School of Engineering Department of Mechanical & Manufacturing Engineering

Developing Capacity to 3D Print Polyether-ether-ketone Using Fused Filament Fabrication

Michael McHugh 10706279

March 2016

A dissertation submitted in partial fulfilment of the degree of MAI (Engineering with Management)

Michael McHugh 10706279

Declaration I have read and I understand the plagiarism provisions in the General Regulations of the University Calendar for the current year, found at http://www.tcd.ie/calendar. I have also completed the Online Tutorial on avoiding plagiarism ‘Ready Steady Write’, located at http://tcd-ie.libguides.com/plagiarism/ready-steady-write.

___________________

i

Michael McHugh 10706279

Abstract In this project a capacity to print Polyether-ether-ketone (PEEK) using 3D Fused Filament Fabrication (FFF) was researched, developed the resulting findings were analysed and discussed. Background research was conducted into the existing scientific literature and into the platform to be adapted for the project, an Ultimaker 2 FFF 3D printer. From the scientific literature an overview of Additive Manufacturing, Fused Filament Fabrication, PEEK material properties and uses related to FFF, and results and observations from a number of key papers in which researchers had manufactured PEEK using FFF were recorded. The research established the main criteria for printing PEEK using a fused filament fabrication process to be a hot end temperature of 410°C, a build plate temperature of 130°C and an ambient temperature of 80°C. (1) The system used as the platform for this project, an Ultimaker 2 FFF 3D printer, was characterised in detail from a systems and process point of view, and a methodology to adapt the machine to print PEEK according to identified criteria was established and carried out successfully. Investigations adhering to ISO 527 were then undertaken using the developed system and the specimens produced were found to have a mean ultimate tensile strength of 91 MPa in the X-Y build axis, comparing well with the results found in the literature and in the technical data. Future works propose further research into the material properties and uses of the FFF PEEK, exploration of the use of the system for printing and researching other high temperature extrusion polymers, and works toward refinement and development of the system and process used.

ii

Michael McHugh 10706279

Acknowledgements

Many thanks to my Supervisor, Garret O'Donnell for his guidance, advice, expertise, patience and understanding throughout this project.

Also a special thanks to Professor Kevin Kelly, the director of the Engineering with Management programme without whom I would certainly not be here today, thank you Kevin for your support and enthusiasm.

Thanks to Michael Cullinan for his know-how, advice and optimism.

Thank you to all in the Parsons Building for making it a great place to study and work.

Thank you to my family for their incredible support through the years.

And thank you for everything Maria, I love you.

iii

Michael McHugh 10706279

Contents Declaration .................................................................................................................................. i Abstract ...................................................................................................................................... ii Acknowledgements................................................................................................................... iii Contents .................................................................................................................................... iv List of Tables ............................................................................................................................ vii List of Figures ............................................................................................................................. 1 1

Introduction ................................................................................................................... 1 1.1

1.1.1

Additive Manufacturing ................................................................................. 1

1.1.2

Fused Filament Fabrication ............................................................................ 1

1.1.3

Polyether-Ether-Ketone ................................................................................. 2

1.1.4

3D Printing PEEK using FFF............................................................................. 2

1.2 2

Overview ................................................................................................................ 1

Objectives............................................................................................................... 3

Background and System Analysis................................................................................... 4 2.1

Literature Review ................................................................................................... 4

2.1.1

Additive Manufacturing ................................................................................. 4

2.1.2

Liquid based AM Processes ............................................................................ 4

2.1.3

Solid Based AM Processes.............................................................................. 6

2.1.4

Powder Based AM Processes ......................................................................... 6

2.1.5

PEEK and FFF .................................................................................................. 9

2.2

FFF PEEK State Of Art ........................................................................................... 10

2.3

System Characterisation ...................................................................................... 11

2.3.1

Ultimaker 2 Overview .................................................................................. 11

2.3.2

Cura GUI ....................................................................................................... 13

2.3.3

Firmware ...................................................................................................... 15

2.3.4

G-code .......................................................................................................... 16

2.3.5

UM2 Control System Schematic .................................................................. 18 iv

Michael McHugh 10706279

3

2.3.6

UM2 Main Board Integrated Circuits Schematic ......................................... 19

2.3.7

UM2 Main Board Connected Components Schematic ................................ 20

2.3.8

UM2 Main Board Expansion Availabilty Schematic ..................................... 21

2.3.9

PEEK Filament .............................................................................................. 22

System and Process Development............................................................................... 23 3.1

Constraints ........................................................................................................... 23

3.2

Methodology........................................................................................................ 23

3.3

Aim 1: Hot End Adaption ..................................................................................... 25

3.3.1

Task 1: Software Maximum Temperature Limit .......................................... 25

3.3.2

Task 2: Thermocouple Installation ............................................................... 26

3.3.3

Task 3: Software Temperature Rise Rate Limit ............................................ 28

3.3.4

Task 4: Heater Cartridge and Hot End Upgrade .......................................... 28

3.3.5

Task 5: Calibration........................................................................................ 29

3.4

3.4.1

Task 1: Firmware Maximum Temperature Limit ......................................... 31

3.4.2

Task 2: Hardware Limit Testing .................................................................... 31

3.4.3

Task 3: Calibration........................................................................................ 32

3.5

4

Aim 2: Build Plate Adaption to 130°C .................................................................. 31

Aim 3: Filament Feeder Adaption to 1.75 mm .................................................... 33

3.5.1

Task 1: Installation of 1.75 mm Bowden tube ............................................. 33

3.5.2

Task 2: Filament Feeder Modification ......................................................... 33

3.5.3

Task 3: Firmware Power Increase to the Feeder Stepper Motor ................ 34

3.6

Aim 4: Ambient Temperature Adaption to 80°C ................................................. 35

3.7

PEEK FFF Printer Testing ...................................................................................... 37

3.7.1

Methodology................................................................................................ 37

3.7.2

Development Testing Methodology ............................................................ 37

3.7.3

Tensile Testing Methodology ....................................................................... 38

Results and Discussion ................................................................................................. 41 4.1

PEEK FFF Printer Development Testing ............................................................... 41 v

Michael McHugh 10706279 4.1.1

Carabiner test print 1 Points of Interest: ..................................................... 41

4.1.2

Carabiner test print 2 Points of Interest: ..................................................... 42

4.1.3

Carabiner test print 3 and 4 Points of Interest: ........................................... 43

4.1.4

Carabiner test print 5 Points of Interest: ..................................................... 44

4.1.5

Carabiner test print 6 Points of Interest: ..................................................... 44

4.1.6

Carabiner test print 7 Points of Interest: ..................................................... 45

4.2

4.2.1

Series 1 Results: 100% Infill PEEK Specimens .............................................. 46

4.2.2

Series 2 Results: 20% Infill PEEK Specimens, Build Plate 130°C .................. 47

4.2.3

Series 3 Results: 20% Infill PEEK Specimens, Build Plate 150°C .................. 48

4.3

5

PEEK FFF Printer Tensile Testing Results ............................................................. 46

PEEK FFF Tensile Testing Discussion .................................................................... 49

4.3.1

Series 1 Discussion: 100% Infill FFF PEEK Specimens .................................. 49

4.3.2

Series 2 & 3 Discussion: Build Plate 130°C Versus 150°C ............................ 51

Summary and Conclusion ............................................................................................ 52 5.1.1

Summary ...................................................................................................... 52

5.1.2

Conclusion .................................................................................................... 53

5.1.3

Future Works ............................................................................................... 53

References ............................................................................................................................... 54

vi

Michael McHugh 10706279

List of Tables Table 1 Ultimaker 2 Specifications .......................................................................................... 12 Table 2 Victrex 450 G PEEK Material Properties...................................................................... 22 Table 3 Ultimaker2 to PEEK Conversion Constraints ............................................................... 23 Table 4 PEEK FFF Test Print Settings ........................................................................................ 37 Table 5 Tensile Test Series Parameters ................................................................................... 38 Table 7 Tensile Test 1 Results 100% Infill PEEK Specimens ..................................................... 46 Table 9 Tensile Test 2 Results 20% Infill PEEK Specimens, Build Plate 130°C ......................... 47 Table 10 Tensile Test 3 Results 20% Infill PEEK Specimens, Build Plate 150°C ....................... 48 Table 11 Combined Values for Tensile Test Results on 100% Infill PEEK Specimens .............. 49

vii

Michael McHugh 10706279

List of Figures Figure 1 AM Processes Overview .............................................................................................. 4 Figure 2 Fused Filament Fabrication .......................................................................................... 5 Figure 3 Stereolithography process .......................................................................................... 6 Figure 4 Selective Laser Sintering Process ................................................................................. 7 Figure 5 Laser Engineered Net Shaping Process ........................................................................ 8 Figure 6 Indmatec 3D Printed PEEK Comparison ..................................................................... 10 Figure 7 Ultimaker 2 front view ............................................................................................... 11 Figure 8 Ultimaker 2 back view ................................................................................................ 12 Figure 9 Screen capture of Arduino IDE showing UltimakerMarlin files ................................. 16 Figure 10 Ultimaker 2 Control System Schematic .................................................................. 18 Figure 11 UM2 Mainboard IC................................................................................................... 19 Figure 12 UM2 Mainboard used ports..................................................................................... 20 Figure 13 UM2 Mainboard Unused ports ................................................................................ 21 Figure 14 Main Goal and Aims ................................................................................................. 24 Figure 15 Software/Hardware Task Approach ........................................................................ 24 Figure 16 Tasks within Hot End Conversion Aim ..................................................................... 25 Figure 17 Installation of Thermocouple into Main Board ....................................................... 26 Figure 18 UM2 Hot End Adaption ............................................................................................ 29 Figure 19 Tasks within Build Plate Conversion Aim ................................................................. 31 Figure 20 Tasks within Filament Feeder Conversion Aim ........................................................ 33 Figure 21 Feeder Mechanism Adaption ................................................................................... 34 Figure 22 Tasks within Ambient Temperature Conversion Aim .............................................. 35 Figure 23 UM2 with Dual Enclosures and Heating Apparatus................................................. 36 Figure 24 ISO 527 Test Specimen Requirements .................................................................... 39 1

