding reductions in the use of resources for tooling manufacturing, given the importance of net-zero climate impact aircraft manufacturing¹⁰. This work achieves reductions in material usage through optimised hybrid manufacturing, which combines additive and subtractive processes. Additive manufacturing allows the production of complex geometries at a competitive cost compared to conventional machining and features high material efficiency, minimising waste compared to traditional methods, making it a cleaner alternative waste^11,12.

The ongoing digitalization and smartization of production processes, driven by Industry 4.0 technologies such as the Internet of Things (IoT) for sensorization and Data Analytics for data analysis and process optimization^13,14, necessitate the integration of digitalization into all manufacturing systems. In line with this trend, the presented work embraces the digitalization and smartization of tooling by incorporating sensors to detect positioning errors and clamping forces. Actuators are integrated into the tooling¹⁵, enabling them to become active tooling systems capable of correcting detected errors. Capturing and recording data from the tooling, including forces, moments, and displacements, facilitates the correlation of these measurements with machining results, such as manufacturing tolerances and surface finish.

The integration of sensors into tooling systems represents a significant advancement in aeronautical part manufacturing, enhancing productivity, precision, and quality¹⁶. For instance, Mahayotsanun et al.¹⁷ demonstrated the importance of in-situ monitoring in stamping processes through mutual inductance-based displacement measurement and distributed contact pressure sensing, which improved the online observability and control of the stamping process. Similarly, Bleicher et al.¹⁸ reviewed the use of various sensors and actuators in machining operations, highlighting their role in improving accuracy, robustness, and productivity, thereby adding intelligence to machine tools. Moreover, Pandey et al.¹⁹ integrated a time domain reflectometry (TDR) sensor line into composite tooling for non-invasive flow and cure monitoring, illustrating the benefits of sensor integration in eliminating ingress/egress issues and reducing maintenance . These studies underscore how sensor integration facilitates real-time monitoring and control, essential for minimizing errors and enhancing efficiency in manufacturing. Integrating sensors into tooling, particularly for complex aeronautical parts, offers significant potential for advancing manufacturing processes by enabling precise control and data-driven optimization¹⁵.

One of the applications of additive manufacturing where the design freedom of AM has been most practically exploited is the construction of innovative structures in production tooling, such as formed cooling structures. These include the pioneering work by Xu et al. in 2001, which demonstrated the potential of solid freeform fabrication technologies to produce tooling with conformal cooling channels, significantly improving production rate and part quality compared to conventional tools²⁰. Zhao et al. recently proposed a design for conformal cooling circuits using metal 3D printing SLM technology, achieving improved temperature distribution and cooling efficiency²¹. Oh et al.²² developed an adaptive conformal cooling method using TPMS structures, reducing cooling time and enhancing mold performance. Additionally, Tan et al. in 2020, optimized internal supports and porous structures to improve cooling efficiency and reduce material costs in LPBF-manufactured molds²³. These studies illustrate the advantages and design flexibility provided by AM in developing high-performance conformal cooling systems for injection molding applications.

In the field of aeronautical accessory manufacturing, the use of metal additive manufacturing has been limited. For example, as reported in the review²⁴, China Eastern Airlines used FDM to rectify misprinted seat signs on its Boeing 777 aircraft, albeit at considerable cost and time. Meanwhile, Northrop Grumman and Moog Aircraft Group used SLA and FDM respectively to manufacture repair kits and manual component maintenance fixtures, streamlining maintenance and reducing downtime. These methods also offer advantages such as consolidation of assembly parts. Advanced Aerials and Honeywell used SLS technology to manufacture rugged parts for unmanned vehicle systems and control pod housings for the RQ-16 T-Hawk UAV, underscoring the versatility and effectiveness of non-metallic additive manufacturing in aerospace applications.

Selecting the appropriate additive manufacturing technology for a given application is a complex decision that involves balancing multiple factors. The most common metal application technologies, with standard acronyms Powder Bead Fusion PBF, Directed Energy Deposition (DED) Laser Beam (DED-LB), Electron Beam (DED-EB), Arc DED-Arc and Metal extrusion (MEX) offer unique advantages and limitations in terms of productivity, part size, geometrical complexity, equipment cost, raw material cost and industrial application. The Table 1 provided summarizes these critical aspects, highlighting the trade-offs involved in the selection process. For instance, while PBF excels in producing high geometric complexity parts, it incurs high raw material costs. On the other hand, DED-Arc, though less capable in terms of geometric complexity, offers high productivity and lower raw material costs, making it suitable for larger parts and applications such as repair and coatings. This comprehensive comparison aids in understanding the strengths and weaknesses of each technology, facilitating an informed decision that aligns with specific manufacturing requirements and constraints.

Table 1.

Summary of qualitative evaluation of Metal Additive Manufacturing technologies.

