Comparison of process performance of AdditiveManufacturing FRTP in a lifecycle perspective for

aerospace application

Diogo Rodriguesdiogo.carvalho.rodrigues@tecnico.ulisboa.pt

Instituto Superior Tecnico, Universidade de Lisboa, Portugal

November 2018

Abstract

Additive manufacturing (AM) has recently been in the spotlight regarding the aerospace industry. Thepossibility of improving critical aspects such as weight, complex designs or material corrosion has ledsome manufacturers into spending some time and resources into the development of this technology, for-merly used almost only in the prototyping industry, with the objective of making end use parts. This thesisfocuses on the design and production of a leaf-type spring landing gear for a UAV, using an AM processknown as Fused Deposition Modeling (FDM), to produce a Fiber Reinforced Thermoplastic (FRTP) com-ponent. The main objective is to develop an FRTP part by FDM from its original design while respectingthe initial functional requirements i.e., the ability to sustain the landing loads. This process of designtakes in account the limitations of the 3D printer, the 3D printer software and the material properties.The part will be compared with a traditional part, in this case an aluminum part manufactured by CNCMilling, a subtractive production process. For that purpose, all the required data from Milling and FDMproduction technology will be collected and a comparison established. The collected data includes thetime of production, energy consumption, amount of material used, among others, and it will be used todevelop a cost model and an environmental impact model. The life cycle costing (LCC) objective is todetermine the unitary cost of one part, for the two distinct production methods, and to compare them, tounderstand if it is viable to produce this component using a 3D printer instead of CNC Milling, that is moreimplemented in the industry. Finally, it is made a life cycle assessment (LCA) to evaluate the environ-mental impact of both technologies for this particular component. This analysis was made using an LCAsoftware. It was concluded that the component produced in aluminum by CNC Milling is better comparingthe mass and volume of the part and has significantly lower production costs. It was also concluded thatthe environmental impact is lower for the FRTP part produced by AM FDM.Keywords: Additive Manufacturing (AM), Fiber Reinforced Thermoplastics, Landing Gear, Life CycleCosting (LCC), Life Cycle Assessment (LCA)

1. Introduction

In a world were sustainability is becoming a keyword in economy, environment and society every-one is always trying to optimize existing proce-dures and technologies or creating new ones [1].Additive Manufacturing (AM) is a manufacturingtechnology that consists in adding material layerupon layer, instead of removing the material, to ob-tain the desired part. This technology has advan-tages and disadvantages compared to other tech-nologies. One example of AM advantages is theability to manufacture components with complex in-ternal geometries, mass personalization, and lowmaterial waste. The main disadvantages are thelong manufacturing time, and the small produc-tion volumes possible. The AM process used in

this thesis is Fused Deposition Modeling (FDM)that consists on extruding melted material trougha heated nozzle. The material is deposed layer bylayer until the final component is formed. The maingoal of this thesis is to develop a FRTP part pro-duced by AM FDM and compare it to an aluminumpart produced by CNC Machining, a subtractiveproduction method. The parts and processes willbe compared in terms of mechanical performance,production costs and environmental impact. Thechosen part to be compared is an UAV (UnmannedAerial Vehicle) landing gear and it must meet thesame mechanical requirements with both materi-als. The process of designing the FRTP landinggear is an iterative process, because of the lack oftools to analyze the behavior of parts made with

1

this material. The goal is to respect the initial re-quirements, the ability to support the landing loads,while exploring the possible aspects where the de-sign could be altered for FDM, while trying to keepthe volume and mass of the part as low as possi-ble. This process is influenced by the 3D printerand its software restrictions and also by the cost ofthe fibers used to reinforce the part. After obtainingthe best design a Life Cycle Costing (LCC) modelwas developed to understand which part is moreexpensive. In this model it is considered a cradle-to-gate assessment, which is an assessment of apartial product life cycle from resource extractionto the factory gate. Because of the lack of data, itwas excluded from the model the use phase andthe disposal phase. So, it was determined the uni-tary cost of a FRTP and an Aluminum part. A LifeCycle Assessment (LCA) model is also developedto determine which component has the higher im-pact on the environment. Unlike the LCC model,in the LCA it was considered a cradle-to-grave as-sessment, taking in account the impact of the usephase and the disposal phase. Finally, the alu-minum part and the FRTP part are compared interms of mass, volume, costs and environmentalimpact. The drawbacks and difficulties of this par-ticular AM process, FDM, and equipment are men-tioned. It was also explored the possible aspectswhere the design could be improved, if the materi-als used explore the full capabilities of the designand some points where the AM equipment and itssoftware can improve. Future work is proposed tocomplement subjects present on this thesis and toexplore some points that were not possible to ad-dress here.

