- Email: [email protected]
Mock-up qualification and prototype manufacture for ITER current leads
Fusion Engineering and Design 96–97 (2015) 388–391
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Mock-up qualification and prototype manufacture for ITER current leads Tingzhi Zhou a,∗ , Kun Lu a , Qingxiang Ran a , Kaizhong Ding a , Hansheng Feng a , Huan Wu a , Chenglian Liu a , Yuntao Song a , Erwu Niu b , Pierre Bauer c , Arnaud Devred c a
Institute of Plasma Physics, Chinese Academy of Sciences, Hefei, China CNDA, Ministry of Science and Technology, Beijing, China c Magnet Division, ITER Organization, Cadarache, France b
h i g h l i g h t s • • • •
Vacuum brazing and electron beam welding qualification. Machine and assembly strategy of fin type heat exchanger. Soldering and joint resistance test of superconducting joint. Pre-preg technology with vacuum bag on insulation.
a r t i c l e
i n f o
Article history: Received 17 September 2014 Received in revised form 15 March 2015 Accepted 23 April 2015 Available online 16 May 2015 Keywords: Current lead Heat exchanger NbTi Bi-2223 Prototype
a b s t r a c t Three types of high temperature superconducting current leads (HTSCL) are designed to carry 68 kA, 55 kA or 10 kA to the ITER magnets. Before the supply of the HTS current lead series, the design and manufacturing process is qualified through mock-ups and prototypes. Seven mock-ups, representing the critical technologies of the current leads, were built and tested successfully in the Institute of Plasma Physics of the Chinese Academy of Sciences (ASIPP) in 2013. After the qualification some design features of the HTS leads were updated. This paper summarizes the qualification through mock-ups. In 2014 ASIPP started the manufacture of the prototypes. The preparation and manufacturing process are also described. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The International Thermonuclear Experimental Reactor (ITER) current leads connect the room temperature busbar and the 4.5 K busbar. 60 HTS current leads transfer 50 GJ of stored magnetic energy into and out of the magnet system. Three types of ITER current leads, Toroidal Field (TF) 68 kA, Poloidal Field (PF)/Central Solenoid (CS) 55 kA and Correction Coil (CC) 10 kA, were designed in 2011 [1]. ASIPP is responsible for all the technologies development and the timely supply of all the current leads. In 2012 ASIPP started the qualification of the technologies for the ITER HTSCL mock-ups and prototypes. In 2013, seven mock-ups (Fig. 1) representing the
∗ Corresponding author. Tel.: +86 18019996936; fax: +86 55165591310. E-mail address: [email protected] (T. Zhou). http://dx.doi.org/10.1016/j.fusengdes.2015.04.050 0920-3796/© 2015 Elsevier B.V. All rights reserved.
critical technologies were qualified: (1) TF HTSCL Electron Beam Welding (EBW) mock-up; (2) TF HTSCL Heat EXchanger (HEX) mock-up; (3) TF HTSCL Low Temperature Superconducting (LTS) mock-up; (4) CC HTSCL instrumentation mock-up; (5) CC HTSCL EBW mock-up; (6) CC HTSCL HEX mock-up; (7) CC HTSCL LTS mock-up. The insulation technologies are still in developing. So the qualification for two insulation mock-ups does not start. Now in 2014 TF and CC prototypes are in manufacturing, except for the insulation. Before the mock-ups and prototypes qualification many documents, including the Manufacture and Inspection Plan (MIP), Manufacture Plan (MP), Quality Plan (QP), including detailed manufacturing procedures were prepared and approved to follow the ITER quality requirements [2]. In 2015, ASIPP plans to finish all the prototypes manufacture and test. After the prototype qualification the project will be pushed into stage III: HTSCL production.
T. Zhou et al. / Fusion Engineering and Design 96–97 (2015) 388–391
389
Fig. 1. HTSCL and mock-ups.