	PBF 25	DED-LB 26	DED-EB 27	DED-Arc 28	MEX 29,30
Productivity	High	Medium- Low	Medium	High	Medium- Low
Part size	Small	Medium	Medium	Medium-Large	Small
Geometric complexity	High	Medium	Medium- Low	Medium- Low	Medium–High
Price of equipment	Medium	High	High	Medium	Medium
Raw material cost	High	High	High	Low	High
Industrial application	Direct manufacture of complex parts	Repair of parts, coatings, direct manufacturing of parts	Repair of parts, coatings, direct manufacturing of parts	Repair of parts, coatings, direct manufacturing of parts	Manufacture of complex prototypes with non-fully dense material

In its place, this paper instead focusses on the use of Direct energy deposition with arc wire, known as DED-Arc, stands as a robust choice for manufacturing medium-sized parts^31–33. Although its application to tooling has not been widely explored, except in a few cases, it does have cases of success in the manufacture of aerospace parts^11,34,35. This technique offers distinct advantages that align well with the aerospace industry's requirements³⁶. One of the primary strengths of DED-Arc technology is its productivity. The rapid deposition rate enables the fabrication of parts within shortened timeframes, a crucial aspect for the efficiency-driven aerospace sector³⁷. While DED-Arc technology is a great alternative in terms of productivity with deposition rates up to 10 kg/h, its precision may not match that of laser and electron beam processes due to a minimum arc width, limiting the creation of thin walls. This discrepancy in accuracy poses challenges for tooling applications, necessitating meticulous post-processing and additional conventional manufacturing processes to meet industry standards³⁸. This kinetic process finds synergy with energy efficiency³⁹, an area where lasers and electron beams falter, operating at a mere 2–5% and 15–20%, respectively. Here, the electric arc dons a cloak of supremacy, boasting an astonishing zenith efficiency of 90%.

After DED-arc additive manufacturing, post-machining is often necessary to achieve final dimensional tolerances and desired surface quality. This additional machining step addresses the limitations of additive manufacturing, ensuring precise dimensions and surface finish required for various applications. The combination of additive manufacturing and machining in an integrated workflow, known as Hybrid Additive Manufacturing (HAM)^38,40,41, optimizes the production process by leveraging the strengths of each technology: additive manufacturing for complex geometry creation and machining for dimensional accuracy and surface quality enhancement.

Additionally, it distinguishes itself through affordability, with lower operational costs compared to other additive technologies utilizing powders. Material utilization efficiency is a strong suit, owing to the direct and efficient use of wire as input material. In a sustainability context, DED-Arc emerges as an ecologically preferable alternative, generating less waste by products and circumventing the use of potentially hazardous powders.

However, it is worth considering its limitations. DED-Arc might face certain restrictions in geometric precision and flexibility for intricate geometries—crucial aspects in aerospace component manufacturing that demand precision and intricate shapes⁴². Nonetheless, the advantages in terms of cost, production speed, and efficient material use solidify DED-Arc as a robust option for medium-sized aerospace components. As technology advances and processes are optimized, current limitations are likely to be addressed through innovative solutions, further strengthening DED-Arc's role in tooling for aeronautical manufacturing.

In conclusion, the aim of this work is to present a concept of intelligent and lightweight tooling based on DED-arc additive manufacturing that offers a novel vision for the anchoring of aeronautical parts. By integrating a posteriori sensor in a weight-optimised designed tooling, additively manufactured by DED-arc with subsequent machining of the critical faces. This work establishes a procedure for redesigning tooling that integrates sensors and takes advantage of the benefits of additive manufacturing.

Objectives

The main objective of this work is to build enlighted tooling this term refers to tool systems equipped with sensors and advanced data processing capabilities that enable real-time monitoring and adaptive control during manufacturing processes. The intelligence aspect arises from the system's ability to autonomously detect, analyze and respond to deviations in process parameters, ensuring optimal performance and quality. Additionally, it is intended to add intelligence to the design itself by adopting a DED arch design approach. This design guidance focuses on the use of directed energy deposition (DED-arc) technology to optimize material usage, reducing the amount of material required for manufacturing while maintaining structural integrity and performance. The Design and Manufacturing of the Directed Energy Deposition (DED)-Arc Turret constitute a crucial aspect of this research. The chosen application for the implementation of redesign and 3D printing techniques involves the creation of a turret used to secure aeronautical components, as it can be seen in Fig. 1. This turret is attached to a metal structure comprised of extruded profiles and includes a linear guide for adjusting the distance between support points. The complexity of the turret is considered moderate. Its purpose is to facilitate the assembly and disassembly of aeronautical parts. Figure 1a illustrates an example of the assembly, depicting four support points for a standard part. The number of support points may vary depending on the length of the specific part. This particular use case has been previously presented by the authors in a paper⁴³, although their approach to redesign was different and they did not validate the mechanical performance of the redesigned turret through experimental testing. Upon analysing the assembly, it is evident that the metallic structure, constructed using extruded and welded bars, is of low complexity and cost-effective. It can be easily manufactured, potentially even at the location where the clamping fixture is needed. The linear carriages and clamping plates for the turrets are primarily commercially available rails. However, the manufacture of the turret itself is more intricate, involving roughing and finishing machining processes due to tighter tolerances. Therefore, the objective of this paper is to assess the feasibility of using 3D printing for the production of the clamping turret. Figure 1b shows the initial and final tooling and the different component parts.