2. Bibliographic ResearchIn the case study ahead, a comparison between acomponent produced in aluminum by CNC Millingand other in FRTP by FDM will be made. This com-parison is focused on the mechanical characteris-tics, a cost analysis and a environmental analysis.

2.1. Additive Manufacturing (AM)Additive Manufacturing is a manufacturing processthat starts from a 3D model data and the designedpart is formed by adding layer upon layer until thepart is completed and the production is finished.Contrary to the more traditional production meth-ods, subtractive methods, the main difference isthat with AM the part is formed without materialwaste.[2].

AM processes main applications are prototypingand personal users (3D home printers), but nowa-days it is gaining space in industry production. Theimprovement of AM technologies is increasing itsuse by companies to produce components [3]. AMcan be used to produce models, prototypes, end

use parts, assemblies or even tooling. AM can beused to produced parts alongside other productionmethods, by alternating printing and machining op-erations. For example, it is possible to print em-bedded components, which consists on one com-ponent of another material being inserted in theAM part during production [4]. This componentis considered in the design of the part and in theprinter software, so that the printing is optimizedto accommodate the new material. In some pro-cesses, the printing can be interrupted in order toinsert a material and then the printing is restarted.The flexibility of materials and procedures aroundAM technologies allow the production of complexshapes and designs using a wide variety of mate-rials. AM parts have a large number of materialsavailable that can be used in their production, frommetals as aluminum or steel alloys, to plastics oreven edible materials as chocolate or sugar [4][5].One important branch of 3D printing is the poly-mer based low-cost 3D printers that allowed thegeneral public to express creativity and produceits own parts as a hobby at their own home. Thistechnology has been improving in quality and theprices are expected to drop in the near future, al-lowing the 3D printing market to grow even moreamong common users [5]. This technology appli-cations cover a large range of industries such asautomotive, aerospace, medical prosthetics, toolsproduction, electronic components, jewelry, furni-ture, sports equipment and numerous other areas.

2.2. Fused Deposition Modeling (FDM)

Fused Deposition Modeling (FDM) is an additivemanufacturing process which consists on extrud-ing the material through a nozzle that also melts it.The extruded material is deposited in the print bed(base) that moves up and down allowing the lay-ered production. This process is one of the mostused within the AM technology [6].

The more common materials used with this pro-cess are thermoplastics, that are polymers whichmain characteristic is that after it has been heatedit becomes moldable, but when it’s cooled it solid-ifies. Some examples of these materials are ABS(Acrylonitrile Butadiene Styrene), Nylon, Polyethy-lene among others. These materials have rela-tively low strength, so, to improve this feature, itwas introduced fibers in the process, such as glassfiber or carbon fiber. The resulting material is aFiber reinforced thermoplastic (FRTP), a compos-ite material, that combines the fiber and the ther-moplastics improving the properties that both ma-terials have by themselves.

The machine used in the case study is the Mark-forged Mark II, that is the first commercial avail-able 3D printer that is able to print Fiber Reinforced

2

Thermoplastics, using the FDM process.

2.3. CNC MillingAs opposed to additive manufacturing, subtractivemanufacturing consists in successively cutting ma-terial from a solid block of material until a 3D part isobtained. This process can be done manually, butnowadays is typically done with a CNC machine.CNC stands for computer numerical control and isthe automation of machine tools that are usuallymanual operated. It includes any type of machining(Drilling, Milling, Water-jet, laser, etc), as long asthe process is automated and controlled by com-puter. Milling is a machining process that uses ro-tary cutting tools to remove material from the initialsolid block. Unlike drilling where the rotary tool onlymoves along the rotation axis, in milling the rotarytool can also move perpendicularly to this axis, so,the cutting occurs with the side of the cutting tool.