2. Welding and brazing qualification The TIG welding was qualified according to the ISO 15617 standard. The vacuum brazing, for which there is no completely defined standard to support the qualification work, the acceptance criteria were discussed and finally approved by IO. There were two major brazed subassemblies, room temperature (RT) terminal and HTS shunt, to be qualified before the prototype manufacture. Because the brazing quality relates to the mass of the component, the real size terminal and shunt were designed for the qualification. Because different mass of the components resulted in the different temperature increasing rates in furnace the thermocouples were mounted on the components to detect the temperature (Fig. 2). When the temperature difference was more than 5 ◦ C the heating would be paused. Without heating the temperature would be converged after some minutes. All the brazed joints were vacuum or pressure leak tested under real operation pressure conditions. The right photos in Fig. 2 are the cross sections of the brazed joint. No voids was found under 50 times magnifier. The tensile stress tests were performed on the shunt because the gravity will give rise to a bending moment here on the operation when the current lead is horizontally placed. The rupture stress was more than 190 MPa on copper section. This shows that the brazing joint has enough strength. 3. Mock-up qualification 3.1. HEX mock-ups A fin type HEX is designed for the ITER current leads following the positive experiences from CERN [3]. The coolant will flow in zig-zag manner. The assembly tolerance H7/g6 between the HEX and the tube is very critical for the current lead performance. To ease the assembly of the close-fitting tube and heat exchanger, the tube was honed to a high precision finish surface. To avoid sticking during assembly the tube was heated to about 180 ◦ C before the assembly. The diameters were increased by 0.25 mm for CC tube, and 0.43 mm for TF tube, which should allow smooth assembly. Given enough gap after the heating the assembly was finished in 10 s. Because of the softness of the copper and the slim core design after machining, the straightness of the CC HEX core could exceed
Fig. 2. Brazing qualification; left: RT terminal, middle: shunt, right: results of destructive examination.
Fig. 3. Nitrogen pressure drop in the 10 kA mock up.
0.12 mm, mainly due to the deformation under the weight. Since this deformation is hardly avoidable, the fin machining is a challenge requiring special tooling. The lower straightness of the HEX core was also a major reason for the choosing of the tube heated method for assembly. Both TF and CC HEX mock-ups were tested in ASIPP after the successful assembly. The mock-up model was calculated by CERN. In Fig. 3 the black points are the calculation results with finite element method. The red points are the test results. A very good agreement was found between the model and the test results (Fig. 3). It proves that the design and manufacture meets the operation requirements. 3.2. EBW mock-ups The EB welding joints of the HEX to the RT terminal and shunt mainly convey the current. EB welding was chosen because of its: (1) larger welding depth to decrease the current density; (2) narrow heat-affected zone to give least deviation on the copper residual resistance ratio (RRR) value and keep the HTS stacks at a safe temperature during welding. Two short size copper parts with the features of the real HEX were welded to the EBW mock-up (Fig. 4). Thermaxes were affixed to the right position of the mock-up to monitor the temperature on the HTS stacks. The recorded maximum temperature 171 ◦ C was less than the melting point 183 ◦ C of 63Sn–Pb solder for the stacks soldered to the shunt. The tensile strength of the four samples ranged from 194 to 205 MPa. The metallographical analysis, left photo in Fig. 4, proved that the welding depth 36 mm is larger than the defined value 30 mm, and no void was found in the joint.
Fig. 4. Tensile and metallography test of EBW joint.
390
T. Zhou et al. / Fusion Engineering and Design 96–97 (2015) 388–391
Fig. 5. Instrumentation mock-up.
The EBW mock-up gives confidence in the procedure for the future prototype welding. 3.3. Instrumentation mock-up This instrumentation, temperature sensors and voltage taps, is enclosed in the envelope of the current lead. There would be a very large sacrifice to open it for repairing after welding and insulation. So the quality of these sensors and their assembly must be very reliable for long operation. The instrumentation mock-up in Fig. 5 gives the features for the assembly simulation of the sensors. The copper block represents the shunt where the sensors are mounted. The bended Stainless Steel (S.S) pipe represents the internal channel for the instrumentation wires. The temperature sensor, a cylindrical PT100, was assembled into a hole in the Cu block and filled by conductive grease. Then a PC board was installed over it and its two probe wires were soldered to the PC board. Four instrumentation wires were then soldered to the PC board for the connection to the control system. To eliminate thermoelectric force and give mechanical support all the wires were wrapped around heat sinks before passing the wires through the instrumentation channel. The assembly of these sensors and threading of the instrumentation wire through the multi-bend instrumentation channel was simulated in this mock-up. Five thermal shock cycles from LN2 temperature to room temperature were applied on the mock-up. The temperature signal was tested in water and ice mixture after every cycle. 3.4. LTS mock-ups The LTS mock-up is mainly composed of the LTS cable, 5 K terminal and the joint box. The LTS cable is of the Cable-In-ConduitConductor (CICC) type. The Nickel from the sub-cable to be soldered with the 5 K copper terminal was removed by chemical method because all the strands and wires will be soldered. The cable jointed to the joint box had the Nickel removed by a mechanical method because only surface soldering is needed here. The pickling, passivation, water washing and ultrasonic cleaning were applied. The conductivity ratio of the cleaning water tested 0.1–1, which is the acceptance criterion. After chemical or mechanical Nickel removing the cable was stored in a bag filled with dry N2 gas. The LTS cable is linked to the cold copper terminal by soldering technology. The joint resistance is the acceptance criterion of the soldering. To fill the voids between the sub-cable and cold copper terminal the soldering with vacuum pressure impregnating (VPI)
Fig. 6. LTS-Cu joint resistance test profile for CC sub-cable sample.
technology was developed on many joint samples before mock-up soldering. In these samples the sub-cable was solidly anchored in the copper block. The samples were tested in liquid helium. The profile of voltage vs. current is shown in Fig. 6. Less than 2.5 n is required for this joint in the ITER CC lead. Given the maximum joint resistance of the sub-cable samples of 6.58 n (Fig. 6), an equivalent of 0.27 n was calculated for the full CC lead configuration. The samples for TF lead were also tested. 0.027 n, far less than the criterion of 1 n, was achieved there. During the soldering of the sub-cable and the 5 K copper terminal on mock-ups, the accumulated time was less than 2 hours with temperature greater than 200 ◦ C in the CC LTS mock-up. But in the TF LTS mock-up the accumulated time was more than 4 hours. The soldering process optimization and N2 cooling will be used to minimize the time in the future. A maximum cable temperature of less than 250 ◦ C was guaranteed during all times in the welding and soldering of both TF and CC LTS mock-ups. The joint box was machined from an explosion-bonded stainless steel-copper bimetal plate after bending and heat treatment. After machining it was ultrasonically, penetration- and leak-tested. The joint box was TIG welded (Fig. 7) under pressure and soldered with the cable using Sn63Pb37 solder. Argon gas was blown into the box for the protection of the cable during welding. In the mockup soldering the temperatures on the cable and outer surface of the box were measured. The soldering proves that there is ∼5 ◦ C temperature difference between the cable and the box. This recorded temperature will be used as criteria for the soldering of prototype where only box temperature are measured and controlled. 3.5. Insulation technologies The insulation technologies are still in qualification on samples before the application on mock-ups. Some progress was achieved in ASIPP. Pre-preg technology with vacuum bag shown in Fig. 8 from ITER was applied on feeder insulation. The vacuum bag was pumped for the better penetration of insulating cement. The temperature sensors were mounted on the work piece and vacuum bag. T3 and T4 were used for the curing. T1 and T2 were recorded
Fig. 7. LTS mock-up and tig welding under pressure.
T. Zhou et al. / Fusion Engineering and Design 96–97 (2015) 388–391
391
Fig. 8. Lay-up for pre-preg technology with vacuum bag. Fig. 10. Resistance on stacks-shunt soldering joint sample.
Fig. 9. Stacks-shunt soldering sample.
for the comparison. The main mechanical properties were meeting the PA requirement. But the void content was higher than 2%. In order to improve the value of void content, autoclave curing technology was brought in. The sample plate was curried in ∼0.6 MPa pressure of nitrogen condition. The void is 1.1–1.4%. The ultimate stress is 910 MPa on 0 degree, 275 MPa on 90 degree. But it needs a big autoclave for the curing. A U bend was insulated and high voltage tested. The leakage current was 0.18 A. ASIPP will test more samples. Eventually, as the technology matures we will start the mock-up qualification. 4. Prototypes manufacture After the approval of the drawing and documents for both TF and CC HTSCL prototypes the manufacture was started in June 2014. To increase the manufacture efficiency and quality a new fiveaxis computerized numerical control machine and honing machine were procured for the TF prototype. The time for the machining of the HEX core has been decreased to one week from three months when the engine lathe was used. The tube was aging treated after machining. There was about 0.1 mm deformation after 2 weeks. So the tube was finished after honing and aging treatment back and forth for many times. The final assembly tolerance reaches H7/g6. Stack-shunt soldering is the critical technology for the current leads that was qualified on some samples before the soldering on the prototypes. To qualify the soldering joints HTS panels were designed and tested. 2 LTS sub-cables developed from busbar were soldered to the two terminals of a small shunt sample with solder Sn3.8Ag0.7Cu. Three Bi-2223 HTS stacks were soldered into one
groove on the Cu-S.S brazed shunt sample by vacuum soldering technology (Fig. 9). The temperatures on the stacks were recorded to control the heating of the system that will be also the temperature reference for the soldering of the prototypes. The soldered samples were thermally shocked and critical current tested in liquid helium. The voltage vs. current is in Fig. 10. It can be calculated that in real CC lead the LTS-HTS joint will be 0.32 n that is 1/7.8 of the required joint resistance 2.5 n. In real TF lead the LTS-HTS joint will be 0.17 n that is 1/5.9 of the required joint resistance 1 n. So the test shows that the soldering joint can meet the ITER requirement. 5. Conclusion The welding, brazing, soldering and mock-ups representing the critical technologies of the ITER HTS leads were successful qualified after one year cooperation efforts of the IO HTSCL working group, ChiNa Domestic Agency (CNDA), ASIPP and the manufacture supplier. The TF and CC HTSCL prototype are manufactured in parallel. According to the planning these prototypes will be tested earlier next year. The PF current lead prototypes will be also qualified during the middle of next year. Acknowledgments We would like to thank the experts invited by IO for the given technology comments. We also wish to thank Hefei Keye and Juneng Inc. for the document preparation, manufacturing and integration of the mock-ups and prototypes. References [1] A. Ballarino, P. Bauer, Y. Bi, A. Devred, K. Ding, A. Foussat, et al., Design of the HTS current leads for ITER, IEEE Trans. Appl. Supercond. 22 (3) (2012) 4800304. [2] P. Bauer, K. Ding, C. Gung, T. Ichihara, G. Shen, T. Taylor, et al., Assurance of sound manufacturing technology for the ITER current leads, IEEE Trans. Appl. Supercond. 23 (3/2) (2013). [3] A. Ballarino, HTS current leads for the LHC magnet powering system, Physica C 373–376 (2002) 1413–1418.
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Mock-up qualification and prototype manufacture for ITER current leads Tingzhi Zhou a,∗ , Kun Lu a , Qingxiang Ran a , Kaizhong Ding a , Hansheng Feng a , Huan Wu a , Chenglian Liu a , Yuntao Song a , Erwu Niu b , Pierre Bauer c , Arnaud Devred c a
Institute of Plasma Physics, Chinese Academy of Sciences, Hefei, China CNDA, Ministry of Science and Technology, Beijing, China c Magnet Division, ITER Organization, Cadarache, France b
h i g h l i g h t s • • • •
Vacuum brazing and electron beam welding qualification. Machine and assembly strategy of fin type heat exchanger. Soldering and joint resistance test of superconducting joint. Pre-preg technology with vacuum bag on insulation.
a r t i c l e
i n f o
Article history: Received 17 September 2014 Received in revised form 15 March 2015 Accepted 23 April 2015 Available online 16 May 2015 Keywords: Current lead Heat exchanger NbTi Bi-2223 Prototype
a b s t r a c t Three types of high temperature superconducting current leads (HTSCL) are designed to carry 68 kA, 55 kA or 10 kA to the ITER magnets. Before the supply of the HTS current lead series, the design and manufacturing process is qualified through mock-ups and prototypes. Seven mock-ups, representing the critical technologies of the current leads, were built and tested successfully in the Institute of Plasma Physics of the Chinese Academy of Sciences (ASIPP) in 2013. After the qualification some design features of the HTS leads were updated. This paper summarizes the qualification through mock-ups. In 2014 ASIPP started the manufacture of the prototypes. The preparation and manufacturing process are also described. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The International Thermonuclear Experimental Reactor (ITER) current leads connect the room temperature busbar and the 4.5 K busbar. 60 HTS current leads transfer 50 GJ of stored magnetic energy into and out of the magnet system. Three types of ITER current leads, Toroidal Field (TF) 68 kA, Poloidal Field (PF)/Central Solenoid (CS) 55 kA and Correction Coil (CC) 10 kA, were designed in 2011 [1]. ASIPP is responsible for all the technologies development and the timely supply of all the current leads. In 2012 ASIPP started the qualification of the technologies for the ITER HTSCL mock-ups and prototypes. In 2013, seven mock-ups (Fig. 1) representing the
∗ Corresponding author. Tel.: +86 18019996936; fax: +86 55165591310. E-mail address: [email protected] (T. Zhou). http://dx.doi.org/10.1016/j.fusengdes.2015.04.050 0920-3796/© 2015 Elsevier B.V. All rights reserved.
critical technologies were qualified: (1) TF HTSCL Electron Beam Welding (EBW) mock-up; (2) TF HTSCL Heat EXchanger (HEX) mock-up; (3) TF HTSCL Low Temperature Superconducting (LTS) mock-up; (4) CC HTSCL instrumentation mock-up; (5) CC HTSCL EBW mock-up; (6) CC HTSCL HEX mock-up; (7) CC HTSCL LTS mock-up. The insulation technologies are still in developing. So the qualification for two insulation mock-ups does not start. Now in 2014 TF and CC prototypes are in manufacturing, except for the insulation. Before the mock-ups and prototypes qualification many documents, including the Manufacture and Inspection Plan (MIP), Manufacture Plan (MP), Quality Plan (QP), including detailed manufacturing procedures were prepared and approved to follow the ITER quality requirements [2]. In 2015, ASIPP plans to finish all the prototypes manufacture and test. After the prototype qualification the project will be pushed into stage III: HTSCL production.
T. Zhou et al. / Fusion Engineering and Design 96–97 (2015) 388–391
389
Fig. 1. HTSCL and mock-ups.
2. Welding and brazing qualification The TIG welding was qualified according to the ISO 15617 standard. The vacuum brazing, for which there is no completely defined standard to support the qualification work, the acceptance criteria were discussed and finally approved by IO. There were two major brazed subassemblies, room temperature (RT) terminal and HTS shunt, to be qualified before the prototype manufacture. Because the brazing quality relates to the mass of the component, the real size terminal and shunt were designed for the qualification. Because different mass of the components resulted in the different temperature increasing rates in furnace the thermocouples were mounted on the components to detect the temperature (Fig. 2). When the temperature difference was more than 5 ◦ C the heating would be paused. Without heating the temperature would be converged after some minutes. All the brazed joints were vacuum or pressure leak tested under real operation pressure conditions. The right photos in Fig. 2 are the cross sections of the brazed joint. No voids was found under 50 times magnifier. The tensile stress tests were performed on the shunt because the gravity will give rise to a bending moment here on the operation when the current lead is horizontally placed. The rupture stress was more than 190 MPa on copper section. This shows that the brazing joint has enough strength. 3. Mock-up qualification 3.1. HEX mock-ups A fin type HEX is designed for the ITER current leads following the positive experiences from CERN [3]. The coolant will flow in zig-zag manner. The assembly tolerance H7/g6 between the HEX and the tube is very critical for the current lead performance. To ease the assembly of the close-fitting tube and heat exchanger, the tube was honed to a high precision finish surface. To avoid sticking during assembly the tube was heated to about 180 ◦ C before the assembly. The diameters were increased by 0.25 mm for CC tube, and 0.43 mm for TF tube, which should allow smooth assembly. Given enough gap after the heating the assembly was finished in 10 s. Because of the softness of the copper and the slim core design after machining, the straightness of the CC HEX core could exceed
Fig. 2. Brazing qualification; left: RT terminal, middle: shunt, right: results of destructive examination.
Fig. 3. Nitrogen pressure drop in the 10 kA mock up.
0.12 mm, mainly due to the deformation under the weight. Since this deformation is hardly avoidable, the fin machining is a challenge requiring special tooling. The lower straightness of the HEX core was also a major reason for the choosing of the tube heated method for assembly. Both TF and CC HEX mock-ups were tested in ASIPP after the successful assembly. The mock-up model was calculated by CERN. In Fig. 3 the black points are the calculation results with finite element method. The red points are the test results. A very good agreement was found between the model and the test results (Fig. 3). It proves that the design and manufacture meets the operation requirements. 3.2. EBW mock-ups The EB welding joints of the HEX to the RT terminal and shunt mainly convey the current. EB welding was chosen because of its: (1) larger welding depth to decrease the current density; (2) narrow heat-affected zone to give least deviation on the copper residual resistance ratio (RRR) value and keep the HTS stacks at a safe temperature during welding. Two short size copper parts with the features of the real HEX were welded to the EBW mock-up (Fig. 4). Thermaxes were affixed to the right position of the mock-up to monitor the temperature on the HTS stacks. The recorded maximum temperature 171 ◦ C was less than the melting point 183 ◦ C of 63Sn–Pb solder for the stacks soldered to the shunt. The tensile strength of the four samples ranged from 194 to 205 MPa. The metallographical analysis, left photo in Fig. 4, proved that the welding depth 36 mm is larger than the defined value 30 mm, and no void was found in the joint.
Fig. 4. Tensile and metallography test of EBW joint.
390
T. Zhou et al. / Fusion Engineering and Design 96–97 (2015) 388–391
Fig. 5. Instrumentation mock-up.
The EBW mock-up gives confidence in the procedure for the future prototype welding. 3.3. Instrumentation mock-up This instrumentation, temperature sensors and voltage taps, is enclosed in the envelope of the current lead. There would be a very large sacrifice to open it for repairing after welding and insulation. So the quality of these sensors and their assembly must be very reliable for long operation. The instrumentation mock-up in Fig. 5 gives the features for the assembly simulation of the sensors. The copper block represents the shunt where the sensors are mounted. The bended Stainless Steel (S.S) pipe represents the internal channel for the instrumentation wires. The temperature sensor, a cylindrical PT100, was assembled into a hole in the Cu block and filled by conductive grease. Then a PC board was installed over it and its two probe wires were soldered to the PC board. Four instrumentation wires were then soldered to the PC board for the connection to the control system. To eliminate thermoelectric force and give mechanical support all the wires were wrapped around heat sinks before passing the wires through the instrumentation channel. The assembly of these sensors and threading of the instrumentation wire through the multi-bend instrumentation channel was simulated in this mock-up. Five thermal shock cycles from LN2 temperature to room temperature were applied on the mock-up. The temperature signal was tested in water and ice mixture after every cycle. 3.4. LTS mock-ups The LTS mock-up is mainly composed of the LTS cable, 5 K terminal and the joint box. The LTS cable is of the Cable-In-ConduitConductor (CICC) type. The Nickel from the sub-cable to be soldered with the 5 K copper terminal was removed by chemical method because all the strands and wires will be soldered. The cable jointed to the joint box had the Nickel removed by a mechanical method because only surface soldering is needed here. The pickling, passivation, water washing and ultrasonic cleaning were applied. The conductivity ratio of the cleaning water tested 0.1–1, which is the acceptance criterion. After chemical or mechanical Nickel removing the cable was stored in a bag filled with dry N2 gas. The LTS cable is linked to the cold copper terminal by soldering technology. The joint resistance is the acceptance criterion of the soldering. To fill the voids between the sub-cable and cold copper terminal the soldering with vacuum pressure impregnating (VPI)
Fig. 6. LTS-Cu joint resistance test profile for CC sub-cable sample.
technology was developed on many joint samples before mock-up soldering. In these samples the sub-cable was solidly anchored in the copper block. The samples were tested in liquid helium. The profile of voltage vs. current is shown in Fig. 6. Less than 2.5 n is required for this joint in the ITER CC lead. Given the maximum joint resistance of the sub-cable samples of 6.58 n (Fig. 6), an equivalent of 0.27 n was calculated for the full CC lead configuration. The samples for TF lead were also tested. 0.027 n, far less than the criterion of 1 n, was achieved there. During the soldering of the sub-cable and the 5 K copper terminal on mock-ups, the accumulated time was less than 2 hours with temperature greater than 200 ◦ C in the CC LTS mock-up. But in the TF LTS mock-up the accumulated time was more than 4 hours. The soldering process optimization and N2 cooling will be used to minimize the time in the future. A maximum cable temperature of less than 250 ◦ C was guaranteed during all times in the welding and soldering of both TF and CC LTS mock-ups. The joint box was machined from an explosion-bonded stainless steel-copper bimetal plate after bending and heat treatment. After machining it was ultrasonically, penetration- and leak-tested. The joint box was TIG welded (Fig. 7) under pressure and soldered with the cable using Sn63Pb37 solder. Argon gas was blown into the box for the protection of the cable during welding. In the mockup soldering the temperatures on the cable and outer surface of the box were measured. The soldering proves that there is ∼5 ◦ C temperature difference between the cable and the box. This recorded temperature will be used as criteria for the soldering of prototype where only box temperature are measured and controlled. 3.5. Insulation technologies The insulation technologies are still in qualification on samples before the application on mock-ups. Some progress was achieved in ASIPP. Pre-preg technology with vacuum bag shown in Fig. 8 from ITER was applied on feeder insulation. The vacuum bag was pumped for the better penetration of insulating cement. The temperature sensors were mounted on the work piece and vacuum bag. T3 and T4 were used for the curing. T1 and T2 were recorded
Fig. 7. LTS mock-up and tig welding under pressure.