Figure 1.

Location of the fixturing turrets for the anchoring of aeronautical parts (a) CAD example ¹⁵, and (b) prior clamping solution ⁴⁴ and new proposed solution.

Materials and methods

This section provides an in-depth explanation of the design process of the DED-Arc Turret, encompassing its mechanical and electrical components. The mechanical design involves selecting materials suitable for the specific properties required by the DED-Arc process, such as metallurgical compatibility and weldability. Additionally, the structural design aims to optimize the turret for enhanced stability and precision during fabrication. The design of the sensor part focuses on the assembly of strain gauges and temperature and humidity sensors to guarantee correct positioning of the part once it is located in the structural turret device, with the turrets manufactured by additive and the profile structures. prefabricated.

The subsequent part of this section focuses on the integration of the turret into the frame for the fixturing of aeronautical parts. A lightweight frame solution is sought. The section describes the selection and placement of the bar structure for the subsequent fixing of the turrets to which the sensors that measure the deformation of the turret are added.

Design and manufacturing of the DED-arc turret

Design for Additive Manufacturing (DfAM) involves a comprehensive understanding of the additive manufacturing process and technology intended for fabricating the target part. The Fig. 2 illustrates the phases involved in DfAM of the fixture turret. Each additive manufacturing technology has a unique set of capabilities and limitations. Therefore, it is essential to have a deep understanding of the chosen technology to design effectively. A crucial aspect of DfAM is material characterization, stage 1. This includes not only considering the mechanical properties of the material but also ensuring the integrity of the microstructure within the manufactured part. These two will be key parameters for finite element analysis (FEA) for optimizing the part geometry, important for volume reduction during stage 2 of design. However, material characterization also includes seam geometry in the case of DED-Arc; this analysis will be critical for feeding stage 3 where computer-aided manufacturing (CAM) tools are used. In this stage, trajectories are defined in each layer and layer increments (slicing), and deposition parameters are also defined. These trajectories and parameters are then translated to a machine or robot where part manufacturing takes place, stage 4. Finally, to verify that the turrets meet mechanical requirements, a mechanical validation test under load is performed.

Figure 2.

Flow design followed in this work for the optimization of the fixation turret.

Among the possible DED-Arc technologies, the one based on gas metal arc welding (GMAW) has been used in this paper. The usage of GMAW in this study is justified by several key factors. First, GMAW is particularly well-suited for printing aluminum, which is the material used for the turret. This welding technique provides excellent control over the deposition process, ensuring high-quality welds and consistent material properties^45,46. Second, GMAW is a coaxial technology that is straightforward to operate in a workshop environment, offering ease of use and accessibility. Additionally, GMAW's versatility makes it integrable with robotic systems, allowing for automated and precise manufacturing processes⁴⁷. These characteristics make GMAW an ideal choice for producing the lightweight, sensor-equipped turrets required in this study, as it supports the desired material properties and manufacturing efficiency. Furthermore, leveraging GMAW in the Directed Energy Deposition (DED-arc) process offers several advantages. The technology enables high deposition rates and excellent control over the thermal input, which are crucial for maintaining the structural integrity and mechanical properties of aluminum components The turrets were fabricated using Al 5356, the composition of which is shown in the Table 2, supplied in the form of a commercial wire with a diameter of 1.2 mm, while flat plates of 15 mm thickness made of Al 5356 were used as the substrate. In terms of parameters, the fabrication process involves the use of the AC PULS welding mode on the Titan XQ 400 AC PULS equipment by EWM. This equipment allows for a wide range of wire feed speeds, ranging from 2 to 12 m/min for this specific material. To ensure optimum shielding and protection during the process, pure argon gas with a flow rate of 30 L/min was chosen.

Table 2.

Composition in weight percentage of Al 5356.