Michael McHugh 10706279 Figure 25 FFF PEEK Test Specimen........................................................................................... 40 Figure 26 Carabiner test prints ............................................................................................... 41 Figure 27 Carabiner Test Print 1 .............................................................................................. 42 Figure 28 Carabiner Test Print 2 .............................................................................................. 42 Figure 29 Carabiner Test Print 4 .............................................................................................. 43 Figure 30 Carabiner Test Print 6 detail .................................................................................... 44 Figure 31 Carabiner Test Print 7 .............................................................................................. 45 Figure 32 Tensile Test Specimens, 100% Infill, Load versus Extension. .................................. 46 Figure 33 Tensile Test Specimens, 130°C Build Plate, Load versus Extension. ........................ 47 Figure 35 Tensile Test Specimens, 150°C Build Plate, Load versus Extension......................... 48 Figure 36 Specimen 4 100% infill post tensile testing ............................................................. 50

2

Michael McHugh 10706279

1 Introduction 1.1 Overview 1.1.1 Additive Manufacturing Additive Manufacturing (AM) is of increasing importance, particularly in biomedical engineering research, and the wider engineering research fields. It has the ability to reduce or remove many limits and obstacles for production and testing of parts as it allows for inexpensive, flexible, on demand high resolution rapid prototyping and rapid manufacturing of custom made parts and biomedical devices. It compares favourably with time, cost and locational dependencies associated with the traditional supply chain of items. (2–9) AM techniques use computer aided design (CAD) technologies to construct a 3d model in software and then encode the model for fabrication in the chosen material by a computer numerically controlled (CNC) machine. Rapid Prototyping (RP) was the first form of AM, used in pre-production to create models for visualisation and parts to be used in prototypes in the 1980's. RP allowed for models that more closely resembled the final parts for evaluation to be produced quickly and at reduced costs, which could help greatly lower the resources needed during the product development cycle. (9–14) Another key advantage of AM is that it enables the manufacture of parts that are difficult or expensive to machine using standard manufacturing methods. Complexity of manufacture is not a cost in AM, and many differing versions of designs can be quickly designed and altered using CAD techniques and rapidly manufactured for test and evaluation purposes, enabling more thoroughly exhaustive research and design processes combined with lower times and costs. (4,7,9,12,15,16)

1.1.2 Fused Filament Fabrication Of the current AM technologies the one that most combines ease of use with low costs is the Fused Filament Fabrication (FFF) method. The technology is available at a consumer level with units that are low cost to purchase and run, easy and safe to use and maintain, and can generate items of increasingly high quality. (4,7,9,10,12,17,18) FFF offers methods to introduce structural properties to modify and control specific functionalities such as porosity that allow for increased biomedical compatibility and integration. (1,3,5,8,19,20)

1

Michael McHugh 10706279 1.1.3 Polyether-Ether-Ketone Polyether-ether-ketone (PEEK) has mechanical , biocompatibility and radiological qualities that have encouraged its wide use in load bearing orthopaedic implants and it is a material of growing interest to the biomedical research field. It is a semi-crystalline polymer with excellent strength, stiffness and toughness. PEEK has been found to be have desirable qualities for biomedical uses compared with a range of currently used metallic alloys, including greater energy absorption efficiency and a higher strength to weight ratio. (1,21– 23)

1.1.4 3D Printing PEEK using FFF Increasing the porosity of PEEK for osseointegration is one of the main areas of research into the use of PEEK for orthopaedic implant use, and 3D printing is a technology that is suited to introducing porosity to a material quickly, easily, and at low cost, making it ideal for prototyping in biomedical research environments. (1,6,22,24,25) FFF currently does not easily or inexpensively allow for the printing of some more novel polymers used in the biomedical field such as PEEK, mainly due to difficulties producing FFF machines that can operate successfully at temperatures capable of printing PEEK, which has a melting point of 343°C and a flow point of 400°C. (1,7,9,17,18,26,27)

2

Michael McHugh 10706279

1.2 Objectives The goal of this research project is to develop the capacity to print Polyether-ether-ketone using 3D Fused Filament Fabrication. An Ultimaker 2 FFF 3D printer was used as the platform for this project, and 1.75mm PEEK filament from Indmatec GMBH was the material source.

The objectives are: 1) Characterise the current Fused Filament Fabrication machine capacity and identify constraints for printing PEEK according to the existing scientific literature. 2) Develop a hardware and software adaption strategy to enable printing of PEEK on the Ultimaker 2 3) Undertake 3D printing investigations using the developed system and analyse results, comparing with existing scientific literature.

In order to develop the capacity to print Polyether-ether-ketone using 3D Fused Filament Fabrication a understanding of PEEK 3D printing and desired characteristics and methodologies will be derived from the existing scientific literature. A FFF 3D printer then will be characterised in detail from a systems and process points of view, and the primary aspects that constrain it regarding 3D printing with PEEK are to be identified. From this an approach to adapting the 3D printer for PEEK regards suitable hardware and software changes will be developed, implemented and documented. The results of the adaption will then be tested and evaluated by analysis that refers to the existing scientific literature. Conclusions and possible future works will then be put forward.

3

Michael McHugh 10706279

2 Background and System Analysis 2.1 Literature Review 2.1.1 Additive Manufacturing Additive Manufacturing (AM) is a term used to describe technologies that produce three dimensional (3D) objects by adding layers of a material at a time to construct the finished object. Materials range from plastics and metals to concrete and tissue. (9,12,18,26,28) Over time the number of AM technologies has grown, but they can be classified under three main headings: Liquid, Solid or Powder based as shown in the figure below.

Figure 1 AM Processes Overview adapted from (9)

2.1.2 Liquid based AM Processes Fused Filament Fabrication (FFF) melts a plastic filament and extrudes it layer by layer to build up the desired 3D object. The print head extrudes in the X-Y axis, whilst the build plate drops down in the Z-axis upon the completion of a layer. This method of printing is 4

Michael McHugh 10706279 becoming widespread in the consumer/hobbyist field as it is inexpensive, safe for home use, the pieces require no post processing and it is simple to use and maintain. Commonly used plastics are Polyactic Acid (PLA) or Acrylonitrile Butadiene Styrene (ABS) which have low melting temperatures, are inexpensive and produce acceptable results for general use. (9,12,13,17)

Figure 2 Fused Filament Fabrication(26)

AM polymerisation techniques include stereolithography and polyjet technologies. In these processes photosensitive polymer resins are laid down and cured into solid form by exposure to ultra violet (UV) light. Stereolithography was the first widely used RP technology, and is still widely used today. A bath of the liquid polymer has a scanner deflecting a UV range laser beam to precise points upon the surface of the liquid above a platform which initially sits just below the surface. Each pass of the laser that generates a layer of the object causes the platform to drop down one layer deeper into the bath, enabling the next layer to be built on top of the preceding layer until the object is completed. (7,9,12,26)

5

Michael McHugh 10706279

Figure 3 Stereolithography process (19)

Polyjet AM technology relies on inkjet technology to deposit a thermo-photopolymer in the X and Y axis, building a single layer, which is then cured by UV lamps. The model under construction is then moved a one layer distance further away from the Ink Jets along the Z axis so that the next layer can be deposited and cured in place. (9,12,26,28)

2.1.3 Solid Based AM Processes Laminated Object Manufacturing is a combination of AM and subtractive manufacturing. Sheets of a material are bonded together one at a time and between each bonding the newest layer is cut to the shape desired for that layer, until the object is built. It can be used for thermal bonding using plastics and metals, or adhesive bonding using paper, which can also contain dyes for detailed and accurately coloured prototyping. This method is inexpensive and can create, depending on the method used, low deformation, high colour fidelity products, but is wasteful of material and has difficulty with internal cavities as they are built with material in place which must be removed after manufacture. (9,12,28,29)

2.1.4 Powder Based AM Processes Powder based AM uses either melting or binding processes. Selective Laser Sintering (SLS) uses Carbon Dioxide lasers to melt a layer of powder that has been rolled one layer thick across a bed. The powder sinters where melted, and then the bed drops one layer and another layer of powder is rolled over for the next layer to be sintered. Because the object is being lowered with powder about the sintered parts there is good support provided for 6

Michael McHugh 10706279 building otherwise troublesome geometries. Thermoplastics, metals and ceramics can be printed using this technology. (9,12,27)

Figure 4 Selective Laser Sintering Process(26)

Electron Beam Melting (EBM) is similar in process to SLS but utilises an electron laser instead of a Carbon Dioxide one, and is situated in a vacuum chamber to prevent oxidisation as it usually builds with metals. (7,9,12) Laser Engineered Net Shaping (LENS) positions and melts injected metal powder using a laser that then solidifies into the desired shape. This technique can be used for a wide variety of metals, and can be used for repair work, for example on turbine blades, but can leave undesirable thermal stresses. (7,9,12)

7

Michael McHugh 10706279

Figure 5 Laser Engineered Net Shaping Process(9)

3DP is a water based binder and starch based powder process in a similar manner to SLS, with the binder being placed onto the powder instead of a laser. This method can bind a variety of polymers. (9) Prometal operates in a similar basis to 3DP but the powder is metal based, for example steel, and the piece needs to be post processed with to cure the binder (low heat), and infiltration (high heat), to infuse another material such as bronze (resulting in a typical 60/40 steel/bronze ration) into the porous specimen. This method can produce parts with more desirable properties than CNC machined items. (9)