T. Zhou et al. / Fusion Engineering and Design 96–97 (2015) 388–391
391
Fig. 8. Lay-up for pre-preg technology with vacuum bag. Fig. 10. Resistance on stacks-shunt soldering joint sample.
Fig. 9. Stacks-shunt soldering sample.
for the comparison. The main mechanical properties were meeting the PA requirement. But the void content was higher than 2%. In order to improve the value of void content, autoclave curing technology was brought in. The sample plate was curried in ∼0.6 MPa pressure of nitrogen condition. The void is 1.1–1.4%. The ultimate stress is 910 MPa on 0 degree, 275 MPa on 90 degree. But it needs a big autoclave for the curing. A U bend was insulated and high voltage tested. The leakage current was 0.18 A. ASIPP will test more samples. Eventually, as the technology matures we will start the mock-up qualification. 4. Prototypes manufacture After the approval of the drawing and documents for both TF and CC HTSCL prototypes the manufacture was started in June 2014. To increase the manufacture efficiency and quality a new fiveaxis computerized numerical control machine and honing machine were procured for the TF prototype. The time for the machining of the HEX core has been decreased to one week from three months when the engine lathe was used. The tube was aging treated after machining. There was about 0.1 mm deformation after 2 weeks. So the tube was finished after honing and aging treatment back and forth for many times. The final assembly tolerance reaches H7/g6. Stack-shunt soldering is the critical technology for the current leads that was qualified on some samples before the soldering on the prototypes. To qualify the soldering joints HTS panels were designed and tested. 2 LTS sub-cables developed from busbar were soldered to the two terminals of a small shunt sample with solder Sn3.8Ag0.7Cu. Three Bi-2223 HTS stacks were soldered into one
groove on the Cu-S.S brazed shunt sample by vacuum soldering technology (Fig. 9). The temperatures on the stacks were recorded to control the heating of the system that will be also the temperature reference for the soldering of the prototypes. The soldered samples were thermally shocked and critical current tested in liquid helium. The voltage vs. current is in Fig. 10. It can be calculated that in real CC lead the LTS-HTS joint will be 0.32 n that is 1/7.8 of the required joint resistance 2.5 n. In real TF lead the LTS-HTS joint will be 0.17 n that is 1/5.9 of the required joint resistance 1 n. So the test shows that the soldering joint can meet the ITER requirement. 5. Conclusion The welding, brazing, soldering and mock-ups representing the critical technologies of the ITER HTS leads were successful qualified after one year cooperation efforts of the IO HTSCL working group, ChiNa Domestic Agency (CNDA), ASIPP and the manufacture supplier. The TF and CC HTSCL prototype are manufactured in parallel. According to the planning these prototypes will be tested earlier next year. The PF current lead prototypes will be also qualified during the middle of next year. Acknowledgments We would like to thank the experts invited by IO for the given technology comments. We also wish to thank Hefei Keye and Juneng Inc. for the document preparation, manufacturing and integration of the mock-ups and prototypes. References [1] A. Ballarino, P. Bauer, Y. Bi, A. Devred, K. Ding, A. Foussat, et al., Design of the HTS current leads for ITER, IEEE Trans. Appl. Supercond. 22 (3) (2012) 4800304. [2] P. Bauer, K. Ding, C. Gung, T. Ichihara, G. Shen, T. Taylor, et al., Assurance of sound manufacturing technology for the ITER current leads, IEEE Trans. Appl. Supercond. 23 (3/2) (2013). [3] A. Ballarino, HTS current leads for the LHC magnet powering system, Physica C 373–376 (2002) 1413–1418.