Element	Si	Fe	Cu	Mn	Mg	Zn	Cr	Ti	Al
% weight	0.25	0.4	0.1	0.05–0.2	4.5–5.5	0.1	0.05–0.2	0.06–0.2	rem

The torch inclination angle with respect to the substrate was set at a fixed 90 degrees, maintaining a consistent Stick-out distance of 15 mm. A 20 mm diameter nozzle was utilized for the welding process. These selected parameters have different conditions in the first layers and a subsequent modification until reaching the upper part of the structure considering the thermal state of the part, as shown in Table 3. The process parameters were chosen using a combination of factors. Parameters selection was based on information provided by the supplier, basic parameters, recommendations, and guidelines from the manufacturer of the power generating equipment, in line with industry best practice. In addition, comparative tests were performed on different transfer modes and parameters for welding in two previous articles by the authors^46,48.

Table 3.

Parameters used for the manufacture of the fixing turret.

Layer number	Mode	Wire speed (m/min)	Deposition velocity (kg/h)	Stick out (mm)	Travel speed (cm/min)	Waiting time (sec)	Layer height (mm)
1	AC PULS	8	1.44	15	114	0 s	1.5
2–8	AC PULS	8	1.44	15	200	0 s	1.4
Remaining	AC PULS	8	1.44	15	168	0 s	1.5

In the material characterization, a detailed comparison of the tensile test results for DED-Arc AA5356 alloy in both the horizontal (HD) and vertical (VD) directions was performed. Tensile specimens were prepared from DED-Arc AA5356 material according to the ASTM E8 standard. The specimens had a diameter of 4 mm and a gauge length of 22 mm. Samples were extracted both in the horizontal direction (HD) and in the vertical direction (VD) to assess any directional variations in mechanical properties. The tensile tests were conducted using Intron testing equipment equipped with a 100 kN load cell. The testing speed was set at 1 mm/min, ensuring a controlled and consistent rate of deformation during the test. This standardized testing procedure allowed for measurement of the mechanical behaviour of the DED-Arc AA5356 alloy in both the HD and VD directions. The mechanical properties, specifically the 0.2% yield strength, ultimate tensile strength (UTS), and elongation, were analysed. In the HD direction, the 0.2% yield strength was found to be approximately 148.25 ± 5 MPa, while in the VD direction, it measured approximately 146.88 ± 4 MPa. The UTS values for the HD and VD directions were 276.67 ± 3 MPa and 264.33 ± 4 MPa, respectively. Moreover, the elongation exhibited by the AA5356 alloy in the HD direction was approximately 81.58 ± 4%, while in the VD direction, it measured approximately 78.60 ± 3%. Based on these results, it can be concluded that the mechanical properties of the DED-Arc AA5356 alloy are quite similar in both the HD and VD directions, as indicated by the comparable values of yield strength, UTS, and elongation.

As for the microstructure of the AA5356 aluminum alloy, the resulting deposited material presents the typical characteristics of DED-arc. One of the main characteristics of the microstructure is the presence of fine equiaxed grains. Equiaxed grains refer to grain structures that have similar dimensions in all directions, giving them a more isotropic nature. These grains are formed due to rapid solidification during the DED process, where the molten material solidifies rapidly as it is deposited. It is also important to note the presence of intermetallic phases. Aluminum alloys, including AA5356, form intermetallic compounds due to the presence of alloying elements. These intermetallic phases have varying compositions and morphologies depending on the specific alloy composition and processing conditions. Overall, the microstructure of aluminum alloy AA5356 produced using DED with wire and arc generally consists of equiaxed grains, columnar grains, and intermetallic phases. Figure 3 shows a mono-seam fabricated wall using the parameters in Table 3. This characterisation is important to feed the FEM model to optimise the geometry and estimate the growth per layer for CAM programming.

Figure 3.

Single-wall macro structure made by DED technology from aluminum alloy.

The original turret had a weight of 6 kg, while the optimized version after applying DED technology weighed only 2.2 kg. The optimized design follows an arc-based approach similar to architecture solutions, and it was previously described in⁴⁹. Finite element modeling (FEM) methodology was employed using Ansys software for linear elastic isotropic modeling. The analysis type conducted was static analysis with solid meshing. The mesh had sixteen Jacobian points for mesh quality and 100,818 nodes for 69,829 elements, with a maximum aspect ratio of 4.6451. Figure 4a shows the basic dimensions of the turret and the axial force and anchorage restrictions. Also included are the datum surfaces which are those that have geometric shape and positional tolerance constraints. These datum surfaces may not be varied in the volume reduction process. Finally, Fig. 4b shows the turret under the tensile and compression validation test to which it has been subjected to establish its suitability.

Figure 4.

Dimensions and constraints: (a) Basic dimensions of the turret, axial force, anchorage restrictions, and datum surfaces for the FEM analysis (b) Turret under tensile and compression validation tests.

The trajectories for the turrets are designed using computer-aided manufacturing (CAM) software, specifically Autodesk PowerMill. The fabrication occurs layer by layer with intermittent stops to allow for temperature stabilization. Oscillation-based trajectories are employed, where contours are first created and then filled using oscillatory movements. The oscillation distance from the contour is set to 2 mm. In the matrix demonstrator consisting of four turrets, the trajectories are designed for simultaneous fabrication of all four turrets. The different stages of manufacturing the turrets that are the subject of this study are shown in the Fig. 5 image sequence.

Figure 5.

Additive manufacturing of the turrets: (a) start of columns manufacturing, (b) start of haunch manufacturing, (c) last step of top plate manufacturing and (d) blanks manufactured by DED in parallel.

Although DED offers significant advantages in terms of weight reduction and design optimization, there are certain drawbacks to consider. One such limitation is that DED-produced parts often require additional machining on critical reference surfaces or areas that interact with other components to meet the needed tolerances. This additional step is necessary to achieve the desired precision and ensure proper functionality of the final piece. Once the turrets were manufactured using additive technologies, mechanical tests were conducted to evaluate their mechanical performance and determine if they met the requirements of the fixtures. One part was subjected to mechanical tests after machining the datum surfaces. However, it is important to note that no additional post-treatments, such as surface finishing or heat treatment, were performed.

In aluminum, the impact of residual stresses is generally less critical compared to other metals, but it still warrants careful consideration. Aluminum has a lower density and higher thermal conductivity than many other metals, which helps dissipate heat more quickly and reduces the extent of residual stresses during the DED Arc AM process⁵⁰. However, aluminum's relatively high coefficient of thermal expansion means that any residual stresses that do form can still lead to dimensional instability, warping, or reduced mechanical performance if not properly managed. In this study, the final machining of reference surfaces helps to mitigate these potential issues by relieving some of the residual stresses and ensuring the turret meets the required dimensional and surface finish specifications. Despite aluminum's generally lower susceptibility to severe residual stress effects compared to other metals, the consideration and management of these stresses remain important to ensure the long-term functionality and reliability of the manufactured part. Consequently, while the mechanical properties of the parts could be improved, no additional enhancements were applied to allow for the completion of the turrets in three phases: manufacturing, machining of datum surfaces, and installation of strain gauges.

Figure 6 presents a mechanical compression test to check the mechanical behaviour of the component, with the constraints described in Fig. 4. The fabricated turrets were subjected to tensile and compression tests to evaluate their behaviour under mechanical stress. The tests were carried out in a 0° orientation using a test rig. Fastening was carried out at the base of the turret by means of five bolts of metric 10 as it would be carried out in the final assembly of the turret. The application of the intended loads is carried out, at the top, by using the 18.5 mm cross hole to accommodate the pulling tool. The range of force applied during the tests was determined by the maximum allowable deformation of the turrets during operational use. To establish this limit, a control point was defined on the front face of the turret. The maximum permissible displacement at this point was set at ± 0.15 mm. The test results provided are for the final turret and include tensile and compression tests performed at 0 degrees along the Z-axis. These tests provide valuable information on the mechanical behaviour and performance of the turrets.

Figure 6.

Part mounted on the test bench for tests oriented at 0°.

In the tensile test (test 0 degrees traction), the applied force (F) increased incrementally from zero up to 500 N. At the end of the test, the displacement along the Z-axis (Z) was recorded as 0.0072 mm. Similarly, in the compression test (test 0 degrees compression), the force (F) also increased gradually from zero to 500 N. The resulting displacement along the Z-axis (Z) was measured as 0.0478 mm.

Structural frame modeling for turret system

Aeronautical tools are traditionally made up of very rigid frames, normally built from mechanically welded steel profiles (Fig. 7c), and of turrets and positioning tools with sometimes complex geometries that either have high manufacturing costs or are oversized parts because trying to optimize the geometry entails additional machining costs that are not affordable. This section focuses on the analysis of the possibility of replacing these positioning frames with lighter structures (aluminum profiles such as Fig. 7d), and the behaviour of all the tooling under gravity and reaction forces has been simulated (Fig. 7b). The previous section studied the lightening of the turrets mounted on the frame, analysing, and optimizing them for their manufacture by means of additive technologies. In this way, it has been possible to design an optimized and lightened aeronautical tooling model. The study is carried out on a reduced model of the tooling shown in Fig. 7a.

Figure 7.

Information concerning the design and modelling of the frame: (a) First proposal for a frame made of aluminum profiles, (b) simplified model with the addition of reinforcements and anchor points, (c) comparison between profiles used for the frame and (d) finite element model of the aluminum profiles.

The proposal entails replacing the mechanically welded tubes with lightweight aluminum profiles to achieve frame weight reduction. Naturally, it is imperative to ensure that the aluminum profile structure intended to replace the welded assembly can withstand the imposed forces without deforming beyond the tolerances of the fixture. In the present context, the "points of control" (P1, P2, P3 and P4), which are those needing to remain within tolerances, correspond to the points where the workpiece to be processed contacts the fixture. These contact points are illustrated on the different turrets affixed to the frame in the provided image.

Consequently, there are four critical points, one on each turret, that necessitate monitoring to ensure that the fixture aligns with tolerance requirements. Specifically, the fixture specifications dictate that these points must stay within a tolerance range of ± 0.15 mm. Regarding the forces acting upon the structure, the ensuing table outlines the reaction forces evident on the turrets.

Furthermore, the streamlined version of the fixture is depicted in the subsequent image. The original framework's square-section tubes, measuring 120 mm by 120 mm with a thickness of 3 mm, were substituted with extruded aluminum profiles measuring 40 mm in dimension. The overall weight of the frame has been significantly reduced, transitioning from 93 kg of the welded tube structure to 33 kg with the aluminum profile configuration, marking a notable 65% reduction.

The subsequent phase involves generating a finite element model of the fixture to verify that the bench's behaviour under stress aligns with specifications. The stresses imposed upon the fixture are relatively low, resulting in modest structural tensions. However, stringent geometric tolerances must be adhered to—recall, ± 0.15 mm of maximum deformation at the points of contact on the turrets. Should the proposed structure fail to meet this requirement, reinforcement will be necessary until deformation values in line with specifications are achieved.

The following outlines the modelling approach for the various elements constituting the fixture. The frame made by aluminum profiles (Fig. 7d) has been modelled using "shell" elements (2D). To facilitate meshing, slight modifications have been applied to the aluminum profile geometry, such as the removal of rounded edges. On the other hand, the towers, to streamline and expedite calculations, have been substituted with rigid bar models (Fig. 7b). These models encompass an equivalent point mass, representing the weight of the workpiece, alongside the corresponding control point (P), where displacement measurements are of interest.

Structural calculations were carried out to verify the admissibility of displacements at the control points. The structure was evaluated under two distinct loading conditions, with displacement measurements recorded at the four turrets (points P1 to P4): gravity (self-weight) and gravity + reaction forces, see Table 4. The obtained displacements at the turrets for both loading scenarios are provided in Table 5.

Table 4.

Displacements obtained in the turrets for both load cases.

Load case	P1 [mm]	P2 [mm]	P3 [mm]	P4 [mm]
Gravity	0.048	0.057	0.035	0.029
Gravity + Reaction forces	0.190 (> 0.15)	0.185 (> 0.15)	0.072	0.078

Table 5.

Weights and centres of gravity of the turrets.

Turret	Weight/Reaction forces [kg]	Position
		X [mm]	Y [mm]	Z [mm]
1	3.6/9.8	148.059	− 254.401	14.215
2	3.6/9.8	148.059	− 704.401	14.215
3	3.6/4	173.27	− 267.046	462.369
4	3.6/4	173.27	− 717.046	462.369

In the case of a structure under gravity and reaction forces, Fig. 8a, the behaviour of the structure (Model a) under its own weight is assessed. The central portion of the structure slightly bends due to the absence of support legs in the middle. One side experience more pronounced deformation due to the significantly greater load of the turrets on that side Fig. 8a. Based on these results; several additional calculations are performed with different types of stiffening techniques:

• Model b (Fig. 8b): Structure with six ground anchor points (two additional legs added in the central section).

• Model c (Fig. 8c): Structural stiffening through transverse profiles and 4 support legs.

• Model d (Fig. 8d): Stiffening through transverse profiles and 6 support legs (combination of stiffening from Models b and c).

Figure 8.

Displacements in the control points under the effect of gravity and load for (a) initial model, (b) two supports at the bottom model b, (c) stiffening bars at the top model c and (d) combination of solutions b and c in model d.

The findings from this analysis provide insights into the deformation behaviour of the initial proposed design. The study indicates areas of concern in terms of displacements and deformation that require addressing to meet the specified tolerances. Subsequent models with varying reinforcement strategies will be explored to identify effective solutions to ensure the structural integrity of the tool under different loading conditions.

The comparative analysis of different design alternatives reveals crucial insights into the structural behaviour of the tooling system under various loading scenarios. Initially, in the preliminary design (Model a), the longitudinal profiles supporting the turrets exhibited torsional deformation under lateral loads, indicating excessive flexibility. Although the addition of six support legs (Model b) managed to mitigate this to some extent, the longitudinal bars continued to display torsion. Table 6 shows a summary of the deformation results of the control points under load and gravity. Subsequent adjustments, however, led to substantial improvements. By adopting transverse profiles between the turrets and reverting to a structure with four ground anchor points (Model c), the undesired torsion of the longitudinal bars was effectively prevented. Despite the absence of central legs, the central flexing phenomenon persisted. As a direct result of the stiffening efforts, deformations under lateral loads were considerably reduced, falling well within the 0.15 mm tolerance range. Model d, incorporating transverse profiles and six support legs, exhibited the most notable improvement in deformation performance, achieving reductions of up to 85% compared to the preliminary design. Notably, the transverse profiles introduced an enhanced level of structural stability, curbing undesired torsional behaviours while maintaining structural integrity.