8

Michael McHugh 10706279 2.1.5 PEEK and FFF Fused filament fabrication of PEEK is of interest not just because it allows for rapid prototyping of parts, but also because it is a method of manufacture that can introduce desirable qualities to the material. For example, in current orthopaedic technology, metallic implants have high strength but because of radiological disadvantages can cause difficulties when performing medical imaging on the subject. Polymers would remove this radiological complication but many lack the strength of the metallic implants. PEEK is a polymer of interest because it has high strength, and a stiffness that resembles that of bone which aids in stress shielding between the implant and the implanted bone matter. However the poor osseointegration of PEEK due to its non-porous finish using traditional manufacturing techniques has hindered its use. FFF can add porosity to PEEK, improving its osseointegration. PEEK is a semi crystalline polymer and the degree of crystallinity changes according to its rate of cooling from a liquid state; a slower cooling rate introduces more crystallinity. (6,21,22,24,30,30–33) In their paper titled "Extrusion Based Additive Manufacturing of PEEK for Biomedical Applications" Vaezi and Yang use Fused Filament Fabrication to produce PEEK structures. They identify key factors for successfully printing PEEK without warping and delamination of the prints, and for avoiding polymer degradation. A recommended hot end temperature of 410°C, a heated bed of 130°C and an ambient temperature of 80°C are identified as optimal settings for printing PEEK. (1) The ambient temperature was provided by heat lamps focussed on the printing area. They found a reduction in ultimate tensile strength from 113 MPa in injection moulded stock to 75.06 in their 100% infill, and therefore expected to be solid, samples, which they attributed to an induced minimum porosity of 14% within the printed specimens due to air gaps caused by the infill pattern and entrapped micro-bubbles inside filaments. (1) They state that heat management during the printing process is an important parameter for adhesion to the build plate and layer to layer PEEK bonding and that environmental heat distribution about the part is important for the level of crystallinity produced in the printed PEEK. (1) Wu et al produce 100% infill FFF PEEK specimens with a tensile strength of 56.6 MPa compared to the 100 MPa tensile strength of the PEEK material before printing. They accredit most of the loss of tensile strength to induced porosity from air gaps introduced during the laying down to the filament layers; they conclude that the mechanical properties of 3D printed PEEK may be improved if refinements to the hardware and software were carried out. They state that future research is needed to find methods and approach to reduce unwanted porosity formation during the printing process and to improve interlayer bonding.

9

Michael McHugh 10706279 PEEK recommended printing temperature is 410°C, 67°C above PEEK melting temperature. This been identified as the point at which the material extrudes a flow suitable for FFF. The other factors relating to printing PEEK using FFF are the temperatures present as it is extruded; the ambient temperature about the extrusion, and the temperature of the build plate onto which the print is built. PEEK FFF printing recommended temperatures are 410° for the hot end, where the filament is extruded, 130°C for the build plate and 80°C for the ambient environment. Consumer FFF printers typically have maximum temperatures of 260°C for the hot end, 100°C for the build plate and the ambient temperature is at room temperature levels. (1,34)

2.2 FFF PEEK State Of Art Indmatec GMBH offer a FFF PEEK printer for sale in the region of €15,000 with an enclosed build chamber, heated build plate and hot end that can reach temperatures of 420°C to enable PEEK to be successfully printed. As of writing this is the only commercially available FFF PEEK printer on the market. The manufacturer claims material properties approaching powder injected moulding methods, an improvement on SLS manufactured PEEK. Indmatec also produce 1.75 mm diameter PEEK filament for use in compatible FFF printers. (35)

Figure 6 Indmatec 3D Printed PEEK Comparison(35)

10

Michael McHugh 10706279

2.3 System Characterisation 2.3.1 Ultimaker 2 Overview The system to be adapted is an Ultimaker 2 Fused Filament Fabrication 3D printer 2. The UM2 printer adapted for this project had been in use as a rapid prototyping tool prior to its assignment to PEEK printing development. (36) The standard UM2 has a maximum hot end temperature of 260°C, the temperature of the heated bed can reach a maximum of 100°C and the ambient temperature within the build area is usually a few degrees higher than the ambient room temperature, depending on printing temperature settings and air currents in the surrounding environment. (37) This model was manufactured by Ultimaker from 2013 to 2016 and cost in the region of €2000 at purchase. Its standard layout and technical specifications are detailed below. (36)

1. 2. 3. 4. 5. 6. 7. 8.

Figure 7 Ultimaker 2 front view

11

Print Head Z-axis spiral screw Build Plate SD-card slot Bowden tube Print Head Cables OLED display Input dial/button

Michael McHugh 10706279

1. 2. 3. 4. 5. 6. 7. 8.

Print Head cables Filament Spool Filament Power Input Bowden tube Filament Feeder USB Input On/Off Switch

Figure 8 Ultimaker 2 back view

Table 1 Ultimaker 2 Specifications (36)

Ultimaker 2 Specifications Layer resolution: Build volume: Position precision X Y Z: Print speed: Travel speed: Filament diameter: Nozzle diameter: Stand-alone WiFi printing Software: Print technology: Ambient Operating Temperature: Frame dimension X Y Z: Operation nozzle temperature: Operation heated bed temperature: AC input: Power requirements:

up to 20 µm 23 cm × 22.5 cm × 20.5 cm 12.5 µm × 12.5 µm × 5 µm 30–300 mm/s 30–350 mm/s 2.85 mm 0.4 mm SD-card ready (future upgradeable) Cura - Official Ultimaker Fused filament fabrication (FFF) 15 to 32°C 49.2 cm × 34.2 cm × 55.8 cm (with filament) 180-260 °C 50-100 °C 100-240 V/~4 A/50–60 Hz/221 watt max. 24 V DC @ 9.2 A

The Ultimaker 2 uses Fused Filament Fabrication (FFF) to build up 3D models from software files. Models can be built or generated many different ways, by using CAD (Computer Aided Design) programs such as Solidworks or Creo, or objects can be scanned using a 3D scanner. 12

Michael McHugh 10706279 There are also online communities with large number of pre-generated models, for example Youmagine and Thingiverse. (38–41) Generally, any 3D design software can be used, once it can export STL, OBJ, DAE or AMF files. These are the formats used in the UM2 conversion software, Cura. (42,43) Printing with the Ultimaker 2 requires some calibration and appropriate settings to be chosen to ensure good quality prints. Good print adhesion is essential to prevent warping during printing. Print bed adhesion is helped greatly if the build plate is properly calibrated so that it is an even and measured distance, 0.1 mm, from the nozzle at any point during printing of the first layer. The printer allows for simple calibration process using a sheet of paper to gauge distance. (44) A brim of one layer in thickness can be added about the print on the first layer, usually about 10 mm about the perimeter of the object, to add surface area in contact with the build plate and help improve adhesion. (37) It is important to choose a width of filament line laid down that is proportional to the nozzle size. Wall thicknesses should be a multiple of the nozzle diameter or they will not print cohesively. (45) Layer thickness is also important; a thinner layer produces better quality prints but is slower to produce a print. Too high a layer thickness will be beyond the ability of the nozzle to fill between layers, producing poor cohesion. Any specific layer heights or thickness chosen, such as the top or bottom parts of a print, should be a multiple of layer height. (45) A general rule is that quality is inversely proportional to speed in FFF, so that choices related to layer height, wall thickness, degree of infill and speed of print head/extrusion flow rate are all affected by this. (45)

2.3.2 Cura GUI Once the CAD model has been obtained it can be prepared and converted for printing on the Ultimaker 2 using Cura. This provides a 3D environment within which the model can be orientated and the print quality and other settings chosen. Cura uses a graphical user interface (GUI) to allow the user to prepare and convert the CAD model so that it is suitable for 3D printing. The figure below shows the main interaction screen of Cura.

13

Michael McHugh 10706279

Figure Cura Main Screen

where: 1. 2. 3. 4. 5.

Menu bar: access to preferences, machine settings and profiles. Quick print profiles: default print profiles of differing quality. Support structure: offers the option to add support to the print Load button: click to locate and load a 3D model into Cura. Save button: Save 3D print files to the computer or directly onto the SD card (if inserted). 6. YouMagine button: use this button to directly share 3D files to the online community on YouMagine.com. 7. View modes: use different view modes to check the printability of your model and to view the print path. 8. Rotate: rotate the model in the X, Y or Z direction. 9. Scale: change the (outer) dimensions of the 3D model. 10. Mirror: Mirror the 3D model in X, Y or Z direction. 11. Visualization: A 3d representation of the platform of the UM2 with the loaded 3D model(s).

Cura also offers several ways of adjusting the model before printing it, for example change to change position and scale of the model. It enables placement of the model into a suitable 14

Michael McHugh 10706279 orientation for printing. Additions or alterations can be made; for example support structures can be added or rescaling of the model can be done. Options are available for changing a wide range of parameters related to the printing such as speed of travel of the print head, or the percentage infill of the model. (37) After the settings have been chosen Cura converts the file into G-code, which can then be saved onto an SD card and the file can now be printed. (37)

2.3.3 Firmware The firmware used in the UM2 is based on open source software for 3D printers called Marlin, and runs on a logic board within the UM2 that simulates an Arduino 2560 and can be programmed as such using the Arduino integrated development environment (IDE) software. Ultimaker provide access to the firmware and all other software and documentation used in their products in a repository on Github, a web based source code management system. A specific fork of the Marlin software is adapted and stored in this depository specifically for the UM2. (46–48) The UM2 firmware source code is constantly updated and revised and is included as part of Cura; as of writing the current official release firmware is version 15.04.4. Previous versions of the firmware are also available. (43,49) The repository can be downloaded and then opened in the Arduino Integrated Development Environment (IDE). This IDE allows both alteration of the UM2 Marlin code, and uploading of that code to the main board. (46) The code is written in C++ and consists of ~80 files depending on the release version, with mainly .cpp (definition) or .h (declaration) extensions and defines and controls all the functions of the UM2 from the stepper motor, heater temperature, user interface and lighting controls. Each file is represented by different tab in the Arduino IDE and can be edited to reconfigure settings. These edits can then be saved and uploaded directly from a computer running the IDE to the Ultimaker via the Main Board USB connection at the rear of the machine. (50)

15

Michael McHugh 10706279

Figure 9 Screen capture of Arduino IDE showing UltimakerMarlin files

2.3.4 G-code The UM2 uses a numerical control programming language called G-code to direct the hardware in the manufacture of 3D prints. The G-code is created by the Cura software bundled with the UM2 from the 3D model to be printed. Below is an example of the start portion of a G-code file for a carabiner to be printed on the UM2. (47,48,51)

;FLAVOR:UltiGCode ;TIME:25534 ;MATERIAL:9510 ;MATERIAL2:0 ;NOZZLE_DIAMETER:0.400000 ;NOZZLE_DIAMETER2:0.400000 ;Layer count: 194 ;LAYER:0 M107 G0 F9000 X72.455 Y96.948 Z0.300 ;TYPE:SKIRT G1 F1200 X71.898 Y97.368 E0.08371 G1 X71.398 Y97.772 E0.16085 G1 X70.880 Y98.222 E0.24319 16