3. Methodology

Figure 1: Case Study Methodology Diagram

The methodology for the case study, where acomparison will be made between a componentproduced in aluminum by CNC Milling and in FRTPby FDM, is present in the diagram of Figure 1. Itstarts with the aluminum design which has a setof reference values of displacement and load toachieve. This design is analyzed with a FEM soft-ware and the results are compared with the initialrequirements. If they don’t match the part has to beredesigned. If they match, the results will be usedto perform an environmental and cost analysis.

3.1. Mechanical PropertiesThe mechanical test performed meant to replicatethe vertical loads applied on the landing gear whenlanding. In Figure 2 it is possible to see that thelower section of the part is fixed with a bolt, not al-lowing any movement. This section, that is bolted,is the connection to the aircraft, while the uppersection, where the force is applied from top to bot-tom, corresponds to the wheel connection. Be-

cause the objective is to compare the FRTP partwith the aluminum one, the load setup is the sameused in the FEM simulation made for the aluminumpart.

Figure 2: Mechanical Test

With this setup, the load applied is increasinggradually and at the same time the displacementis measured. So, we can determine what loads areapplied in the range of displacement defined andcompare them to the loads the aluminum part en-dures within the same displacement range.

3.2. Life Cycle Costing (LCC)Machine Cost: The machine cost of the Mark-forged Mark II is calculated based on Equation 1,considering annual cost of the equipment and theannual productive hours.

Ceq = Caq.eq

[i(1 + i)n(

1 + i)n − 1

]1

dyearhdayhunit (1)

Ceq is the cost of the equipment that we wantto calculate, Caq.eq is the cost of acquisition of themachine, i is the opportunity cost, n is the depreci-ation time in years, dyear is the productive days ofthe year, hday is the productive hours of the day andhunit is the hours of production of one unit.

Energy Cost: The energy cost is calculated bysimply multiplying the value of the cost of the en-ergy per KWh by the Energy used by each ma-chine, as in Equation 2.

Cenergy = CKWhEused (2)

Labor Cost: The labor costs for the Markforgedmark II are calculated using Equation 3, in whichthe first term is the hourly cost of a worker to thecompany and the second term is the hours theworkers spend with the production of one part, re-sulting in the cost of labor per unit. It will be consid-ered both specialized and not specialized workers,so this equation needs to be calculated for each

3

type of worker and the final value is the sum of thetwo.

Clabor =SmonthNsalCsoc + 11Alunch

dyearhdayfphwork (3)

Smonth is the salary of the worker per month andhwork is the time in hours spent by one worker toproduce one piece (note that despite the time ofproduction of one piece being more than 11 hours,as the system is automatic, the workers are not100% dedicated to the part the whole time), andboth are different if the worker is specialized or not.Nsal is the number of salaries per year including va-cations salaries. Csoc is the social contribution thecompany makes for each employee. dyear is theworking days in a year. hday is the working hoursper day. fp is a factor of productivity. Alunch is thelunch allowance.

Finishing Cost: After printing, the parts are notyet ready to be used, needing some post process-ing. In the case of the aluminum part, first it needsa rotary tool to remove shavings and excess mate-rial. Then the part is sanded to provide a smoothersurface and prepare for painting. Later, the partis painted providing an extra protection and an im-provement in aesthetics. Finally, is applied an UVcoat for protection. In the case of the FRTP part,the post processing consists in removing the sup-port material. Because this part is not prone to cor-rosion and the final color of the part can be deter-mined by the material used in the printing process,for this part the preparation for painting and paint-ing costs are not applied. The costs for the FRTPpart are calculated for each process, by dividingthe cost of purchase of the tool by the number ofparts that one tool can produce.

Cfinishing =Crot

Nunits+

Csand

Nunits+

CUV

Nunits+

Cpaint

Nunits(4)

3.3. Life Cycle Assessment (LCA)A life cycle assessment was made for this workcase study using the software Simapro developedby Pre Sustainability.