Table 6.

Simulated deformations at the control points in the different design models.

Design alternatives	P1 [mm]	P2 [mm]	P3 [mm]	P4 [mm]
Model a (preliminary design)	0.190	0.185	0.072	0.078
Model b (6 support legs)	0.177	0.172	0.078	0.085
Model c (transverse profiles and 4 support legs)	0.055	0.055	0.031	0.027
Model d (transverse profiles and 6 support legs)	0.028	0.022	0.010	0.013

This comparative analysis underscores the significance of proper stiffening strategies in achieving the desired structural performance. The iterative design process, culminating in Model d, showcases the effectiveness of targeted modifications in ensuring that the tooling deformation remains well within the specified tolerances under various loading conditions.

Result of intelligence on the fixturing of aeronautical parts

The work undertaken thus far has focused on the development of sensor installation in the fixtures used for precise assembly of aeronautical components. The fixtures, designed for high-precision placement of components, have traditionally been passive tools. However, in response to the current slowdown in the aerospace sector, there is limited demand for fixtures of alternative types in the near future. Nevertheless, this paper aimed to enhance engineering capabilities by exploring monitoring systems for various variables, such as deformations and forces. By incorporating control electronics and sensors, it becomes possible to monitor and acquire data in real time, opening possibilities for operator assistance, corrective actions, and historical analysis. This is primarily due to limited knowledge of the mechanical demands imposed on the assembled components using these fixtures. The sensor installation process involved selecting and mounting sensors on the fixtures. The sensors were connected to a central hub, enabling data acquisition. This centralized hub, referred to as the W-BOX, facilitated the connection of all the sensors present in each fixture. The W-BOX, in turn, was connected to a main control cabinet. To ensure proper installation, strain gauges capable of measuring real-time deformations were employed on each fixture's four turrets. These strain gauges were strategically positioned to monitor and record deformation data during fixture operations.

In addition to sensor installation, an online expert system was developed to collect and visualize the data captured by the sensors in real time. The monitoring of these parameters can be observed through dedicated software, providing immediate access to the collected data and enabling further analysis and insights. The range of force applied during the test was determined based on the maximum allowable deformation in the piece during its operation. A control point on the front face of the turret was defined, and the maximum allowable displacement at this point was set at ± 0.15 mm.

Consequently, an in-depth study has been undertaken to investigate methods and sensor systems for real-time monitoring of assembly variables, covering forces, moments, displacements, vibrations, and temperatures, among others. This study has led to the formulation of an overall system that integrates sensors to interconnect fixtures and monitor a few relevant parameters. The first actions consisted of selecting and integrating the sensors into the fixtures, as shown in Fig. 9a. The illustration shows the incorporation of a vibration sensor into the device. The illustration highlights the incorporation of a W-BOX in each luminaire, which serves as a hub for all sensors in the respective luminaire; these W-BOX units are connected to a central cabinet, as shown in the diagram. The detailed connections between the W-BOX and the sensors, shown in Fig. 9b, reveal the intricate network of sensors in each luminaire. Finally, the installation procedures for each fixture's sensors are shown in Fig. 9c, following the sequential order of actions. These sensors, in the form of strain gauges, actively measure the real-time deformations in the four interconnected mounting turrets of the sensor-equipped luminaire.

Figure 9.

Displacements in the control points under the effect of gravity and load for (a) initial model, (b) two supports at the bottom model b, (c) stiffening bars at the top model c and (d) combination of solutions b and c in model d.

The communications box that connects to the main cabinet is a W-BOX (wireless box), which has a Lord V-LINK-200 multi-function sensor module, an 8-channel analogue input wireless sensor node for acquisition and transmission, capable of interfacing with sensors and transmitting data wirelessly. The strain gauge specified is the Vishay Micro Measurement CEA-06-250UNA-350 strain gauge, with a resistance of 350 ohms, arranged in a Wheatstone bridge. The wireless communication protocol is IEEE 802.15.4 for efficient data transmission. In addition, ambient temperature and humidity sensors are connected directly to the V-LINK-200.

The W-Box sends the data to the central cabinet where another wireless communication unit, the WSDA-2000 Network Connected Wireless Gateway, is available for data acquisition from wireless and inertial sensors. The WSDA-2000, with 4 GB of onboard memory for data logging and an Ethernet connection, connects to a Beckhoff industrial PC. This PC is connected to a 4G router and uploads the data to the cloud platform.