Michael McHugh 10706279 G1 X70.399 Y98.671 E0.32215 G1 X69.913 Y99.157 E0.40463 G1 X69.464 Y99.638 E0.48359

Within the header code at the start of the file the flavor tag relates to setup instructions regarding the printer used; ulticode specfies this is an UM2 and so certain parameters relating to size of build area, 0-axis points for the build plate and print head are taken into account. (47) The time refers to the model print time in seconds; the material refers to settings for the plastic used, though this is often over ridden by settings already stored on the UM2;the nozzle diameter is 0.4 mm in this setup. The material 2 and nozzle diameter 2 parts of the header code are related to printers where a second extruder is mounted; the setup in this example only uses one extruder. (47) After the header code the G-code begins to state the instructions for building the model layer by layer; layer 0 is the skirt about the model; a one nozzle diameter wide loop of filament which is laid down a centimetre outside the outer circumference of the model to be printed. This acts as a prime to ensure the extruder head is printing successfully when it begins work on the model proper. The G0 code is for rapid positioning, moving the print head into place for the initial deposition of the material for the skirt. The G1 code is linear interpolation, where the print head is moved from one point to another in a straight line along the x and y axes to the stipulated position. There are many more G-codes, such as G2 which refers clockwise circular interpolation, G3 which refers to counter clockwise circular interpolation. (51) F refers to feed-rate, or speed at which the print head is moved. X,Y and Z refer to the positional axes; from the front of the UM2 X is left and right, Y is forward and back and Z is up and down. The G-code also controls the amount of plastic being fed through the extruder during the print. (51)

17

Michael McHugh 10706279 2.3.5 UM2 Control System Schematic The main board of the UM2 powers and controls all the other elements of the printer. It is a custom built board based on the Arduino Mega 2560 board, and the Original Ultimaker printer. It integrates and controls the stepper motors, axis limits temperature, cooling, lighting, power in, data in and User Interface into one central unit as shown in the figure below. (50,52)

Figure 10 Ultimaker 2 Control System Schematic (50)

18

Michael McHugh 10706279 2.3.6 UM2 Main Board Integrated Circuits Schematic The main processor is the ATMEGA2560-16AU which communicates with and controls the other main board integrated circuit units which are detailed in the figure below. (50)

No. 1 2 3 4 5 6 7

Unit ATMEGA16U2-MU NE555D a4403geutr-t 3 X INA826 ATMEGA2560-16AU 3 X IRF8736PbF 4 X Allegro A4988

Type USB Controller Precision Timer Buck Converter Instrumentation Amplifier Micro Controller Power MOSFET Stepper Motor Driver Carrier Figure 11 UM2 Mainboard IC

19

Function USB to Serial for firmware Safety cut off and reset Voltage step down converter Amplify thermistor signals Main controller, 256KB Flash Step up for heating elements Translate and drive steppers

Michael McHugh 10706279 2.3.7 UM2 Main Board Connected Components Schematic The standard main board has 19 different components connected to it as detailed in the figure below. (50)

No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Connector Input 220W AC/DC Power Adaptor USB Type B Cable Rocker Switch Fan LED Strip Light X- Axis Limit Switch Y- Axis Limit Switch Z- Axis Limit Switch PT100 B Thermistor PT100 B Thermistor EXP1 EXP2 Heater Cartridge 24Vdc 40w Heated Bed 24Vdc 130W X-Axis Stepper Motor Y-Axis Stepper Motor Z-Axis Stepper Motor Filament Feed Stepper Motor 5Vdc Fan

Function Supply Board with 24Vdc Power Connect Board to PC for Firmware, Tuning. Power Printer On/Off Power Control for Hot End Fans Power Control for LED Strip Lighting Enable end point for X-Axis traverse Enable end point for Y-Axis traverse Enable end point for Z-Axis traverse Measure Temperature at Hot End Measure Temperature at Build Plate Connection to User Interface Board Connection to User Interface Board Heating of Hot end Heating of Build Plate X-Axis Movement of Print Head Y-Axis Movement of Print Head Z-Axis Movement of Build Plate Filament Feed and Retraction Fan cooling for heat-sink

Figure 12 UM2 Mainboard used ports

20

Michael McHugh 10706279 2.3.8 UM2 Main Board Expansion Availability Schematic The board as standard has 16 empty connections for expansion and adaption as detailed below. (50)

No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Connector Label J4 ICSP2 SAFETY1 SAFETY2 ANALOG TEMP2 HEATER2 8/16 STEP E2 FAN 19-24V ICSP1 EXT/IO EXP3 SERIAL ADC2 ADC1

Functionality Interrupt pins to USB Controller In-Circuit Serial Programming input for USB controller chip For safety switch/cut off device For safety switch/cut off device 3-pin header with 5V+, Gnd and ADC 15 via 1K R pins available For temperature sensor if second extruder added For heater cartridge if second extruder added Z-Axis microstep setting: 8 steps open (default), 16 steps closed. For stepper motor if second extruder added Capability to add additional fan In-Circuit Serial Programming input for main Microcontroller 5V, ground and 2 digital input pins 5V, ground and a digital input pin 5V, ground and a transmit and a receive pins 5V, ground and 3 analog/digital input pins 8 analog/digital input pins Figure 13 UM2 Mainboard Unused ports

21

Michael McHugh 10706279 2.3.9 PEEK Filament The PEEK filament used is Indmatec GMBH 1.75 mm diameter PEEK 3D FFF Printer Filament. This filament is extruded from Victrex PEEK 450G granules for which the material properties are detailed in the figure below. (35,53)

Table 2 Victrex 450 G PEEK Material Properties

22

Michael McHugh 10706279

3 System and Process Development 3.1 Constraints To print PEEK filament on the Ultimaker 2 according to requirements derived from the literature means substantial increases to some limits of the printer as shown in the table below. (1)

Table 3 Ultimaker2 to PEEK Conversion Constraints

Hot End maximum (°C) Build Plate maximum (°C) Ambient in printer (°C) Filament Intake (mm)

Ultimaker 2 unmodified 260 100 ~30 2.85

Desired for PEEK Printing 400-430 130+ ~80 1.75

Recommended settings for printing PEEK using a filament based extrusion process state a hot end temperature of up to 410°C, a build plate temperature of 130°C and an ambient temperature of 80°C. (1) The PEEK filament used has a diameter of 1.75 mm, whereas the UM2 uses filaments of 2.85 mm diameter. (35)

3.2 Methodology Due to the complexity of the development process an approach methodology was devised and implemented that allowed for a structured breakdown, analysis and implementation of the main tasks within the conversion and adaption of the UM2 for PEEK. The methodology was to first divide the main goal into four logical individual aims as shown in the figure below.

23

Michael McHugh 10706279

Figure 14 Main Goal and Aims

These aims were then divided into tasks that would be approached from either software related or hardware related aspects as shown in the figure below.

Figure 15 Software/Hardware Task Approach

The tasks were then carried out and implemented in an iterative process, assessing after each task had been completed if the aim had been achieved. Once one aim had been achieved the next aim was approached. Upon completion of the four aims the UM2 conversion to PEEK was complete. This process is detailed below.

24

Michael McHugh 10706279

3.3 Aim 1: Hot End Adaption

Figure 16 Tasks within Hot End Conversion Aim

3.3.1 Task 1: Software Maximum Temperature Limit To test how to best modify the UM2 for printing PEEK the hard limits of the standard set-up needed to be established. To test these the soft limits put in place by the firmware needed to be removed. To change these limits the Arduino IDE was installed onto a computer and adapted for use with the UM2. The Arduino IDE needs to be altered to accommodate compiling of the Marlin fork used by Ultimaker so that it will work on the UM2 main board. Specifically: open the file called "twi.c" located in the Arduino IDE install location, and locate the file via arduino-1.0.3\libraries\Wire\utility". Open the file with a text editor, remove all code from the line within the file "SIGNAL(TWI_vect)" and below. (54) The Ultimaker 2 firmware version 15.04.4 can then be loaded into the Arduino IDE. The configuration.h file, which defines many of the main settings for the hardware, was edited at the portion of code expressing the max temperature settings shown below : #define HEATER_0_MAXTEMP 275 #define HEATER_1_MAXTEMP 275 #define HEATER_2_MAXTEMP 275 25

Michael McHugh 10706279 The max temperature setting achievable via user input is then defined further as 15°C less than this maxtemp setting. So for testing limits, and to enable a User Interface settable temperature of 410°+°C the code was changed to: #define HEATER_0_MAXTEMP 600 #define HEATER_1_MAXTEMP 600 #define HEATER_2_MAXTEMP 600 The code was then compiled and loaded onto the Ultimaker 2 main board. This changed the software limit on temperature successfully from 260°C to 575°C within the settings on the UM2. A test was performed to see if the previous upper limit of 260°C could be surpassed on the hardware; 340°C was reached using the hardware on the UM2.