The goal of the study is to assess the environ-mental impact of the production of two componentswith the same function and similar characteristicsbut produced in two different materials each usinga different production process. The component isa leaf-type spring gear that is produced in FRTP(nylon and fiberglass) by FDM and is going to becompared to another one produced in Aluminumby CNC Milling.

For this study the functional unit was defined asone part (leaf-type spring gear) produced. In the

case study, the reference flow are 500 parts. Thisnumber is, as used before in the LCC, obtained byconsidering a low-scale production of 100 parts ayear for a period of 5 years.

In the case study it is used a “cradle to grave”approach, that is defined as an assessment fromthe resource extraction to the disposal phase. Thisapproach was considered rather than the cradle togate approach, because the component analyzedin the case study is part of an aircraft, thus it isimportant to consider the use phase of this com-ponent because it may have implications in the en-ergy (fuel consumption) used by the aircraft as aresult of the component final weight.

In Figure 3 and in Figure 4, are present theflowcharts for the CNC Milling process and the AMFDM process respectively.

Assumptions: The first assumption is that thesuppliers of both aluminum, fiberglass and nylonwould be within a range of 100 Kilometers of theproduction facilities, so the transport of the raw ma-terials used to produce the parts is considered tobe 100 Km. To measure the impact that the weightof the part has in the fuel consumption of the air-craft it will equip over the period of use, it is con-sidered that the part has a life of 5 years and thatthe aircraft flies 10 000 Km per year. Fiber rein-forced thermoplastics recycling is not yet a com-mon practice. Usually these materials are dis-posed into landfills. Although many studies regard-ing the recycling of FRTP have been made, cur-rently it is not economically attractive to opt for thisprocess. However, environmental legislation is be-coming more restrictive over the time, so, the re-cycling of these composites may become a realityin the future. In this case study, it is consideredthat the FRTP landing gear is disposed into a mu-nicipal landfill. The method of disposal chosen forthe Aluminum part, in this case study, is recycling.Recycling is considered not only for the aluminumlanding gear, but also for the aluminum scraps orig-inated from the CNC Milling, resulting in a 100%rate of recycling of the original material. In this pro-cess it is considered the amount of energy used inthe recycling.

4. Case Study4.1. Restrictions and requirementsThe first restriction is related to the equipmentavailable to manufacture the AM part, the dimen-sions of the part should be within the printing areaof the Markforged Mark II. With the purpose of es-tablishing a relevant comparison the chosen partshould meet some requirements, such as: Thepart should be a structural part, meaning that itshould be a part essential to the aircraft operationand should support considerable loads. This is animportant requirement because one of the FRTP

4

Figure 3: CNC flowchart

Figure 4: AM FDM flowchart

main characteristics is to have high mechanical re-sistance, so it will enable the AM part to achieveits full capabilities; The production scale should besmall, as the rate of production of current state ofthe art technology in 3D printing is low. It is impor-tant to compare two parts that have similar produc-tion scale at the present time, despite what evolu-tion will the AM technology experience; The partshould be customizable, this means it should bechosen a part whose design could be altered, toimprove the part performance, appearance or costeven if for certain parts, it is very difficult to alter thedesign because of the adjacent parts, or becauseof the already simplified designs.

4.2. Landing Gear Type and Characteristics

The part to be compared in the case study is themain component of a Leaf-type spring gear.

A Leaf-type spring gear, shown in Figure 5, is atype of landing gear that equips many smaller air-crafts such as the broadly used Cessna airplanes.This type of landing gear consists on a strut that re-ceives and dissipates the impact of the landing bytransferring the impact forces to the airframe at asmaller rate than the Fixed Landing Gear. This dis-sipation of the impact forces is obtained becausethe part flexes when the forces are applied andthen it returns to the original position. This type oflanding gear is usually made of steel or aluminum,

Figure 5: Leaf-type spring gear

and sometimes is also made of composite materi-als that are light weight, more flexible and preventcorrosion [7].

The component of the leaf-type spring gear thatis going to be studied in this work can be seen inFigure 6, pointed at with the red arrow.

Due to the restrictions of the available 3D printerdimensions, it is not possible to study a full-scalepart that could be used on a passengers airplane.So, the part was down-scaled, maintaining the pro-portions and mechanical characteristics, with thepossibility of being used in smaller aircraft like anUnmanned Aerial Vehicle (UAV), because leaf-typespring gear is commonly used in UAVs [8].

5

Figure 6: Leaf-type spring gear 2

So, after selecting the part to be analyzed in thecase study, the leaf-type spring gear belonging to asmall size UAV, it was decided that the UAV wouldhave a Maximum Take Off Weight (MTOW) of 17.5Kg. The landing gear meant to be compared isexpensive and not easy to obtain, so because ofthe difficulty to obtain an aluminum landing gearand test it, this part was designed with a 3D CADsoftware (Solidworks) and tested with a Finite Ele-ments Method Software (Siemens NX), that usu-ally obtains an approximation of the mechanicalproperties and behavior of the part formed by anisotropic metal with a relatively low error [8]. FEMis used by major established industries such asaerospace, automobile, electronics, among othersto aid design and manufacturing [8] [9].

Using as reference for a hard landing an accel-eration of 1.27g [10]. We can calculate the verticallanding force. The tire is a part of the landing sys-tem that also absorbs energy and decreases thestress inflicted on the landing gear leg. In order todecrease the complexity of the system in study, thetire effects are not considered. Usually a safety fac-tor is considered when studying an aircraft struc-ture, but because this is an unmanned aircraft andto compensate for the absence of the tire influencein the system, no safety factor is considered.

F = m[Kg] · a

[ms2

] = m · g · 1.27 = 218N (5)

Using Equation 5 we multiply the mass of theaircraft by the acceleration (a=9.81 m/s2). We aregoing to test only one side of the landing gear thatis formed by 2 symmetric legs, assuming the planetouches the ground with both wheels ate the same

time, we can divide this force by 2 and obtain avertical landing force of 109N.

4.3. Design EvolutionAs this work evolved it was perceptible that the de-sign was one of the most important points in theprocess of making a 3D printed part. This is notonly because of the changes in the original designto improve costs, performance or aesthetics, butalso the changes needed because of the machineand its software settings and capabilities.

In this sub-chapter it is going to be shown theevolution of the part design starting from the designof Figure 6, that is the design of a Cessna’s landinggear. Then, by an iterative process the aluminumdesign was obtained. And from that, the designevolved to comply with AM FRTP. This evolution ofthe design is mainly based on 3 factors: part re-quirements, 3D printing machine restrictions anddesign improvement over original part. The initialaluminum design is iteratively designed and testedto withstand the vertical load caused by the landingof an aircraft with the pretended MTOW. The workgoal is to achieve a similar part, that can replacethe original steel or aluminum part, that matchesits performance while trying to improve some char-acteristics such as weight, volume, cost.

Aluminum Design:In the aerospace industry the most predominant

aluminum alloys are 2024-T3 and 7075-T6 [11][12]. For this case study the aluminum part is con-sidered to be from 2024-T3 aluminum alloy.

The first design considered was the original de-sign (Figure 6), it was replicated and analyzedwith the FEM software. The maximum Von MisesStress obtained was 318 Mpa (MegaPascal), veryclose to the yield stress of 324 MPa. This may indi-cate that the part is operating very close to its limit.So, the part needed to be redesigned.

With the second design it was achieved a bettermaximum stress and a better displacement. But,the main gains were in the volume and mass ofthe part that were reduced.This mass and volumeimprovement is very important to the aircraft oper-ation.

Figure 7: Second Design Dimensions

6

In the second design, in Figure 7, the thicknesswas increased and the width was reduced progres-sively from the aircraft link to the wheel link. As aconsequence of this process it was impossible tomaintain the four holes configuration in the wheellink, so it was reduced to only two bigger holes.

Figure 8: Aluminum part displacement

Figure 9: Aluminum part streesses

So, with an applied load of 109N, the maximumVon Mises Stress dropped from 318 MPa to 309MPa (Figure 9), pulling this result away from the324 MPa of the yield stress. The maximum dis-placement obtained (Figure 8) is 24.30 mm similarto other results of the literature [8] [11], where dis-placement is within a range of 20mm and 30mm.

This second design meets all the requirementsestablished (Load, displacement, stresses) andalso shows improvement in the part properties, sothis is defined as the aluminum part design that willbe used as base of comparison for the AM FRTPpart.

FRTP Design:

This aluminum design was sliced with Eiger soft-ware, the Markforged software, that allows the userto determine which layer is fiber and which is ny-lon. With this software, the fiber placement, thepart layout for printing, the amount of fiber in eachlayer and the type of fiber distribution in each layercan be personalized to obtain the best configura-tion and characteristics possible for the part.

The design used for the aluminum part was notsuitable for this specific AM machine, becausethe placement of the holes was too close to theedges, and the machine software could not layfiber around the holes, resulting in the part to befragile in an essential location.

Because the Finite Element Method is not ableto analyze the FRTP part produced in the Mark-forged, the analysis has to be made by trial anderror, printing a new design and testing on the lab-oratory.

So, various designs evolved from the initial onewith the objective of matching the sustained loadand having a displacement in the same range ofvalues as the aluminum part. In the process ofdesign, besides the mechanical behavior intended,it was also always taken in consideration the costof the materials. For example, it would be easyto match the mechanical characteristics of the alu-minum part by simply adding more fiber to the ini-tial design. But, because the fiber is expensive,this increase in fiber would increase the costs andalso the weight of the part, making the part non-competitive. Therefore, the design process was fo-cused on making a functional part in a sustainableway.

The objective was to reinforce the mostly nylonpart with fiberglass in some specific locations suchas around the holes, in the edges were torque isapplied and maintaining some flexibility in the cen-ter section of the part to allow the energy dissipa-tion when the aircraft lands.

Figure 10: Final design fiber location

In Figure 11, we can see the 8th and final designbeing tested. On the left the test is starting withdisplacement zero, and on the right the part is al-ready near the maximum displacement. This partmanaged to sustain a load of 107N in the range of20 to 30 mm of displacement. This is a result veryclose to the required of 109 N.

7

Figure 11: Final design test

Table 1: Simulation parametersAluminumCNC

FRTP AM

Load Sustained (N) 109 107Displacement (mm) 24.3 25.13Part Volume (cm2) 11.86 57.29Part Mass (g) 32.96 49.06Raw material Vol. (cm2) 100.8 44.66Production time (h) 1 11.5Energy used (KWh) 1.6 2.06

5. Results & discussionThe volumes of Nylon and Fiberglass used to pro-duce the AM part are given by the Eiger software.Then the mass is calculated using the density ofeach material given by Markforged, 1.1 g/cm3 forNylon and 1.5 g/cm3 for Fiberglass. In table 1 it ispossible to see that the volume of the raw materialis smaller than the volume of the part, this occursbecause the FRTP part, unlike the aluminum one,is not fully filled with material, resulting in some freespaces inside the part and consequently an highervolume.

The aluminum part has a raw material volumealmost ten times higher than the final part volume,this is because, in CNC Milling the production ofthe part starts with a solid block of material fromwhich it is removed material to obtain the final form.

From table 1 it is possible to see that the partsbehave similar in terms of loads sustained and dis-placement. The part volume and mass is higherfor the FRTP part, this may result in a worst perfor-mance when flying, considering that the weight is avery important factor in fuel consumption and alsothe bigger volume can cause aerodynamic prob-lems. From the table 1 we can also see that theproduction time for the FRTP part is much higherthan the aluminum part, forcing this type of parts toonly be considered to small production batches.

In 2, all the costs of material acquisition, ma-

Table 2: CostsCosts(e/unit) Aluminum

2024 - CNCNylon +Fiberglass -FDM

Material 11.6 23.20Machine 15.26 23.62Energy 0.16 0.21Labor 10 8.60Finishing 1.76 0.1Total 38.78 55.73

chines acquisition and production of both parts aresummarized, and the total cost is calculated. Is im-portant to refer that these costs only consider theproduction phase. The costs of the use phase (air-craft fuel consumption) and the costs of recyclingare not considered in this analysis.