This sensor and communication network enables comprehensive monitoring capabilities. Parameters that can be monitored in real time within the installation include system-level metrics, strain gauge measurements, and environmental parameters (temperature and humidity). System-level metrics allow the identification of correct system operation and the detection of voltage drops and overload events, facilitating the localization of possible installation errors. Additionally, real-time visualization of the strain results, measured with strain gauges, is graphically represented in Fig. 9d.

Once the assembly of the tool sensorization was described, the system for collecting and visualizing the parameters measured by the sensors in real-time was developed. This cloud-based software application consists of four tabs. Firstly, there is the Dashboard related to tool sensorization and its status. In this screen, the connection status of the different elements that make up the connection system is observed. Figure 10a shows a screenshot of the Dashboard screen. Then there is a tab named "Download," where the electrical diagrams for the sensor mounting, W-Box, and general communication cabinet can be accessed. Next are those related to data inquiry, named "Data" (Fig. 10b), where historical data can be queried between two dates. Lastly, there is the Expert Knowledge System (abbreviated as SEC acronym in Spanish for Sistema Experto de Conocimiento), where control and monitoring algorithms for clamping are implemented. Currently, the algorithm has a threshold for deformation set at 0.15 and − 0.15, which triggers when the deformation of any of the turrets deviates from the tolerance during the process. The screen for inquiry would be the same as the one shown in the previous Fig. 9d

Figure 10.

Screenshots of the cloud-based software application. (a) Dashboard screen and (b) Data screen.

In short, until now, the initial fixture employed a passive tooling solution used for precise component placement in high-precision assembly processes. In the short term, due to the temporary hiatus in the aerospace sector, demand for alternative types of tooling is not foreseen. However, through this initiative, efforts in this work have been directed towards exploring systems for monitoring various variables (such as deformations and forces), achieved through sensorization. These efforts expand the possibilities for operator assistance, corrections and historical data collection to facilitate the study of fastener behaviour during operation. This introduces challenges in enacting active responses due to the lack of knowledge of the mechanical demands placed on the assembly used.

Conclusions and future lines

This paper investigates the application of design for additive manufacturing (DfAM) to sensor tooling for aeronautical parts. The DfAM necessitates a comprehensive grasp of additive manufacturing processes and technologies. Each additive manufacturing technique possesses distinct capabilities and limitations, emphasizing the importance of intimate knowledge of the chosen method to ensure effective design strategies. Integral to DfAM is the meticulous characterization of materials, encompassing mechanical properties and microstructure integrity. This understanding empowers designers to make informed choices, optimizing designs for additive manufacturing. Concurrently, meticulous consideration of additive manufacturing parameters, down to layer-by-layer growth, is paramount. Selecting appropriate layer paths guarantees successful deposition, fostering favourable fabrication outcomes.

DfAM capitalizes on additive manufacturing technology's unique features, allowing designers to exploit its strengths to enhance designs or reduce weight. By harnessing the inherent advantages of additive manufacturing, designs can be optimized in ways unattainable through traditional manufacturing routes. Notably, the DfAM process dovetails into manufacturing tests vital for ensuring the desired part construction. Informed by additive manufacturing insights, design shapes the selection of suitable testing procedures and evaluation benchmarks. This iterative process fine-tunes the design and manufacturing parameters, culminating in high-quality, functional components.

This study's focus on gas metal arc welding (GMAW)-based DED arc technology for fabricating turret structures underscores its multifaceted complexity. The aluminum alloy used as a material supports the manufacturing process. The installation of sensors that feed data in real time for informed decision making contributes to the correct positioning of parts in the aeronautical sector. Material characterization reveals consistent mechanical properties, making the alloy suitable for its intended applications.

Furthermore, the investigation extends to integrating intelligence into the fixture of aeronautical parts. The proposed approach replaces traditionally rigid frames with lightweight aluminum profiles, underpinned by finite element modelling. Sensor installation into fixtures amplifies monitoring capabilities, enabling real-time data acquisition, operator assistance, and historical analysis. The study successfully demonstrates how sensor networks and cloud-based software contribute to real-time data collection, parameter visualization, and system monitoring.

Ultimately, this holistic approach not only advances DfAM methodologies but also showcases the intricate interplay of technology, design, material, and process optimization within the realm of additive manufacturing and aeronautical part fabrication.

Author contributions

Conceptualization, VU, FV, AS and TB; Data curation, VU, FV, IG and AL; Formal analysis, VU, FV, IG and AL; Investigation, VU, FV, IG and AL; Methodology, VU, FV and AL; Project administration, AL; Supervision, AS; Validation, TB and AS; Writing—original draft, VU, FV and TB; Writing—review & editing, VU, FV, TB and AS.

Funding

The authors thank the Basque Government for funding the ADIFIX, HAZITEK 2019 (ZL-2019/00738) programs.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-68786-w.