3.3.2 Task 2: Thermocouple Installation Upon testing it was found that the thermistor units were below specification, and from ~330°C upwards they could not sustain repeated operation and would fail. It was decided to incorporate a K-type thermocouple into the UM2 instead. This required the addition of an AD597 amplifier and conditioner unit into the main board as shown below. (55–57)

Figure 17 Installation of Thermocouple into Main Board

After installation into the main board the AD597 was powered and grounded via pins on the SERIAL connector on the main board and sending the thermocouple readings via a pin on EXP3. 26

Michael McHugh 10706279 This installation required altering of the firmware to recognise the input as a K-type thermocouple and to reassign the pin for the new location of the sensor input. The code below refers to the type of sensor used for the hot end temperature measurement, and is located in the Configuration.h file: #define TEMP_SENSOR_0 20

where TEMP_SENSOR_0 is defined as the hot end sensor, and the number following it, in this case a 20, refers to the sensor type used as stated below: //// Temperature sensor settings: // -2 is thermocouple with MAX6675 (only for sensor 0) // -1 is thermocouple with AD595 // 0 is not used // 1 is 100k thermistor - best choice for EPCOS 100k (4.7k pullup) // 2 is 200k thermistor - ATC Semitec 204GT-2 (4.7k pullup) // 3 is mendel-parts thermistor (4.7k pullup) // 4 is 10k thermistor !! do not use it for a hotend. It gives bad resolution at high temp. !! // 5 is 100K thermistor - ATC Semitec 104GT-2 (Used in ParCan) (4.7k pullup) // 6 is 100k EPCOS - Not as accurate as table 1 (created using a fluke thermocouple) (4.7k pullup) // 7 is 100k Honeywell thermistor 135-104LAG-J01 (4.7k pullup) // 8 is 100k 0603 SMD Vishay NTCS0603E3104FXT (4.7k pullup) // 9 is 100k GE Sensing AL03006-58.2K-97-G1 (4.7k pullup) // 10 is 100k RS thermistor 198-961 (4.7k pullup) // 20 is PT100 with INA826 amp in Ultiboard v2.0

which means a change in the sensor from the PT100 to the Thermocouple with an AD597 amplifer requires the change #define TEMP_SENSOR_0 20

to #define TEMP_SENSOR_0 -1

as the AD597 is equivalent to the AD595 for this purpose. Pins on the Main Board now needed to be reassigned via the firmware as the thermocouple sensor was now connected differently from the PT100 connector. The file pins.h designates the pins on the main board according to the type of board used. In configuration.h the main board is defined as: 27

Michael McHugh 10706279 // 72 = Ultiboard v2.0 (includes Ultimaker 2)

within the Ultiboard v2.0 section in the pins.h file the temperature sensor for the hot end is defined as: #define TEMP_0_PIN 8

Now the temperature readings are being fed via the thermocouple into the pin assigned as 0 on the main board and so the code is changed accordingly: #define TEMP_0_PIN 0

This amended firmware was uploaded to the UM2 whereupon the Thermocouple provided temperature readings for the hot end.

3.3.3 Task 3: Software Temperature Rise Rate Limit Upon testing it was found that the UM2 firmware had safeguards in place that required the temperature of the hot end to have a measured rise of 10°C every 10 seconds when rising towards the set temperature. This was in case of a failure or decoupling of the thermocouple which would then return a false temperature reading and result in runaway heating of the hot end. However in this case the temperature was rising less than 10°C every ten seconds because the printer was attempting much higher temperatures than in normal usage and the heater cartridge output was not enough to keep pace at these higher temperatures and so would return the error. Because this code was deeply integrated into the current releases of the firmware the solution chosen was to revert to an earlier edition of the firmware, to version 14.07.0 on the Github repository, which lacked the safeguards. The older firmware version was altered to include the earlier changes to the maximum temperature and the installation of the thermocouple. (47)

3.3.4 Task 4: Heater Cartridge and Hot End Upgrade The standard hot end uses a Polytetrafluoroethylene (PTFE) coupler which has a melting point of 327°C, is designed for 2.85mm filament, uses a fixed nozzle, has inadequate heat isolation from the plastic print head, and has many different parts made of different materials which it was felt could cause issues going from room temperature to 400°+ and back. A hot-end which could overcome these issues was assembled from parts obtained from E3D, a 3D printing parts manufacturer, and integrated into the standard print head as shown in the figure below. (58) 28

Michael McHugh 10706279

Figure 18 UM2 Hot End Adaption

The all metal new hot end incorporates a heat sink which accommodates 1.75 mm filament through its central bore, and effectively thermally isolates the hot end from the rest of the print head. The heat sink attaches to a heater block with fixing points for the heater cartridge, thermocouple and nozzle. During this hot end installation the heater cartridge was also upgraded from a 25 Watt unit to a 40 Watt unit as the previous cartridge was unable to pass temperatures of 375°C. (59)

3.3.5 Task 5: Calibration The hot end was now in a lower position in the Z-axis and so the build plate would impact upon it if rising to position before beginning a print. Within the firmware the settings for the limit positions are defined as: #define #define #define #define #define #define

X_MAX_POS X_MIN_POS Y_MAX_POS Y_MIN_POS Z_MAX_POS Z_MIN_POS

230 0 230 0 225 0

The line #define Z_MAX_POS 225

was changed to #define Z_MAX_POS 180

29

Michael McHugh 10706279

to prevent contact between the hot end and the build plate. The firmware uses proportional–integral–derivative (PID) control to power the heater cartridge. PID values for the new 40 watt heater cartridge were set in the firmware by means of the PID auto tune setting in Cura, via File->Preferences->Set window to Pronterface This allows the entering of the following G-code command: M303 C8 S175

which activates PID auto-tune via the firmware and provides new PID values. The new hot end setup was tested and registered a temperature of 420°C on the display screen of the UM2. Upon checking the accuracy of this temperature with a calibrated external digital thermometer inserted into the hot end it was found that this reading was low; the actual temperature at this setting was 460°C. It was found that by setting the UM2 hot end temperature to 390°C in the settings menu the desired actual temperature of 410°C was attained and usable while printing.

30

Michael McHugh 10706279

3.4 Aim 2: Build Plate Adaption to 130°C

Figure 19 Tasks within Build Plate Conversion Aim

3.4.1 Task 1: Firmware Maximum Temperature Limit The adaption of the build plate to 130°C first required the lifting of software limits similar to those removed in raising the maximum temperature for the hot end. Within the configuration.h file #define BED_MAXTEMP 100

was changed to #define BED_MAXTEMP 300

This successfully lifted the build plate software constraint.

3.4.2 Task 2: Hardware Limit Testing At this point the hardware limitations of the build plate were reached, the maximum temperature attainable was 110°C. It was decided to test if raising the ambient temperature would raise the build plate temperaure A rudimentary cardboard enclosure was built upon the top and front openings of the UM2, and the hot end and build plate were set to maximum. The heat created by the hot end and build plate, and contained within the cardboard enclosure, successfully raised the ambient temperature to 50°C. Within this environment the build plate reached a temperature of 126°C. 31

Michael McHugh 10706279 The approach then taken was to return to this aim and take readings once permanent enclosures had been constructed, and the ambient temperature had been raised to 85°C as it was felt likely the heated enclosure would enable the build plate to raise its temperature sufficiently.

3.4.3 Task 3: Calibration Once the enclosures were successfully in place and an ambient temperature of 85°C had been reached the build plate reached temperatures of 152°C. This reading is from an external digital thermometer as the Thermistor attached to the build plate was found to be unreliable at these temperatures; it would give readings 40°C to 200°C higher than the actual readings. Because of this it was found that, unlike the hot end, this meant that the build plate temperature setting could not be accurately relied on as an offset. The external digital thermometer was used as the setting temperature, and the software build plate temperature was set from this and monitored for accuracy during a print.

32

Michael McHugh 10706279

3.5 Aim 3: Filament Feeder Adaption to 1.75 mm

Figure 20 Tasks within Filament Feeder Conversion Aim

3.5.1 Task 1: Installation of 1.75 mm Bowden tube The PEEK filament available is 1.75 mm in diameter; the UM2 is set up for 2.85 mm filament, and has a bowden tube with an inner diameter of 3 mm. To adapt the printer for the PEEK filament it was necessary to swap out the bowden tube for one with an diameter of 1.85 mm diameter as the wider inner tube would cause losses of pressure from the filament feeder due to the narrower filament being able to occupy this space during feeding. The hot end chosen helped enable this adaption as it accommodates the narrower bowden tubing.

3.5.2 Task 2: Filament Feeder Modification The feeder mechanism is constructed for 2.85 mm filaments, and it was found that the 1.75 mm filament was not being gripped and fed sufficiently, and was also travelling laterally away from the grooved feeder wheel within the mechanism. To adapt the feeder mechanism for 1.75 mm filament the following changes were made:

33

Michael McHugh 10706279

Figure 21 Feeder Mechanism Adaption

1.

Bowden Clip installed for narrower Bowden

2 & 3. Bowden tube inserted into feeder to aid alignment onto feeder wheel 4.

Adjustable Pressure Lever Filed to enable gripping of 1.75 mm filament

3.5.3 Task 3: Firmware Power Increase to the Feeder Stepper Motor After this adaption the feeder motor power was increased to aid gripping and extrusion pressures by increasing the feeder stepper motor current limit from 1250 mA to 1600 mA in the firmware. In the configuration_adv.h file the lines // Default motor current for XY,Z,E in mA #define DEFAULT_PWM_MOTOR_CURRENT {1300, 1300, 1250}

were altered to // Default motor current for XY,Z,E in mA #define DEFAULT_PWM_MOTOR_CURRENT {1300, 1300, 1600}

which successfully increased feeder power and removed extrusion difficulties with the 1.75 mm filament.

34

Michael McHugh 10706279

3.6 Aim 4: Ambient Temperature Adaption to 80°C

Figure 22 Tasks within Ambient Temperature Conversion Aim

To enable an ambient environment of 80°C within the build area a design that enclosed the UM2 with a door over the front and a hood on top, and then enclosed the entire unit in a larger enclosure was chosen versus heat lamps, which it was felt would not create a homogenous heat environment. The outer enclosure would be heated by means of circulating heated air, and the entire UM2 would then be heated to this temperature, and accordingly its inner ambient temperature would rise, whilst also retaining and using the heat generated from the build plate and hot end. This method was chosen for stability of temperature and air current within the inner enclosure so that printing would be unaffected by air currents and temperature fluctuations when heating the ambient temperature, and also when operating the controls of the machine. An aluminium frame and clear 10 mm polycarbonate walls were chosen for the construction of the outer enclosure, with polycarbonate chosen for the inner door and hood as these materials would be stable under any temperatures reached, had reasonable thermal insulation qualities, and would allow good visibility of the machine at all times. The hood for the top of the UM2 was constructed so as to allow the bowden tube and print head electronics freedom of movement during printing. The doors to both enclosures used rubber sealing to aid in heat retention and stability.

35

Michael McHugh 10706279 The heating apparatus used was a hairdryer as after investigation it was the simplest solution to provide an adjustable temperature with a safety cut off if the temperatures went too high. The enclosures and heating apparatus are shown in the figure below.

Figure 23 UM2 with Dual Enclosures and Heating Apparatus

where 1. 2. 3. 4.

Polycarbonate outer enclosure with aluminium frame Heating apparatus Polycarbonate Hood to enclose top of UM2 Polycarbonate door to enclose front of UM2

The dual enclosure set up provided a stable temperature of 85°C within the UM2 build area. The time for this setup to reach this temperature is about an hour.

36

Michael McHugh 10706279

3.7 PEEK FFF Printer Testing 3.7.1 Methodology A series of investigations were undertaken focussing on two phases, first the printing of a simplistic 3D part in order to examine the output from the printer during development in order to gauge progress, effects of changes, and aid development decisions; and secondly a series of investigations focussing on material characteristics resulting from the newly implemented FFF PEEK printing solution. All test prints were printed from STL files that were imported in Cura Version 15.04.4 and exported as G-code files with the following print settings:

Table 4 PEEK FFF Test Print Settings

Setting Nozzle Diameter (mm) Layer height (mm) Shell thickness (mm) Bottom/Top thickness Travel speed (mm/s) Top/bottom speed (mm/s) Infill speed (mm/s) Outer shell speed (mm/s) Width of 1-layer Brim (mm)

Value 0.4 0.1 0.8 0.6 150 15 80 30 10

3.7.2 Development Testing Methodology Testing of the prints whilst constructing the prototype was done using a carabiner print with a 20% infill setting. The objective was to have a simple guide by which we could measure if the printer was producing parts accurately, or if the prototype needed refinement. The carabiner print was chosen because it was simple, quick to print, but enabled easily definable analysis of print quality and part functionality All carabiner prints were produced using a hot end temperature of 410°C and a layer of Kapton tape used to provide grip between the print and the build plate. They were printed on normal setting within Cura, and have maximum dimensions of 61mm long by 30mm wide by 4 mm deep. (60) The most informative carabiner test results are discussed in the next chapter.

37

Michael McHugh 10706279 3.7.3 Tensile Testing Methodology Three series of tests were performed to evaluate output from the printer by means of tensile testing according to ISO 527-1 Plastics-Determination of Tensile Properties and ISO 527-2 Plastics-Determination of Tensile Properties, Test Conditions for Moulding and Extrusion Plastics. (61,62) Test 1 uses 100% infill and is to ascertain the tensile strength of printed PEEK produced by the FFF printer using the conditions found in the literature, and comparing these results with the literature and technical data. Test 1 and 2 are to test and compare the tensile quality of parts printed without 100% infill and also to compare results printed below and above the glass transition temperature of PEEK, which is 143°C. (53)

Table 5 Tensile Test Series Parameters

Test 1 Parameters Test 2 Parameters Test 3 Parameters Hot End Temperature 410°C 410°C 410°C Build Plate Temperature 130°C 130°C 150°C Ambient Temperature 85°C 85°C 85°C Infill Amount 100% 20% 20%

The material tested is Indmatec GMBH 1.75 mm diameter PEEK 3D FFF Printer Filament. This filament is extruded from Victrex PEEK 450G granules. The printed specimens were produced from a CAD file which had been converted to print on the printer by Cura software. The settings for the print were normal print settings from the quick print menu, with the infill percentage altered to 100% to gain a solid sample piece. The test specimen dimensions in CAD form were as follows:

38

Michael McHugh 10706279

Figure 24 ISO 527 Test Specimen Requirements (62)

where, in mm, L=50

h=2

L0=20

b1=4

l1=25

b2=12.5

l2=45

r1=8

l3=75

r2=12.5

The test specimens were then extruded by means of Fused Filament Fabrication through a nozzle of 0.4 mm diameter at a temperature of 410°C. This material was extruded onto a build platform with a temperature of 130°C or 150°C, and into an ambient environment of 85°C. A layer of Kapton tape was used to provide grip between the print and the build plate. The specimens were printed laying flat, with the X axis traversing the length, the Y axis traversing the breadth, and the Z-axis of the printer relating to the height of the specimens. The specimens were printed with a brim of 10 mm about the piece to ensure adhesion to the build plate during the build. After printing this brim was removed and the pieces dimensions were checked. There was some variations in the region of 0.1 mm from the CAD dimensions but these were within the tolerances set out by ISO 527-2 for Specimen Type 5A. The actual physical dimensions for each test specimen were used for testing results and calculations. (62) 39

Michael McHugh 10706279

Figure 25 FFF PEEK Test Specimen

Five specimens were used for the testing of the tensile qualities for each of the three parameters. The testing was performed in an open environment with an ambient temperature of 18°C. The test machine used was an Instron 3366 running on Instron Bluehill software. The specimens were held in parallel clamping jaws and tension was applied by moving the jaws apart at a rate of 40 mm/minute. The results of these tests are tabulated and discussed in the next chapter.

40

Michael McHugh 10706279

4 Results and Discussion 4.1 PEEK FFF Printer Development Testing

Figure 26 Carabiner test prints (60)

Carabiner test print 1 was printed without a heated environment and with a build plate temperature of 100°C to test if a heated environment was required. With these settings most prints would fail; out of ten prints this was the most successful. The primary failure was the hot end catching on printed filament as these pieces tended to warp upwards during printing. This piece almost completed but attached itself to the hot end near completion. It lacked the final layer only.

4.1.1 Carabiner test print 1 Points of Interest:  Poor adhesion between the layers, delaminates easily suggesting cooling too abruptly upon extrusion to adhere layer to layer. (1)  Very brittle and amber in colour suggesting tendency toward amorphous structure. (1)  Warped upward as shown in figure below suggesting high thermal stresses induced during cooling. (1)  Resolution of the print was good.  Limited functionality as a carabiner 41

Michael McHugh 10706279

Figure 27 Carabiner Test Print 1

Carabiner test print 2 is from the same batch as test print 1 but was treated with a heat gun set at 400°C to see what effect post process heat treatment would have on the print.

4.1.2      

Carabiner test print 2 Points of Interest: Post Processed with heat gun at 400°C Post Processing improved part solidity Colour changed from amber to stock PEEK beige suggesting increased crystalline structure. (1) Heat Gun introduced loss of resolution and more warping suggesting this method too severe for any useful post processing Little or no flex before breaking Non functional

Figure 28 Carabiner Test Print 2

Carabiner test print 3 and 4 are some of the first prints from within the heated chamber. The chamber enabled and ambient temperature of 85°C and a build plate temperature of 130°C. At this point the printer was producing prints with sponge like appearance as shown 42

Michael McHugh 10706279 in the figure below. This is related to under extrusion. It was found that this was due to slipping of the filament at the feeder which was adapted to better handle the filament and environment. The prints did exhibit less brittleness and warping, and were flexible which pointed to better layer adhesion.

4.1.3      

Carabiner test print 3 and 4 Points of Interest: Heated environment improved Z-axis cohesion, much stronger in hand Underextrusion produces sponge like prints Dimensional accuracy very low Colour very close to the PEEK filament and stock PEEK suggesting good crystallinity versus earlier prints which were amber and suggested amorphous states. No Functionality

Figure 29 Carabiner Test Print 4

The adapted filament feeder produced the next three runs of prints during which the machine was tuned for better results. Test print 5 shows the quality improving but none of the prints were adhering to the plate during the build due to warping, which was likely an effect of PEEKs shrinkage as it cooled due to its coefficient of thermal expansion of 120. It was decided to use a brim of 1 cm about the prints to create more surface area for adhesion. (53) Test print 6 shows a carabiner that has material spread between the opening of the carabiner, which suggests over extrusion, again likely due to effects related to PEEKs coefficient of thermal expansion, see figure below. The flow rate of the material was adjusted to reduce the amount being extruded. It was found that a flow rate of 95% produced better results.

43

Michael McHugh 10706279 4.1.4     

Carabiner test print 5 Points of Interest: Underextrusion issues resolved Quality of print now high Print warping off buildplate suggesting thermal stresses still high Not functional

4.1.5 Carabiner test print 6 Points of Interest:  Brim introduced which removes warping effect  Printer overextruding which is likely related to PEEKs coefficient of thermal expansion  Not functional 

Figure 30 Carabiner Test Print 6 detail

The final Carabiner test print runs provided satisfactory results from the prototype. Print failures were much reduced. The resolution and quality were as expected from initial print settings, the piece exhibited no warping or layer delamination, see figure below. The carabiner was fully functional. At this point it was decided to test the material properties of the PEEK using standard tensile tests.

44

Michael McHugh 10706279 4.1.6      

Carabiner test print 7 Points of Interest: No warping of part Good cohesion between layers Detail and resolution of print are of good quality No extrusion issues Colour indicates high level of crystallinity within structure compared to early prints Part functional

Figure 31 Carabiner Test Print 7

45

Michael McHugh 10706279

4.2 PEEK FFF Printer Tensile Testing Results 4.2.1 Series 1 Results: 100% Infill PEEK Specimens

Table 6 Tensile Test 1 Results 100% Infill PEEK Specimens

Specimen No.

Maximum Load [N]

Extension at Break (Standard) [mm]

Tensile stress at Maximum Load [MPa]

Tensile strain (Extension) at Maximum Load [mm/mm]

Load at Break (Standard) [N]

1

738.23

1.86796

97.14

0.07

738.23

2

710.58

1.73451

89.8

0.07

710.58

3

649.69

1.3681

82.03

0.05

649.69

4

717.02

1.70147

90.93

0.07

717.02

5

763.89

1.80143

95.73

0.07

763.89

900 800 700

Load (N)

600 500 400 300 200 100

0 0

0.2

0.4

0.6

0.8

1

1.2

1.4

Extension (mm) Figure 32 Tensile Test Specimens, 100% Infill, Load versus Extension.

46

1.6

1.8

2

Michael McHugh 10706279 4.2.2 Series 2 Results: 20% Infill PEEK Specimens, Build Plate 130°C

Table 7 Tensile Test 2 Results 20% Infill PEEK Specimens, Build Plate 130°C

Specimen No.

Maximum Load [N]

Extension at Break (Standard) [mm]

Tensile stress at Maximum Load [MPa]

Tensile strain (Extension) at Maximum Load [mm/mm]

1

436.24

1.90117

61.07

0.07

2

483.91

1.90145

66.38

0.07

3

308.46

1.0013

52

0.04

4

519.31

1.63461

70.88

0.06

5

414.79

1.40137

61.77

0.05

600

500

Load (N)

400

300

200

100

0 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Extension (mm) Figure 33 Tensile Test Specimens, 130°C Build Plate, Load versus Extension.

47

1.8

2

Michael McHugh 10706279

4.2.3 Series 3 Results: 20% Infill PEEK Specimens, Build Plate 150°C

Table 8 Tensile Test 3 Results 20% Infill PEEK Specimens, Build Plate 150°C

Specimen No.

Maximum Load [N]

Extension at Break (Standard) [mm]

Ultimate Tensile Strength [MPa]

Tensile strain (Extension) at Maximum Load [mm/mm]

1

505.45

0.18377

68.6

0.01

2

464.43

1.50116

64.77

0.06

3

420.89

1.26803

61.45

0.05

4

429.11

1.43457

64.43

0.06

5

456.06

1.60151

61.92

0.06

600

500

Load (N)

400

300

200

100

0 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Extension (mm) Figure 34 Tensile Test Specimens, 150°C Build Plate, Load versus Extension.

48

1.8

2

Michael McHugh 10706279

4.3 PEEK FFF Tensile Testing Discussion 4.3.1 Series 1 Discussion: 100% Infill FFF PEEK Specimens A range of values derived from the combined test results of the five specimens series one are shown in the table below.

Table 9 Combined Values for Tensile Test Results on 100% Infill PEEK Specimens

Values from Combined Specimen Results

Tensile stress at Maximum Load [MPa]

Tensile strain (Extension) at Maximum Load [mm/mm]

Maximum

97.14

0.07

Mean

91.13

0.07

Median

90.93

0.07

Minimum

82.03

0.05

Range

15.1

0.02

5.95328

0.00774

Standard deviation

Regards tensile strength of the 100% infill test specimens we see that the tensile strength is 91.13 MPa versus 98 MPa for the Victrex PEEK 450G from which the filament is produced. A range between the tensile strength values of 15.1 MPa and a standard deviation of 5.95 MPa shows a lack of uniformity across the specimens; this can be attributed to the nature of FFF and that defects attributed to air gaps and porosity is present in some specimens more than others. The mean tensile strength of 91.13 Mpa compares well with the figure of 75.06 MPa for the data obtained by Vaezi and Yang in their 2015 paper, which is the closest equivalent to our FFF PEEK methodology in the literature. Vaezi and Yangs results were produced using a PEEK filament extruded from PEEK stock with the same technical specifications to that used in this project and so provides for interesting comparisons. (1,35,53) A porosity of 14% was introduced into Vaezi and Yangs 100% infill prints that is stated is due to an inability to prevent the formation of air gaps between deposited filaments within layers; it is that in these tests evidence is provided that there is an ability to reduce the air gap formation and therefore the amount of porosity to lesser levels. (1) A similar finding was reported by Wu et al who attributed a drop in tensile strength from 100 MPa to 56 Mpa to the influence of raster angle and layer thickness, and the inclusion of air gaps due to the same issues identified by Vaezi and Yang. (1,20) 49

Michael McHugh 10706279 The higher yield strengths, and therefore assumed reduction in air gaps and porosity, in the tests conducted is perhaps down to careful calibration of a few key variables related to FFF 3D printing practices, including: flow rate, that is, the amount of material being extruded over time, the shell thickness of object being printed, the line width of the extruded filament of the infill, and a layer height between layers on the Z axis that ensure tight packing of filament as it is laid, combined with ensuring that these variables, excluding the layer height, are ratios of nozzle diameter. The reduction in porosity is a positive result as it increases the tensile strength of the 3D printed PEEK. Increased porosity if desired can be added to portions of an object being printed by changing settings for different layers, so that part of the object has less porosity and greater mechanical strength, say within the internal structure of a biomedical component, than another part of the object where the increased porosity might be desired, say at the surface of the same biomedical component. Further testing is necessary, however, to accurately determine the levels of porosity within the prints. The specimens exhibited brittleness with little to no plastic deformation, as we can see from the load/extension graph, and the picture below, which perhaps could be attributed to a high level of crystallinity in the specimens, however further testing is necessary to determine the crystalline nature of the printed PEEK.

Figure 35 Specimen 4 100% infill post tensile testing

50

Michael McHugh 10706279 4.3.2 Series 2 & 3 Discussion: Build Plate 130°C Versus 150°C Two series of specimens were printed with 20°% infill settings; test series 2 onto a build plate of 130°C, and test series 3 onto a build plate of 150°C to determine the effects of printing above the glass transition temperature of PEEK, 143°C. The prints produced onto a build plate above the glass transition temperature are more closely grouped together in their results in terms of extension, from 1.3 to 1.68 mm, a range of 0.34 mm, compared to the extensions of the 130°C build plate test specimens, from 1.07 to 1.9 mm, a range of .81 mm. This suggests more conformity in the 150°C build plate printed specimens, and again is perhaps related to the underlying crystallinity being formed. Further research is needed.

51

Michael McHugh 10706279

5 Summary and Conclusion 5.1.1 Summary The goal of this research project was to develop the capacity to print Polyether-etherketone using 3D Fused Filament Fabrication, which was successfully achieved. Initial research into the literature laid a background for an understanding of the central topics related to the research project goal: an overview of Additive Manufacturing, Fused Filament Fabrication, PEEK material properties and uses related to FFF, and results and observations from a number of key papers that had manufactured PEEK using FFF, including that porosity was attainable using FFF for PEEK, and also that the porosity control issues when attempting to print solid specimens, particularly in relation to tensile strength. (1,20) It was found from the literature that the main criteria for printing PEEK using a filament based extrusion process state to be a hot end temperature of up to 410°C, a build plate temperature of 130°C and an ambient temperature of 80°C. (1) An Ultimaker 2 FFF 3D printer was used as the platform for this project, and 1.75mm PEEK filament from Indmatec GMBH was the material source. This system was characterised in detail from a systems and process point of view, and a methodology to adapt the machine to print PEEK was established. Four individual aims were identified that would act as a logical breakdown of the main goal. A simple testing procedure was used to assess progress throughout the development. Each aim was achieved through an iterative process that typically alternated between hardware and software related tasks, thereby satisfying the main goal of developing the capacity to print PEEK. A methodology to test the PEEK FFF printer according to ISO 527 was developed and carried out. The results proved that the capacity to print PEEK using FFF had been successfully developed, and that print specimens had been produced with tensile strengths that compared well to results found in the literature.

52

Michael McHugh 10706279 5.1.2 Conclusion The objectives for this project were: 1) Characterise the current Fused Filament Fabrication machine capacity and identify constraints for printing PEEK according to the existing scientific literature. 2) Develop a hardware and software adaption strategy to enable printing of PEEK on the Ultimaker 2 3) Undertake 3D printing investigations using the developed system and analyse results, comparing with existing scientific literature. After research into the literature that established the context and requirements of a PEEK FFF printer, an Ultimaker 2 FFF was successfully characterised, developed and tested for PEEK printing. It was found that that an Ultimaker 2 FFF 3D printer can, with the adaptions detailed in this paper, produce PEEK specimens by means of a hot end temperature of 410°C, onto a build plate of 130°C, into an ambient temperature of 85C. Investigations were undertaken using the developed system and the specimens produced were found to have a mean ultimate tensile strength of 91 MPa, comparing well with the results found in the literature. It can be concluded the the objectives of the project were completed, though further works detailed below would add greater value and insight to the topic.

5.1.3 Future Works Future research is needed to establish further the mechanical properties and possible uses of the printed PEEK, including investigations into porosity and changes at the level of the crystalline structure. Tensile testing was conducted in the X-Y axes direction; testing in Z-axis direction would be informative particularly regarding layer to layer adhesion. Testing on the effects of print settings such as layer height and raster angle should be carried out. Developing the capacity to 3D print PEEK using FFF also opens up the opportunity to investigate other materials, for example other high temperature extrusion polymer materials, using the process. Future works should explore these possibilities. Future Works are needed to establish the long term robustness of the developed system as it is operating in an environment with elevated temperatures which are far above its maximum recommended working conditions. These future works should work towards improving the reliability and accuracy of the process, system and results.

53

Michael McHugh 10706279

References 1. Vaezi M, Yang S. Extrusion-based additive manufacturing of PEEK for biomedical applications. Virtual Phys Prototyp. 2015 Jul 3;10(3):123–35. 2. Bogers M, Hadar R, Bilberg A. Additive manufacturing for consumer-centric business models: Implications for supply chains in consumer goods manufacturing. Technol Forecast Soc Change. 2016 Jan 1;102:225–39. 3. Chhaya M, Poh P, Balmayor E, van Griensven M, Schantz J, Hutmacher D. Additive manufacturing in biomedical sciences and the need for definitions and norms. EXPERT Rev Med DEVICES. 2015 Sep;12(5):537–43. 4. Gao W, Zhang Y, Ramanujan D, Ramani K, Chen Y, Williams CB, et al. The status, challenges, and future of additive manufacturing in engineering. Comput-Aided Des. 2015 Dec;69:65–89. 5. Hengsbach S, Lantada A. Rapid prototyping of multi-scale biomedical microdevices by combining additive manufacturing technologies. Biomed MICRODEVICES. 2014 Aug;16(4):617–27. 6. Jaekel DJ, MacDonald DW, Kurtz SM. Characterization of PEEK biomaterials using the small punch test. J Mech Behav Biomed Mater. 2011 Oct;4(7):1275–82. 7. Mahamood RM, Akinlabi ET, Shukla M, Pityana S. Revolutionary Additive Manufacturing: An Overview. Lasers Eng Old City Publ. 2014 Jun;27(3/4):161–78. 8. Murr LE, Gaytan SM, Medina F, Lopez H, Martinez E, Machado BI, et al. Next-generation biomedical implants using additive manufacturing of complex, cellular and functional mesh arrays. Philosophical Transactions: Mathematical, Physical and Engineering Sciences. 2010;(1917):1999. 9. Wong KV, Hernandez A, Wong KV, Hernandez A. A Review of Additive Manufacturing, A Review of Additive Manufacturing. Int Sch Res Not Int Sch Res Not. 2012 Aug 16;2012, 2012:e208760. 10. D’Aveni R. The 3-D Printing Revolution. Harv Bus Rev. 2015 May;93(5):40–8. 11. Elwany A. Making 3-D futures reality. Ind Eng IE. 2014 Sep;46(9):32–5. 12. Slotwinski JA. Additive Manufacturing: Overview and NDE Challenges. AIP Conf Proc. 2014 Feb 18;1581:1173–7. 13. Venekamp NJR, Le Fever HT. Application Areas of Additive Manufacturing : From Curiosity to Application. IEEE Technol Soc Mag. 2015 Sep;34(3):81–7.

