Additive Manufacturing for Xband applications Alexej Grudiev 5022014

Additive Manufacturing for Xband applications Alexej Grudiev 5/02/2014 CLIC 14 workshop

Acknowledgements • BE-RF • Miriam Colling • Alexej Grudiev • EN-MME • Said Atieh • Ofelia Capatina • Floriane Leaux • Raphael Leuxe • Thomas Sahner • Ignacio Santillana • TE-VSC • Sergio Calatroni • Ivo Wevers

Additive manufacturing Process 1. Model designed in CAD 2. CAD file sent to additive manufacturing system 3. Model divided into slices 4. 3 D product sculpted from powdered material layer by layer using the specified technique EOS – SLS, http: //www. eos. info/additive_manufacturing/for_technology_interested

Number of good reasons to try

Typical materials

Technical data Ti 64 DC electrical conductivity: 600000 S/m, two times lower than stainless steel • Relatively low accuracy • Big roughness (much larger than skin depth) • Low DC conductivity Obvious RF application is a broad-band all-metal dry RF load

High power/E-field performance of Ti DC breakdown thresholds 30 GHz high power performance, PAC 2007

Three stages of the project 1. Low power prototype for material and fabrication characterization 1. 2. 3. 4. DC conductivity and RF losses UHV compatibility: leak tightness and outgassing Shape accuracy and Roughness ere h e r a Mechanical strength and Metallurgy We 2. Prototypes for high power tests 1. Integration of cooling 2. High power performance 3. Design of the RF load 4. RF load prototype

Prototype Design Ø Prototype modelled in HFSS Ø WR 90 end tapered to 13 mm by 2 mm middle 200 mm

Waveguide manufacturing methods Waveguide 1: EOS – Selective laser sintering Waveguide 2: Grenoble INP – Electron beam melting 5 Waveguides 3, 4 and 5: Concept – Selective laser melting Ø Length: 20 cm Ø Material: Titanium alloy

RF measurements using VNA • Obtained data for S(1, 1), S(1, 2), S(2, 1) and S(2, 2) parameters for each of the five waveguides • Measurements required careful handling - movement in cables cause readings to change • Measurements repeated three times for each waveguide for reliable results

RF Results Comparison of S(1, 1) parameter 0 -5 -10 -15 S(1, 1)/d. B -20 -25 -30 HFSS simulation -35 Waveguide 1 Waveguide 2 -40 Waveguide 3 Waveguide 4 -45 Waveguide 5 -50 11, 3 11, 5 11, 7 11, 9 Frequency/GHz 12, 1 12, 3 12, 5 12, 7

RF Results Comparison of S(1, 2) parameter 0 -5 -10 S(1, 2)/d. B -15 HFSS simulation -20 Waveguide 1 Waveguide 2 -25 Waveguide 3 Waveguide 4 Waveguide 5 -30 -35 11, 3 11, 5 11, 7 11, 9 Frequency/GHz 12, 1 12, 3 12, 5 12, 7

DC conductivity measurements Titanium alloy conductivity: 6 E+5 S/m Two types of DC measurements: 1. Four probes in contact with middle section 6 mm 2. Two probes in contact with middle while clamps on flange provide voltage difference 62 mm

DC Results Titanium alloy conductivity: 6 E+5 S/m Method 1 Method 2 Waveguide No. Conductivity/ S/m Waveguide No. Conductivity [S/m] 1 - EOS 641025 1 - EOS 716093 2 - INP Grenoble 368595 2 - INP Grenoble 480179 3 - Concept laser 496771 3 - Concept laser 571880 4 - Concept laser 500000 4 - Concept laser 557176 5 - Concept laser 522739 5 - Concept laser 580343 Apply all DC cond. to HFSS waveguide and obtain values for all 3 parameters in each case aqw Nominal HFSS values: ‘aqw’=14. 2 mm ‘a’=13 mm Roughness=0µm a

Determining parameters Determining 'aqw' of waveguide 1 (cond=641025 S/m) 0 -10 S(1, 1)/d. B -20 -30 -40 HFSS Nominal: aqw='14. 2 mm' Measured: Waveguide 1 -50 Matched: aqw='13. 9 mm' -60 11, 3 11, 5 11, 7 11, 9 Frequency/GHz 12, 1 12, 3 12, 5 12, 7