From table 2 it is possible to see that despitethe big amount of aluminum waste the materialcost is still higher for the FRTP part. The machinecosts are also significantly higher for the FRTP partmainly because of the time needed to produce onepart being higher. Summing all parcels the result isthat the aluminum part is 16.95 e cheaper to pro-duce.This result demonstrates that, for this specificcomponent produced with these materials, CNCMachining is the less expensive technology of thetwo. To calculate the machine costs of the FRTPpart it was considered that the machine was non-dedicated and that the hours of work per day were8. However, the costs per unit drop considerableby increasing the work hours, for example from 8h to 12h the cost decreases from 55.72 e/unit to47.85e/unit. From 12 hours on, the costs continuedecreasing achieving a value of 39.98 e/unit if themachine worked permanently, 24 hours a day.

Midpoint AnalysisIn the graphic of Figure 12, it is present a nor-

malization of the midpoint categories results. It ispossible to understand that the AM FDM procedurehave a lower impact on the environment than theCNC Milling in every single category.

Figure 12: LCA Midpoint Analysis

Endpoint Analysis

8

In Figure 13 is present a single score impactassessment that compares the production of theleaf-type spring gear by CNC Milling and AM FDM.The results give the cumulative environmental im-pact for the two different processes, consideringthat each uses a different material, aluminum andFRTP. These results include three categories: hu-man health, ecosystems and resources. From thegraphic it is perceptible that, for this case study,CNC Milling has a much higher cumulative impactthan AM FDM.

Figure 13: LCA Endpoint Analysis

From the graphics presented before, it is possi-ble to extract as a result that this component, theleaf-type spring gear of an UAV, produced in alu-minum by CNC milling has a significantly higherenvironmental impact than the same componentproduced in FRTP by FDM. One factor that con-tributes largely for this result is the fact that the ra-tio between the amount of raw material needed toproduce the component and the amount of materialpresent in the component after production, is muchhigher for CNC milling. In practical terms this trans-late into more material that has to be extracted,more material that has to be transported to the pro-duction site and more waste material that will be re-cycled. Considering the use phase, that is the influ-ence the weight of the part has on its end use, theflight of the aircraft, the aluminum part has a lowerweight, so less impact on the fuel consumption andthe pollution. Despite the use phase, and the factthat the FRTP part is considered to be disposed of,compared to the aluminum part and waste materialthat are recycled, the FRTP still have a significantlylower impact on the environment.

6. ConclusionsFrom the results it is not possible to determinewhich technology performed better in this casestudy. On one hand, the component produced inaluminum by CNC Milling is better comparing themass and volume of the part and has significantlylower production costs. On the other hand, theenvironmental impact is remarkably lower for theFRTP part produced by FDM. Despite the contrast-

ing results it is possible to make some conclusions.In first place, it is important to emphasize that CNCmilling is a technology that is already well devel-oped and has its place in the industry for a longtime, while Fused Deposition Modeling, with fiberreinforced thermoplastics, is still a relatively newtechnology with a large amount of improvementsto be made, so this big gap that exists, in termsof production costs, can be reduced. The mainparcels of the cost where FDM is more expensiveis in the material acquisition and machine costs,and both of these costs can be reduced if AM getsmore popular and starts to gain market share. Thiswill increase the number of machines available inthe market, the number of suppliers of materialsand consequently dropping the prices of the equip-ment and the materials used. In terms of the partvolume and mass, the FRTP part sustained thesame loads as the aluminum part, displaying thesame range of displacement. The only drawbackwas that the FRTP part, presented a higher massand volume. This point could be improved by twoways: changing the materials used or improvingthe design. In terms of materials, with this partic-ular machine, the Markforged Mark II, one way ofmaintaining the same behavior of the part and re-ducing the mass and volume could be by using car-bon fiber instead of fiberglass. The use of this fiberwould increase the material cost, so, it may not bethe best solution as the FRTP part is already moreexpensive than the aluminum part. In terms of de-sign, with the right expertise and resources, a bet-ter design could be generated in order to better ex-ploit the capabilities of the FRTP and AM. Since thedesign topic was brought up, design flexibility wasone of the general advantages of AM technology,but it is important to mention that for this particularproduction method and machine (FDM and Mark-forged) the design has large number of restrictions.As an example of some restrictions are the printerdimensions; the printer software, that has limitedways of placing the fibers, since each layer has tobe of only one material, sometimes fiber is neededin one section of the part but is not in another sec-tion and both are in the same layer, leading some-times to the use of more fiber than needed; and thevalidation of results, because nowadays it is com-mon in industry to develop a product and evolv-ing its design before production, resorting to FEMsoftware or other numerical methods, but for FRTPdeposited by Markforfed way there are not yet nu-merical methods able to predict the behavior of thepart, so the process of design has to be made byiterations, producing and testing prototypes. Re-garding the production size or scale, it is assumedfrom the start that AM is only viable for a smallnumber of parts, and in this case study, it is shown

9

that even if the production scale is increased thecost of the part is higher for the FRTP part. Themain element holding back the production size isthe production time of the part, that is much higherthan CNC machining, but with the improvement ofthe FDM machines it may be possible to increasethe production size and consequently reduce theproduction costs. Environmentally, the more effi-cient of both processes, with a big difference, isFDM. This result is largely because FDM, beingan additive process, only consumes the amountof material that forms the final part, while CNCmilling requires a block of material from where thepart is sculpted. So, even though FRTP part hasmore mass and more volume than the aluminumpart, the aluminum part, in total, uses much morematerial causing a higher impact on the environ-ment. Environmental awareness is becoming asignificant factor for companies, governments andthe public in general when making decisions. Be-sides the increasingly restrictive laws on this sub-ject, a fair amount of companies also want to trans-mit to the public the image that they care aboutthe environment and their products are sustain-able. This factor amplifies the value of the environ-mental results of the case study, even though thecosts were almost always the more important fac-tor in product development. Furthermore, with thedevelopment of the technology around FDM withFRTP almost all the causes for the inflated produc-tion costs, such as production times or machinesoftware restrictions, can have their impact dimin-ished. In conclusion, parts of Nylon reinforced withFiberglass produced with Fused Deposition Mod-eling are not yet an alternative to aluminum partsmade by CNC milling, mainly because of the costsinvolved and the software and machine limitationsthat do not allow for the design to be improved indetail for AM. However, mainly because of its lowerenvironmental impact and the room for improve-ment, it is expected for it to be an alternative overtraditional production methods in the near future.

References[1] M. Gebler et al. A global sustainability per-

spective on 3D printing technologies. EnergyPolicy. 2014.

[2] ASTM International. Standard terminology foradditive manufacturing technologies.

[3] S. Ford et al. Additive manufacturing and sus-tainability: an exploratory study of the advan-tages and challenges. Journal of Cleaner Pro-duction, 2016.

[4] M.K. Thompson et al. Design for AdditiveManufacturing: Trends, opportunities, consid-erations, and constraints. 2016.

[5] W. Gao et al. The status, challenges, and fu-ture of additive manufacturing in engineering.2015.

[6] P. Pecas et al. Additive manufacturing ofcomposites - design, manufacturing and re-sources efficiency challenges. 2017.

[7] Federal Aviation Administration U.S. Depart-ment of Transportation. Aviation MaintenanceTechnician Handbook-Airframe. InternationalJournal for Numerical Methods in Engineer-ing, 2012.

[8] Z. Xue et al. A Simplified Flexible MultibodyDynamics for a Main Landing Gear with Flexi-ble Leaf Spring. Hindawi Publishing Corpora-tion, Shock and Vibration, 2014.

[9] J.N. Reddy. An introduction to the Finite Ele-ment Method. McGraw Hill, 2006.

[10] A. Pereira et al. Relatorio Final de Acidente,Ministerio das obras publicas, transportes ecomunicacoes, Relatorio n.33,ACCID,2009.2009.

[11] M. Imran et al. Static and Dynamic ResponseAnalysis for Landing Gear of Test Air Crafts.International Journal of Innovative Researchin Science, Engineering and Technology, 3(5),2014.

[12] A.K. Matta et al. Design Optimization of Land-ing gear’s Leg for an Un-Manned Aerial Vehi-cle. International Journal of Engineering Re-search and Applications (IJERA), 2(4):2069–2075, 2012.

10