54

Michael McHugh 10706279 14. Weller C, Kleer R, Piller FT. Economic implications of 3D printing: Market structure models in light of additive manufacturing revisited. Int J Prod Econ. 2015 Jun;164:43– 56. 15. Newman ST, Zhu Z, Dhokia V, Shokrani A. Process planning for additive and subtractive manufacturing technologies. CIRP Ann - Manuf Technol. 2015;64(1):467–70. 16. Yang S, Zhao YF. Additive manufacturing-enabled design theory and methodology: a critical review. Int J Adv Manuf Technol. 2015 Mar 24;80(1-4):327–42. 17. Bland S, Conner B. Mapping out the additive manufacturing landscape. Met Powder Rep. 2015 May;70(3):115–9. 18. Conner BP, Manogharan GP, Martof AN, Rodomsky LM, Rodomsky CM, Jordan DC, et al. Making sense of 3-D printing: Creating a map of additive manufacturing products and services. Addit Manuf. 2014 Oct;1-4:64–76. 19. Stansbury JW, Idacavage MJ. 3D printing with polymers: Challenges among expanding options and opportunities. Dent Mater. 2016 Jan;32(1):54–64. 20. Wenzheng Wu, Peng Geng, Guiwei Li, Di Zhao, Haibo Zhang, Ji Zhao. Influence of Layer Thickness and Raster Angle on the Mechanical Properties of 3D-Printed PEEK and a Comparative Mechanical Study between PEEK and ABS. Mater 1996-1944. 2015 Sep;8(9):5834–46. 21. Chen F, Ou H, Lu B, Long H. A constitutive model of polyether-ether-ketone (PEEK). J Mech Behav Biomed Mater. 2016 Jan;53:427–33. 22. Evans NT, Torstrick FB, Lee CSD, Dupont KM, Safranski DL, Chang WA, et al. Highstrength, surface-porous polyether-ether-ketone for load-bearing orthopedic implants. Acta Biomater. 2015 Feb;13:159–67. 23. Garcia-Gonzalez D, Rusinek A, Jankowiak T, Arias A. Mechanical impact behavior of polyether–ether–ketone (PEEK). Compos Struct. 2015 Jun;124:88–99. 24. Evans NT, Irvin CW, Safranski DL, Gall K. Impact of surface porosity and topography on the mechanical behavior of high strength biomedical polymers. J Mech Behav Biomed Mater. 2016 Jun;59:459–73. 25. Sobieraj MC, Murphy JE, Brinkman JG, Kurtz SM, Rimnac CM. Notched fatigue behavior of PEEK. Biomaterials. 2010 Dec;31(35):9156–62. 26. Bikas H, Stavropoulos P, Chryssolouris G. Additive manufacturing methods and modelling approaches: a critical review. Int J Adv Manuf Technol. 2015 Jul 24;1–17. 27. The Types of 3-D Printing. Libr Technol Rep. 2014 Jul;50(5):8–12. 28. Monzón M, Ortega Z, Martínez A, Ortega F. Standardization in additive manufacturing: activities carried out by international organizations and projects. Int J Adv Manuf Technol. 2015 Feb;76(5-8):1111–21. 55

Michael McHugh 10706279 29. Mcor ARKe 3D Photorealistic Colour Printer [Internet]. Mcor Technologies. [cited 2016 Mar 19]. Available from: http://mcortechnologies.com/3d-printers/arke-3dphotorealistic-colour-printer/ 30. Berretta S, Evans KE, Ghita O. Processability of PEEK, a new polymer for High Temperature Laser Sintering (HT-LS). Eur Polym J. 2015 Jul;68:243–66. 31. Lu SX, Cebe P, Capel M. Thermal stability and thermal expansion studies of PEEK and related polyimides. Polymer. 1996 Jul;37(14):2999–3009. 32. Manufacture and thermal deformation analysis of semicrystalline polymer polyether ether ketone by 3D printing. Mater Res Innov. 2014 Aug 2;18(S5):S12. 33. Maksimov RD, Kubat J. Time and temperature dependent deformation of poly(ether ether ketone) (PEEK). Mech Compos Mater. 33(6):517–25. 34. Ultimaker 2+ Specifications | Ultimaker [Internet]. Ultimaker.com. [cited 2016 Mar 20]. Available from: https://ultimaker.com/en/products/ultimaker-2-plus/specifications 35. PEEK FFF 3D Printer [Internet]. Indmatec GmbH. [cited 2016 Mar 19]. Available from: http://www.indmatec.com/en/peek-3d-printer-solution 36. Ultimaker 3D Printing Products | Ultimaker [Internet]. Ultimaker.com. [cited 2016 Mar 23]. Available from: https://ultimaker.com/en/products 37. The basics | Ultimaker [Internet]. Ultimaker.com. [cited 2016 Mar 23]. Available from: https://ultimaker.com/en/manuals/software/cura-15-04/the-basics 38. http://www.zarr.com W designed and developed by Z-. SOLIDWORKS 3D CAD Products | Solid Solutions in Ireland [Internet]. [cited 2016 Mar 23]. Available from: http://www.solidsolutionsireland.ie/solidworks/default.aspx 39. PTC Creo [Internet]. PTC Creo. [cited 2016 Mar 23]. Available from: http://creo.ptc.com/ 40. YouMagine - Where makers collaborate on 3D designs � [Internet]. [cited 2016 Mar 23]. Available from: https://www.youmagine.com/ 41. Thingiverse.com. MakerBot Thingiverse [Internet]. [cited 2016 Mar 23]. Available from: http://www.thingiverse.com/ 42. Designing for 3D printing: Guidlines - Ultimaker Wiki [Internet]. [cited 2016 Mar 23]. Available from: http://wiki.ultimaker.com/Designing_for_3D_printing:_Guidlines 43. Ultimaker Cura Release notes [Internet]. Ultimaker.com. [cited 2016 Mar 7]. Available from: https://ultimaker.com/en/products/cura-software/release-notes 44. Warping | Ultimaker [Internet]. Ultimaker.com. [cited 2016 Mar 23]. Available from: https://ultimaker.com/en/troubleshooting/19537-warping

56

Michael McHugh 10706279 45. Full settings | Ultimaker [Internet]. Ultimaker.com. [cited 2016 Mar 23]. Available from: https://ultimaker.com/en/manuals/software/cura-15-04/full-settings 46. Arduino - Software [Internet]. [cited 2016 Mar 7]. Available from: https://www.arduino.cc/en/Main/Software 47. Ultimaker Github Ultimaker2Marlin firmware [Internet]. GitHub. [cited 2016 Mar 7]. Available from: https://github.com/Ultimaker/Ultimaker2Marlin 48. Home · MarlinFirmware/Marlin Wiki [Internet]. [cited 2016 Mar 23]. Available from: https://github.com/MarlinFirmware/Marlin/wiki 49. Ultimaker/Ultimaker2Marlin Version 14.07.0 [Internet]. GitHub. [cited 2016 Mar 7]. Available from: https://github.com/Ultimaker/Ultimaker2Marlin 50. Ultimaker2 Main Board Schematics [Internet]. GitHub. [cited 2016 Mar 23]. Available from: https://github.com/Ultimaker/Ultimaker2 51. G-code - RepRapWiki [Internet]. [cited 2016 Mar 23]. Available from: http://reprap.org/wiki/G-code 52. Ultimaker2 Main Board precedents. [Internet]. [cited 2016 Mar 9]. Available from: https://github.com/Ultimaker/Ultimaker2/blob/master/1091_Main_board_v2.1.1_(x1) /readme.txt 53. Datasheet Downloads - Victrex [Internet]. [cited 2016 Mar 21]. Available from: https://www.victrex.com/en/datasheets 54. Ultimaker 2 Arduino IDE adjustment [Internet]. Ultimaker.com. [cited 2016 Mar 7]. Available from: https://ultimaker.com/en/community/8807-main-links-for-marlinfirmware 55. External Thermocouple Board v1.0 [Internet]. [cited 2016 Mar 23]. Available from: http://e3d-online.com/External-Thermocouple-Board-v1.0 56. Type K Thermocouple (Welded Tip) [Internet]. [cited 2016 Mar 23]. Available from: http://e3d-online.com/Type-K-Thermocouple 57. Ultimaker’s v1.5.7 PCB - Thermocouple Wiring [Internet]. [cited 2016 Mar 9]. Available from: http://reprap.org/wiki/Ultimaker’s_v1.5.7_PCB 58. v6 HotEnd Full Kit - 1.75mm Universal (with Bowden add-on) (24v) [Internet]. [cited 2016 Mar 7]. Available from: http://e3d-online.com/E3D-v6/Full-Kit?product_id=381 59. Heater Cartridge - 24v - 40w [Internet]. [cited 2016 Mar 23]. Available from: http://e3donline.com/Heater-Cartridge-24v-40w?search=24%2040 60. carabiner by venkel – YouMagine � [Internet]. [cited 2016 Mar 20]. Available from: https://www.youmagine.com/designs/carabiner 57

Michael McHugh 10706279 61. BS EN ISO 527-1:2012 Plastics. Determination of tensile properties. General principles. [Internet]. BSI Standards Limited; 2012. Available from: http://elib.tcd.ie/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=e dsbsi&AN=edsbsi.30216857&site=eds-live 62. BS EN ISO 527-2:2012 Plastics. Determination of tensile properties. Test conditions for moulding and extrusion plastics. [Internet]. BSI Standards Limited; 2012. Available from: http://elib.tcd.ie/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=e dsbsi&AN=edsbsi.30216860&site=eds-live

58

Michael McHugh 10706279

59