Determining parameters Determining 'a' and roughness of waveguide 1 (cond=641025 S/m) 0 -5 S(1, 2)/d. B -10 -15 HFSS Nominal: a='13 mm' aqw='14. 2 mm' rough='0 um' Measured: Waveguide 1 -20 Matched: a='12. 91 mm' aqw='13. 9 mm' rough='50 um' -25 11, 3 11, 5 11, 7 11, 9 Frequency/GHz 12, 1 12, 3 12, 5 12, 7

Table of parameters HFSS Nominal: a=‘ 13 mm’ aqw=‘ 14. 2 mm’ rough=‘ 0 um’ Waveguide No. DC cond. (S/m) (method 1) ‘a’ (mm) ‘aqw’ (mm) HFSS roughness (µm) d. B % difference at 12 GHz 1 -EOS 641025 12. 91 13. 9 >50 -13. 12 2 -Grenoble 368595 12. 74 14. 0 >50 -2. 78 3 -Concept 496771 12. 75 14. 0 >50 -33. 90 4 -Concept 500000 12. 79 14. 0 >50 -19. 71 5 -Concept 522739 12. 80 14. 1 >50 -88. 03 Waveguide No. DC cond. (S/m) (method 2) ‘a’ (mm) ‘aqw’ (mm) HFSS roughness (µm) d. B % difference at 12 GHz 1 -EOS 716093 12. 90 14. 0 >50 -20. 29 2 -Grenoble 480179 12. 74 14. 0 >50 -15. 56 3 -Concept 571880 12. 74 14. 0 >50 -40. 02 4 -Concept 557176 12. 79 14. 0 >50 -25. 79 5 -Concept 580343 12. 80 14. 1 >50 -95. 53 Tables show: • Values for each modified parameter: ‘a’ , ‘aqw’ and HFSS roughness • 100 -300 micron differences • Change in parameters required to provide HFSS results which agree with those produced by the VNA for each waveguide

Metrology Microtomography – X-ray non destructive testing Radioscopic image acquisition 1 2 Post processing 3 Images: RX solutions gallery http: //www. rxsolutions. fr/#!untitled/zoom/cjjm/i 47 og 1 3 D reconstruction

Metrology Results Ø Blue lines show lack of material Ø Red lines show excess material

Metrology Results Ø First three waveguides were measured using microtomography to determine dimensions ‘a’ and ‘aqw’ Ø Measured ‘a’ at 3 points and ‘aqw’ and 2 points and an average was found Ø Waveguide 1 and 3 < 100 micron difference from nominal RF+DC measurements: Waveguide No. DC cond. (S/m) (method 1) ‘a’ (mm) ‘aqw’ (mm) HFSS roughness (µm) d. B % difference at 12 GHz 1 -EOS 641025 -13. 12 368595 >50 -2. 78 3 -Concept 496771 13. 9 14. 0 >50 2 -Grenoble 12. 91 12. 74 12. 75 >50 -33. 90

Vacuum Ø Waveguide 1, 3, 4 and 5 are leak tight, OK for UHV Ø Waveguide 2 was not be able to pump down due to presence of small holes

Mechanical testing and metallographic observations Waveguide 2 Waveguide 3 After etching Before etching Waveguide 1 W 1 shows least porosity W 2 shows large porosity W 3 shows different microstructure

Summary of the results obtained after tensile tests of the samples WG # Company Emod (GPa) Rp 0. 2 (MPa) A% UTS (MPa) 1 3 T 112 ± 1 1097 ± 8 2± 0 1139. 8 ± 3. 0 2 IPN 51 ± 0 830 ± 36 11 ± 2 904. 8 ± 20. 4 3 Concept Laser 108 ± 1 825 ± 11 12 ± 3 893. 4 ± 10. 7 120 910 - 958 12 -16 972 - 1030 Table value for standard material (www. matweb. com) OK OK

Summary • Laser melting fabrication is validated for two manufacturers • EBM fabrication requires some improvements • Next step: