- Email: [email protected]
ITER Vacuum Vessel design and construction
Fusion Engineering and Design 87 (2012) 828–835
Contents lists available at SciVerse ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
ITER Vacuum Vessel design and construction夽 K. Ioki a,∗ , C.H. Choi a , E. Daly a , S. Dani a , J. Davis a , B. Giraud a , Y. Gribov a , C. Hamlyn-Harris a , L. Jones b , C. Jun a , B.C. Kim c , E. Kuzmin d , R. Le Barbier a , J.-M. Martinez a , H. Pathak e , J. Preble a , J. Reich a , J.W. Sa c , A. Terasawa a , Yu. Utin a , X. Wang a , S. Wu a a
ITER Organization, Route de Vinon sur Verdon, 13115 Saint Paul Lez Durance, France F4E, c/Josep Pla, n.2, Torres Diagonal Litoral, Edificio B3, E-08019 Barcelona, Spain c NFRI, 52 Yeoeundong Yuseonggu, Daejeon 305-333, South Korea d NTC “Sintez”, Efremov Inst., 189631 Metallostroy, St. Petersburg, Russia e ITER-India, A-29, GIDC Electronic Estate, Sector -25, Gandhinagar 382025, India b
a r t i c l e
i n f o
Article history: Available online 15 March 2012 Keywords: Vacuum Vessel Structural analysis Support In-wall shielding Material Dynamic test
a b s t r a c t After implementing a few design modifications (referred to as the “Modified Reference Design”) in 2009, the Vacuum Vessel (VV) design had been stabilized. The VV design is being finalized, including interface components such as support rails and feedthroughs for the in-vessel coils. It is necessary to make adjustments to the locations of the blanket supports and manifolds to accommodate design modifications to the in-vessel coils. The VV support design is also being finalized considering a structural simplification. Design of the in-wall shielding (IWS) has progressed, considering the assembly methods and the required tolerances. The detailed layout of ferritic steel plates and borated steel plates was optimized based on the toroidal field ripple analysis. A dynamic test on the inter-modular key to support the blanket modules was performed to measure the dynamic amplification factor (DAF). An R&D program has started to select and qualify the welding and cutting processes for the port flange lip seal. The ITER VV material 316 L(N) IG was already qualified and the Modified Reference Design was approved by the Agreed Notified Body (ANB) in accordance with the Nuclear Pressure Equipment Order procedure. © 2012 Published by Elsevier B.V.
1. Introduction The Vacuum Vessel (VV) is a key component of the ITER facility. The primary functions of the VV are to provide the first confinement barrier, withstand postulated accidents without losing confinement, remove the nuclear heating, provide a boundary consistent with the generation and maintenance of a high quality vacuum and support in-vessel components and their loads. This paper explains the status of the VV design and R&D, and preparation for its construction. 2. VV design and other activities 2.1. VV design completion The VV remains a double-walled torus-shaped structure composed of SS 316 L(N)-IG (ITER Grade) [1]. The main vessel consists
夽 The views and opinions expressed herein do not necessarily reflect those of the ITER Organization. ∗ Corresponding author. Tel.: +49 89 3299 4428; fax: +49 89 3299 4422. E-mail address: [email protected] (K. Ioki). 0920-3796/$ – see front matter © 2012 Published by Elsevier B.V. doi:10.1016/j.fusengdes.2012.02.023
of an inner and outer shell, poloidal and toroidal ribs and inwall shielding (IWS). Flexible support housings (FSHs) and keys are welded to the vessel shells (as shown in Fig. 1) [2], since the blanket modules are directly supported by the VV. Electron beam (EB) welding will be used as much as possible for joints between the inner shell and these structures. The layout of welds on the inner and outer shells is very tight considering accessibility for welding and non-destructive examination requirements defined in the design codes. The triangular support plays an important role in the plasma vertical stability control during minor disruptions, and its position and configuration is optimized based on the plasma vertical stability analysis. The VV design (3D CAD models) was distributed to the Domestic Agencies (DAs) and their suppliers in Nov 2010 for their development of the manufacturing design. In 2010, several design modifications/optimizations were made for the VV, as explained below (a)–(f) (see Fig. 2). The designs of interfacing components drove most of these changes. (a) Supporting rails for in-vessel coils (IVC) and Blanket manifolds—Layout of the support rails was optimized and the rail structure was simplified. Additional local tolerance requirements were introduced.
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
829
Fig. 3. In-wall shielding (left: 20◦ sector, right: whole structure).
2.2. Other activities for VV design, analysis and R&D
Fig. 1. ITER Vacuum Vessel.
(b) Inter-modular keys and stub keys—The keys supporting blanket modules were reinforced to withstand higher EM loads. (c) Upper port plug supports (#4–7)—Support structures for port shield blocks were added considering difficulties in the access from the neutral beam injector cell. (d) Port stub configuration—Port stub configuration (upper ports) was optimized/modified based on an Assembly Group request. (e) Cooling/drain lines of triangular supports (close to the sector edges)—A design modification is required to avoid a conflict with the toroidal transport of the divertor cassette. (f) Lower port—One of the lower ports was modified to avoid interference with a PF4 feeder.
Fig. 2. Design modifications explained in Section 2.1.
2.2.1. IWS design and analysis The space between the double walls will be mostly filled with shield structures mainly made of an austenitic stainless steel containing boron. To minimize induced eddy currents, the IWS is made of ∼8000 blocks (see Fig. 3). The addition of boron to SS 304 was adopted to improve neutron shielding efficiency. Two types of borated steels are used: (i) Steel type 304B4 with 1.0–1.24% of boron for the outboard region of the VV, (ii) Steel type 304B7 with 1.75–2.25 wt.% of boron for the inboard region. This steel (SS304B7) has low ductility (minimum total elongation of 6%) and low fracture toughness. However, this issue will be mitigated by the selection of a powder metallurgy (P/M) route for manufacturing of SS304B7. A ferritic stainless steel type 430 is used under the TF coils in the outboard area to reduce the toroidal field ripple. This steel has a saturated magnetization at ∼1.6 T. The borated and ferritic steels have high corrosion resistance in water and acceptable fabrication characteristics. The toroidal field ripple calculation was completed and the maximum ripple in the plasma region was reduced from 1.16% to 0.30% (in the nominal toroidal field) and −0.67% (in half of the nominal toroidal field) by the ferromagnetic inserts in the regular sectors (see Fig. 4). Electromagnetic (EM) loads on IWS blocks due to the magnetization effect have been analyzed using a 3D solid model (see Fig. 5). The maximum load per IWS block is 200 kN (toroidal direction), 45 kN (poloidal direction) and 30 kN (radial direction). Detailed IWS design has progressed also taking into account the VV design modifications. (i) IWS blocks have been shifted according
Fig. 4. Distribution of toroidal field ripple (left: nominal toroidal field, right: half of nominal toroidal field).
830
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
Fig. 5. Analysis model (left), and magnetization electromagnetic load on each IWS block (right).
to the new VV inner shell location (MDR: Reference Design). (ii) The IWS design has been modified for consistency with design changes in the Vacuum Vessel such as upper ports and flexible support. (iii) The gaps between IWS blocks have been adjusted based on nuclear calculation results. The basic scheme of IWS assembly stays the same (see Fig. 6(a)) and it also shows compatibility with the Tshape adapters of the VV poloidal/toroidal ribs for welding with the outer shell (see Fig. 6(b) and (c)). Tools for IWS block assembly have been designed. The lifting tool includes an eye bolt, a cover cap and a nut (see Fig. 7). Before lifting the block, the eye bolt is screwed with the IWS bolt head. Then the cover cap covers the IWS
Fig. 7. IWS blocks in outboard and lifting tools.
bolt head and the nut is fixed onto the cover cap. The cover cap and nut give an additional structural margin to support the IWS block against the torque generated by gravity. Structural analysis has been performed to confirm the structural integrity of the IWS. In the most severe case (CAT III event: EM and seismic loads), limit analysis was performed. The calculated load factor (LF) was 2.7 where the required minimum value of LF for category III events is 1.25. This shows that there is sufficient structural margin in the IWS design. 2.2.2. VV support Design of the dual-hinge type VV supports has progressed. The number of supports is nine and they are connected to an even number of lower ports (see Fig. 8). The dual hinge support has a primary hinge and a secondary hinge. The primary hinge is connected with upper and lower blocks through dowels (200 mm diameter) and provides all constraints for vertical downwards, upwards, and toroidal forces as well as radial & toroidal moment. The secondary hinge contacts upper and lower blocks and provides constraint for vertical downward forces. The upper block is connected to the port basement by bolts and the lower block is connected to the cryostat pedestal ring by bolts. Structural analysis has been performed and limit analysis shows a load factor of 2.0 where the RCC-MR code [3] requirement is 1.25 as the minimum. Global structural analysis results (including a fault mode) of the VV supports are shown in Fig. 9 and local support analysis results are in Fig. 10. R&D testing was performed for the selection and qualification of the dowel coating materials (see Fig. 11). MoS2 and WS2 coating materials were tested under a vacuum condition under maximum pressure 200 MPa at 100–200 ◦ C, and the measured friction factor was 0.14–0.28.
Fig. 6. Assembly procedure of IWS blocks. (a) Inboard, (b) Poloidal T-rib Area, and (c) Toroidal T-rib Area.
2.2.3. Dynamic test of keys Electromagnetic loads on each blanket module are supported by the VV. The radial torque and poloidal force are supported by the inter-modular key in the inboard region. There is a small gap (<1 mm) between the key and blanket module. This gap facilitates the installation of the blanket modules. The reaction force on the key from the transient EM load on the blanket is amplified and the
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
831
Fig. 10. Local support analysis results (CAT III events).
Fig. 11. (i) Friction testing for VV support dowel and (ii) MoS2 surface after testing.
Fig. 8. (a) VV lower port and support and (b) VV support.
amplification factor is defined as the Dynamic Amplification Factor (DAF). An R&D program was launched to measure the DAF. As shown in Fig. 12, the test set-up consists of a drop mass {1} which impacts on the free end of a large beam . This beam is fixed to a ‘hinge’ at the other end {4}. The ‘hinge’ is a steel plate which has a relatively low bending stiffness and bends very little during the test. The beam is used to amplify and shape the force pulse, generated by the drop mass, which is applied on the key {3}. Furthermore, an auxiliary mass {5} is mounted on the beam above the key fixture. This mass represents the mass of a blanket module. The DAF in halo current load conditions was measured by changing the gap height, the contact area of impact on the key and the duration of the input pulse. A nominal duration was 50 ms and nominal amplitude was 1.4 MN. The maximum measured DAF in halo current load conditions was 1.36. A different test set-up is prepared for DAF measurements during major disruptions due to the nature of the applied load (a torque). 2.2.4. Lip seal welding and cutting for port flange joint An R&D program has started to select and qualify the welding and cutting process for the port flange lip seal. It also includes
Fig. 9. Global analysis results (including fault modes) of the VV supports. Fig. 12. Dynamic load test on key.
832
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
Table 1 Comparison of TIG and laser welding methods for lip seal of port flange.
Practically achievable penetration Welding speed Allowable deviations/misalignments Air gap Vertical misalignment of lip plates Welding head distance to lip joint Welding head transversal position Welding head angular position a b c
Laser welding
TIG weldings
>3 mm 1–2 m/min
2–2.5 mma 250–500 mm/mim
<0.2 mm 0.5 mm ±2–3 mm ±0.2–0.3 mmb ±3◦ –4◦
0.5–1 mm 1 min ±0.2–0.3 mmc ±0.5–1 mm ±5◦ –10◦
May be increased with current pulsing techniques. May be increased with special welding head optics or by using a linear scanner. Requires on-line sensor system. Fig. 15. Lip-seal cutting equipment by dry milling (left) and chip collecting system (right).
for Laser welding. The 2 mm thick lips were tack welded before each robot driven welding. TIG and Laser mock-ups were welded and cut, respectively. 2 and 4 times. The mock-ups were both leak tested to verify the leak rate requirement (<1.0 × 10−10 Pa m3 /s). Dry milling was selected as the cutting method in this R&D program (see Fig. 15). Several runs are necessary to remove a 3 mm deep weld. The tool is equipped with a chip collecting system which satisfies the cleanliness requirements.
Fig. 13. Profiles of welds in lips. (a) Laser welding, Focus position: −3 mm, (b) Laser welding, Focus position: 0 mm, (c) TIG welding.
the development of NDT methods of welds and the demonstration of the adaptability of a robot arm by using real scale mock-ups of the lip seal. TIG welding and YAG Laser welding were selected based on initial tests and an assessment. The two welding processes were compared (see Table 1 and Fig. 13). Two full scale mockups were tested: one for TIG welding (see Fig. 14) and the other
Fig. 14. Lip seal welding on full-scale mock-up.
2.2.5. Thermal hydraulic analysis Thermal hydraulic analysis is under way for the main vessel and ports in collaboration with the DAs. A preliminary result for the main vessel is shown in Fig. 16. In most of the areas, the temperature is below 120 ◦ C and the heat transfer coefficient is much higher than the minimum requirement (500 W/m2 /K). Further analysis is being performed to improve cooling performance in specific areas in the triangular support or near ports. 2.2.6. Design of in-vessel coils (ELM/VS coils) and interfaces with the VV The ELM control coils include three rows of 9 sets corresponding to toroidally 40◦ spanning VV sectors (see Fig. 17). Each row of the ELM control coils is independently installed and supported to the VV inner surface. The VS coils are prepared inside the VV as toroidally continuous coils along the upper and lower legs. Based on
Fig. 16. Thermal hydraulic analysis result of main vessel (temperature distribution of inner shell surface).
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
Fig. 17. In-vessel (ELM/VS) coils installed on VV sectors (coils are shown in green). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
R&D results on irradiation damage of electrical insulation material, the coil design was changed to the mineral-insulated (MI) cable type. ELM and VS coil conductors are attached to a central support structure and fixed to the VV by its support rail. The ELM coils (including coil feeders) have been designed to be remotely maintainable. The coils are bolted to coil supports welded to the VV inner shell. A square cross-section rail (instead of a L-shaped rail) is used where possible, considering that the weld structures shall satisfy RCC-MR [3] requirements for weld type and non-destructive examination (Ultrasonic Testing is to be used) on welds in the VV inner shell. The upper VS coil was shifted upward above the upper port to optimize the feeder routing and interfaces with blanket modules. The ELM and upper VS coil feeders are routed out of the Vacuum Vessel through the upper ports and the lower VS coil feeders are routed out of the vessel through two small dedicated ports. A blanket manifold design was developed to fit with the ELM/VS coil design into the outboard region. A blanket manifold consisting of multiple pipes is used in the outboard region while retaining the option for the replacement by remote handling. 3. Activities towards the VV construction 3.1. Conformity assessment of the VV design The ITER Vacuum Vessel (VV) is nuclear pressure equipment according to the French Decree of December 12th 2005 (ESPN). Therefore the VV manufacturer must demonstrate that the applicable essential safety requirements and radioprotection requirements are satisfied. The procedure for licensing the ITER VV according to ESPN is in progress and conformity assessment is being performed by AIB-Vinc¸otte International selected by ITER Organization as the Agreed Notified Body (ANB). The ITER Organization (IO) is the Manufacturer (under the ESPN definition) of the complete VV (double shell structure and ports) and responsible for design, manufacture and conformity assessment. The VV is an assembly of ESPN components including sectors and ports. As part of the Phase I, devoted to design of this process, IO submitted to the ANB a set of documents (design description, load specification, hazard analysis, stress report, drawing package, radioprotection guide, approach for pressure test, particular material appraisals, material procurement specifications, etc.) and gained a preliminary design approval and approval of material specifications. Due to design modifications, a set of complementary documents on the Modified Reference Design were further assessed by ANB to check that the design changes are
Fig. 18. Vacuum Vessel mock-ups (KODA/HHI).
833
834
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
Fig. 19. Lower port mock-up (KODA/HHI) (upper) and upper port mock-up (RFDA) (lower).
Fig. 20. Assembled inboard IWS mock-ups (upper) and outboard IWS mock-ups (lower) (INDA/ATL).
structurally acceptable. The process continues with a full update of the documentation package necessary before starting the first fabrication activities of the VV components.
assembling four individual poloidal segments and then final assembly of these together into the sector. Manufacturing and material qualification documents required before the start of fabrication were prepared by DAs and VV main suppliers and reviewed by IO and the ANB. Various mock-ups are in progress by Korea and Europe for the main vessel sectors, by Korea and Russia for the lower port and the upper port, respectively, and by India for the In-Wall Shielding to establish in detail manufacturing sequences and validate in particular tolerance aspects (see Figs. 18–20).
3.2. VV material qualification The procurements for the ITER Vacuum Vessel and Ports will be “in-kind” and materials for these components are produced by different DAs. In accordance with the regulatory requirements and quality requirements for operation, common material specifications were prepared in collaboration with the DAs and their industries. Documents to qualify plate/forging material manufacturing by each material supplier selected by the DAs and their main fabricators are submitted to IO and the ANB for review and approval. Several material suppliers were qualified and fabrication has started for plates for KO DA after approval of all required documentation. 3.3. Manufacturing design by DAs and VV suppliers Conceptual or preliminary manufacturing designs have been prepared by DAs and VV main suppliers and have been reviewed by IO and the ANB. Structural analysis has been performed to justify various proposed manufacturing design details including reduction of the minimum shell thickness of the VV. When the manufacturing design features require design changes, Project Change Requests (PCR) or Deviation Requests (DR) are submitted and assessed by IO and the ANB. A PCR or DR impacting structural behavior is required to include dedicated stress reports. Prior to the start of manufacturing, Manufacturing Designs shall be approved by the ANB and are planned by Korea and Europe for late this year. In Korea and in Europe, the same basic sequence will be used for the assembly of the sectors. Sector assembly will be accomplished by first
4. Conclusion Several design modifications were made during 2010 and 2011 according to requests from in-vessel coil interfaces, blanket module supports and manifolds, other interfaces, and the DAs and their suppliers. Designs of the IWS and VV supports have progressed including detailed electromagnetic, thermal and structural analyses. The ITER fabrication technology has progressed based on results of the L-3 project and additional R&D [4–7] by the Domestic Agencies. The fabricated sectors and ports will be transported to the ITER site in 2015–2016 for assembly. References [1] K. Ioki, P. Barabaschi, V. Barabash, W. Daenner, F. Elio, et al., Design improvements and R&D achievements for VV and in-vessel components towards ITER construction, Nuclear Fusion 43 (2003) 268–273. [2] K. Ioki, V. Barabash, J. Cordier, M. Enoeda, G. Federici, et al., ITER Vacuum Vessel, in-vessel components and plasma facing material, Fusion Engineering and Design 83 (2008) 787–794. [3] Design and Construction Rules for Mechanical Components of Nuclear Installation, RCC-MR, French Association for the Design, Construction and Operating Supervision of the Equipment for Electro-Nuclear boilers (AFCEN), 2007.
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835 [4] K. Koizumi, M. Nakahira, Y. Itou, E. Tada, G. Johnson, et al., Design and development of the ITER vacuum vessel, Fusion Engineering and Design 41 (1998) 299. [5] L. Jones, J.-F. Arbogast, A. Bayon, S. Galvan, B. Giraud, et al., European preparations for the ITER VV sectors manufacture, Fusion Engineering and Design 86 (2011) 2096–2100.
835
[6] L. Jones, J. Duhovnik, M. Ginola, J. Huttunen, K. Ioki, et al., Results from iter vacuum vessel sector manufacturing development in Europe, Fusion Engineering and Design 82 (2007) 1942. [7] B.Y. Kim, H.J. Ahn, M.S. Ha, Y.K. Kim, H.S. Kim, et al., Design analysis of the hinge support for the ITER vacuum vessel, Fusion Engineering and Design 86 (2011) 2003–2007.
Contents lists available at SciVerse ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
ITER Vacuum Vessel design and construction夽 K. Ioki a,∗ , C.H. Choi a , E. Daly a , S. Dani a , J. Davis a , B. Giraud a , Y. Gribov a , C. Hamlyn-Harris a , L. Jones b , C. Jun a , B.C. Kim c , E. Kuzmin d , R. Le Barbier a , J.-M. Martinez a , H. Pathak e , J. Preble a , J. Reich a , J.W. Sa c , A. Terasawa a , Yu. Utin a , X. Wang a , S. Wu a a
ITER Organization, Route de Vinon sur Verdon, 13115 Saint Paul Lez Durance, France F4E, c/Josep Pla, n.2, Torres Diagonal Litoral, Edificio B3, E-08019 Barcelona, Spain c NFRI, 52 Yeoeundong Yuseonggu, Daejeon 305-333, South Korea d NTC “Sintez”, Efremov Inst., 189631 Metallostroy, St. Petersburg, Russia e ITER-India, A-29, GIDC Electronic Estate, Sector -25, Gandhinagar 382025, India b
a r t i c l e
i n f o
Article history: Available online 15 March 2012 Keywords: Vacuum Vessel Structural analysis Support In-wall shielding Material Dynamic test
a b s t r a c t After implementing a few design modifications (referred to as the “Modified Reference Design”) in 2009, the Vacuum Vessel (VV) design had been stabilized. The VV design is being finalized, including interface components such as support rails and feedthroughs for the in-vessel coils. It is necessary to make adjustments to the locations of the blanket supports and manifolds to accommodate design modifications to the in-vessel coils. The VV support design is also being finalized considering a structural simplification. Design of the in-wall shielding (IWS) has progressed, considering the assembly methods and the required tolerances. The detailed layout of ferritic steel plates and borated steel plates was optimized based on the toroidal field ripple analysis. A dynamic test on the inter-modular key to support the blanket modules was performed to measure the dynamic amplification factor (DAF). An R&D program has started to select and qualify the welding and cutting processes for the port flange lip seal. The ITER VV material 316 L(N) IG was already qualified and the Modified Reference Design was approved by the Agreed Notified Body (ANB) in accordance with the Nuclear Pressure Equipment Order procedure. © 2012 Published by Elsevier B.V.
1. Introduction The Vacuum Vessel (VV) is a key component of the ITER facility. The primary functions of the VV are to provide the first confinement barrier, withstand postulated accidents without losing confinement, remove the nuclear heating, provide a boundary consistent with the generation and maintenance of a high quality vacuum and support in-vessel components and their loads. This paper explains the status of the VV design and R&D, and preparation for its construction. 2. VV design and other activities 2.1. VV design completion The VV remains a double-walled torus-shaped structure composed of SS 316 L(N)-IG (ITER Grade) [1]. The main vessel consists
夽 The views and opinions expressed herein do not necessarily reflect those of the ITER Organization. ∗ Corresponding author. Tel.: +49 89 3299 4428; fax: +49 89 3299 4422. E-mail address: [email protected] (K. Ioki). 0920-3796/$ – see front matter © 2012 Published by Elsevier B.V. doi:10.1016/j.fusengdes.2012.02.023
of an inner and outer shell, poloidal and toroidal ribs and inwall shielding (IWS). Flexible support housings (FSHs) and keys are welded to the vessel shells (as shown in Fig. 1) [2], since the blanket modules are directly supported by the VV. Electron beam (EB) welding will be used as much as possible for joints between the inner shell and these structures. The layout of welds on the inner and outer shells is very tight considering accessibility for welding and non-destructive examination requirements defined in the design codes. The triangular support plays an important role in the plasma vertical stability control during minor disruptions, and its position and configuration is optimized based on the plasma vertical stability analysis. The VV design (3D CAD models) was distributed to the Domestic Agencies (DAs) and their suppliers in Nov 2010 for their development of the manufacturing design. In 2010, several design modifications/optimizations were made for the VV, as explained below (a)–(f) (see Fig. 2). The designs of interfacing components drove most of these changes. (a) Supporting rails for in-vessel coils (IVC) and Blanket manifolds—Layout of the support rails was optimized and the rail structure was simplified. Additional local tolerance requirements were introduced.
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
829
Fig. 3. In-wall shielding (left: 20◦ sector, right: whole structure).
2.2. Other activities for VV design, analysis and R&D
Fig. 1. ITER Vacuum Vessel.
(b) Inter-modular keys and stub keys—The keys supporting blanket modules were reinforced to withstand higher EM loads. (c) Upper port plug supports (#4–7)—Support structures for port shield blocks were added considering difficulties in the access from the neutral beam injector cell. (d) Port stub configuration—Port stub configuration (upper ports) was optimized/modified based on an Assembly Group request. (e) Cooling/drain lines of triangular supports (close to the sector edges)—A design modification is required to avoid a conflict with the toroidal transport of the divertor cassette. (f) Lower port—One of the lower ports was modified to avoid interference with a PF4 feeder.
Fig. 2. Design modifications explained in Section 2.1.
2.2.1. IWS design and analysis The space between the double walls will be mostly filled with shield structures mainly made of an austenitic stainless steel containing boron. To minimize induced eddy currents, the IWS is made of ∼8000 blocks (see Fig. 3). The addition of boron to SS 304 was adopted to improve neutron shielding efficiency. Two types of borated steels are used: (i) Steel type 304B4 with 1.0–1.24% of boron for the outboard region of the VV, (ii) Steel type 304B7 with 1.75–2.25 wt.% of boron for the inboard region. This steel (SS304B7) has low ductility (minimum total elongation of 6%) and low fracture toughness. However, this issue will be mitigated by the selection of a powder metallurgy (P/M) route for manufacturing of SS304B7. A ferritic stainless steel type 430 is used under the TF coils in the outboard area to reduce the toroidal field ripple. This steel has a saturated magnetization at ∼1.6 T. The borated and ferritic steels have high corrosion resistance in water and acceptable fabrication characteristics. The toroidal field ripple calculation was completed and the maximum ripple in the plasma region was reduced from 1.16% to 0.30% (in the nominal toroidal field) and −0.67% (in half of the nominal toroidal field) by the ferromagnetic inserts in the regular sectors (see Fig. 4). Electromagnetic (EM) loads on IWS blocks due to the magnetization effect have been analyzed using a 3D solid model (see Fig. 5). The maximum load per IWS block is 200 kN (toroidal direction), 45 kN (poloidal direction) and 30 kN (radial direction). Detailed IWS design has progressed also taking into account the VV design modifications. (i) IWS blocks have been shifted according
Fig. 4. Distribution of toroidal field ripple (left: nominal toroidal field, right: half of nominal toroidal field).
830
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
Fig. 5. Analysis model (left), and magnetization electromagnetic load on each IWS block (right).
to the new VV inner shell location (MDR: Reference Design). (ii) The IWS design has been modified for consistency with design changes in the Vacuum Vessel such as upper ports and flexible support. (iii) The gaps between IWS blocks have been adjusted based on nuclear calculation results. The basic scheme of IWS assembly stays the same (see Fig. 6(a)) and it also shows compatibility with the Tshape adapters of the VV poloidal/toroidal ribs for welding with the outer shell (see Fig. 6(b) and (c)). Tools for IWS block assembly have been designed. The lifting tool includes an eye bolt, a cover cap and a nut (see Fig. 7). Before lifting the block, the eye bolt is screwed with the IWS bolt head. Then the cover cap covers the IWS
Fig. 7. IWS blocks in outboard and lifting tools.
bolt head and the nut is fixed onto the cover cap. The cover cap and nut give an additional structural margin to support the IWS block against the torque generated by gravity. Structural analysis has been performed to confirm the structural integrity of the IWS. In the most severe case (CAT III event: EM and seismic loads), limit analysis was performed. The calculated load factor (LF) was 2.7 where the required minimum value of LF for category III events is 1.25. This shows that there is sufficient structural margin in the IWS design. 2.2.2. VV support Design of the dual-hinge type VV supports has progressed. The number of supports is nine and they are connected to an even number of lower ports (see Fig. 8). The dual hinge support has a primary hinge and a secondary hinge. The primary hinge is connected with upper and lower blocks through dowels (200 mm diameter) and provides all constraints for vertical downwards, upwards, and toroidal forces as well as radial & toroidal moment. The secondary hinge contacts upper and lower blocks and provides constraint for vertical downward forces. The upper block is connected to the port basement by bolts and the lower block is connected to the cryostat pedestal ring by bolts. Structural analysis has been performed and limit analysis shows a load factor of 2.0 where the RCC-MR code [3] requirement is 1.25 as the minimum. Global structural analysis results (including a fault mode) of the VV supports are shown in Fig. 9 and local support analysis results are in Fig. 10. R&D testing was performed for the selection and qualification of the dowel coating materials (see Fig. 11). MoS2 and WS2 coating materials were tested under a vacuum condition under maximum pressure 200 MPa at 100–200 ◦ C, and the measured friction factor was 0.14–0.28.
Fig. 6. Assembly procedure of IWS blocks. (a) Inboard, (b) Poloidal T-rib Area, and (c) Toroidal T-rib Area.
2.2.3. Dynamic test of keys Electromagnetic loads on each blanket module are supported by the VV. The radial torque and poloidal force are supported by the inter-modular key in the inboard region. There is a small gap (<1 mm) between the key and blanket module. This gap facilitates the installation of the blanket modules. The reaction force on the key from the transient EM load on the blanket is amplified and the
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
831
Fig. 10. Local support analysis results (CAT III events).
Fig. 11. (i) Friction testing for VV support dowel and (ii) MoS2 surface after testing.
Fig. 8. (a) VV lower port and support and (b) VV support.
amplification factor is defined as the Dynamic Amplification Factor (DAF). An R&D program was launched to measure the DAF. As shown in Fig. 12, the test set-up consists of a drop mass {1} which impacts on the free end of a large beam . This beam is fixed to a ‘hinge’ at the other end {4}. The ‘hinge’ is a steel plate which has a relatively low bending stiffness and bends very little during the test. The beam is used to amplify and shape the force pulse, generated by the drop mass, which is applied on the key {3}. Furthermore, an auxiliary mass {5} is mounted on the beam above the key fixture. This mass represents the mass of a blanket module. The DAF in halo current load conditions was measured by changing the gap height, the contact area of impact on the key and the duration of the input pulse. A nominal duration was 50 ms and nominal amplitude was 1.4 MN. The maximum measured DAF in halo current load conditions was 1.36. A different test set-up is prepared for DAF measurements during major disruptions due to the nature of the applied load (a torque). 2.2.4. Lip seal welding and cutting for port flange joint An R&D program has started to select and qualify the welding and cutting process for the port flange lip seal. It also includes
Fig. 9. Global analysis results (including fault modes) of the VV supports. Fig. 12. Dynamic load test on key.
832
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
Table 1 Comparison of TIG and laser welding methods for lip seal of port flange.
Practically achievable penetration Welding speed Allowable deviations/misalignments Air gap Vertical misalignment of lip plates Welding head distance to lip joint Welding head transversal position Welding head angular position a b c
Laser welding
TIG weldings
>3 mm 1–2 m/min
2–2.5 mma 250–500 mm/mim
<0.2 mm 0.5 mm ±2–3 mm ±0.2–0.3 mmb ±3◦ –4◦
0.5–1 mm 1 min ±0.2–0.3 mmc ±0.5–1 mm ±5◦ –10◦
May be increased with current pulsing techniques. May be increased with special welding head optics or by using a linear scanner. Requires on-line sensor system. Fig. 15. Lip-seal cutting equipment by dry milling (left) and chip collecting system (right).
for Laser welding. The 2 mm thick lips were tack welded before each robot driven welding. TIG and Laser mock-ups were welded and cut, respectively. 2 and 4 times. The mock-ups were both leak tested to verify the leak rate requirement (<1.0 × 10−10 Pa m3 /s). Dry milling was selected as the cutting method in this R&D program (see Fig. 15). Several runs are necessary to remove a 3 mm deep weld. The tool is equipped with a chip collecting system which satisfies the cleanliness requirements.
Fig. 13. Profiles of welds in lips. (a) Laser welding, Focus position: −3 mm, (b) Laser welding, Focus position: 0 mm, (c) TIG welding.
the development of NDT methods of welds and the demonstration of the adaptability of a robot arm by using real scale mock-ups of the lip seal. TIG welding and YAG Laser welding were selected based on initial tests and an assessment. The two welding processes were compared (see Table 1 and Fig. 13). Two full scale mockups were tested: one for TIG welding (see Fig. 14) and the other
Fig. 14. Lip seal welding on full-scale mock-up.
2.2.5. Thermal hydraulic analysis Thermal hydraulic analysis is under way for the main vessel and ports in collaboration with the DAs. A preliminary result for the main vessel is shown in Fig. 16. In most of the areas, the temperature is below 120 ◦ C and the heat transfer coefficient is much higher than the minimum requirement (500 W/m2 /K). Further analysis is being performed to improve cooling performance in specific areas in the triangular support or near ports. 2.2.6. Design of in-vessel coils (ELM/VS coils) and interfaces with the VV The ELM control coils include three rows of 9 sets corresponding to toroidally 40◦ spanning VV sectors (see Fig. 17). Each row of the ELM control coils is independently installed and supported to the VV inner surface. The VS coils are prepared inside the VV as toroidally continuous coils along the upper and lower legs. Based on
Fig. 16. Thermal hydraulic analysis result of main vessel (temperature distribution of inner shell surface).
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
Fig. 17. In-vessel (ELM/VS) coils installed on VV sectors (coils are shown in green). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
R&D results on irradiation damage of electrical insulation material, the coil design was changed to the mineral-insulated (MI) cable type. ELM and VS coil conductors are attached to a central support structure and fixed to the VV by its support rail. The ELM coils (including coil feeders) have been designed to be remotely maintainable. The coils are bolted to coil supports welded to the VV inner shell. A square cross-section rail (instead of a L-shaped rail) is used where possible, considering that the weld structures shall satisfy RCC-MR [3] requirements for weld type and non-destructive examination (Ultrasonic Testing is to be used) on welds in the VV inner shell. The upper VS coil was shifted upward above the upper port to optimize the feeder routing and interfaces with blanket modules. The ELM and upper VS coil feeders are routed out of the Vacuum Vessel through the upper ports and the lower VS coil feeders are routed out of the vessel through two small dedicated ports. A blanket manifold design was developed to fit with the ELM/VS coil design into the outboard region. A blanket manifold consisting of multiple pipes is used in the outboard region while retaining the option for the replacement by remote handling. 3. Activities towards the VV construction 3.1. Conformity assessment of the VV design The ITER Vacuum Vessel (VV) is nuclear pressure equipment according to the French Decree of December 12th 2005 (ESPN). Therefore the VV manufacturer must demonstrate that the applicable essential safety requirements and radioprotection requirements are satisfied. The procedure for licensing the ITER VV according to ESPN is in progress and conformity assessment is being performed by AIB-Vinc¸otte International selected by ITER Organization as the Agreed Notified Body (ANB). The ITER Organization (IO) is the Manufacturer (under the ESPN definition) of the complete VV (double shell structure and ports) and responsible for design, manufacture and conformity assessment. The VV is an assembly of ESPN components including sectors and ports. As part of the Phase I, devoted to design of this process, IO submitted to the ANB a set of documents (design description, load specification, hazard analysis, stress report, drawing package, radioprotection guide, approach for pressure test, particular material appraisals, material procurement specifications, etc.) and gained a preliminary design approval and approval of material specifications. Due to design modifications, a set of complementary documents on the Modified Reference Design were further assessed by ANB to check that the design changes are
Fig. 18. Vacuum Vessel mock-ups (KODA/HHI).
833
834
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835
Fig. 19. Lower port mock-up (KODA/HHI) (upper) and upper port mock-up (RFDA) (lower).
Fig. 20. Assembled inboard IWS mock-ups (upper) and outboard IWS mock-ups (lower) (INDA/ATL).
structurally acceptable. The process continues with a full update of the documentation package necessary before starting the first fabrication activities of the VV components.
assembling four individual poloidal segments and then final assembly of these together into the sector. Manufacturing and material qualification documents required before the start of fabrication were prepared by DAs and VV main suppliers and reviewed by IO and the ANB. Various mock-ups are in progress by Korea and Europe for the main vessel sectors, by Korea and Russia for the lower port and the upper port, respectively, and by India for the In-Wall Shielding to establish in detail manufacturing sequences and validate in particular tolerance aspects (see Figs. 18–20).
3.2. VV material qualification The procurements for the ITER Vacuum Vessel and Ports will be “in-kind” and materials for these components are produced by different DAs. In accordance with the regulatory requirements and quality requirements for operation, common material specifications were prepared in collaboration with the DAs and their industries. Documents to qualify plate/forging material manufacturing by each material supplier selected by the DAs and their main fabricators are submitted to IO and the ANB for review and approval. Several material suppliers were qualified and fabrication has started for plates for KO DA after approval of all required documentation. 3.3. Manufacturing design by DAs and VV suppliers Conceptual or preliminary manufacturing designs have been prepared by DAs and VV main suppliers and have been reviewed by IO and the ANB. Structural analysis has been performed to justify various proposed manufacturing design details including reduction of the minimum shell thickness of the VV. When the manufacturing design features require design changes, Project Change Requests (PCR) or Deviation Requests (DR) are submitted and assessed by IO and the ANB. A PCR or DR impacting structural behavior is required to include dedicated stress reports. Prior to the start of manufacturing, Manufacturing Designs shall be approved by the ANB and are planned by Korea and Europe for late this year. In Korea and in Europe, the same basic sequence will be used for the assembly of the sectors. Sector assembly will be accomplished by first
4. Conclusion Several design modifications were made during 2010 and 2011 according to requests from in-vessel coil interfaces, blanket module supports and manifolds, other interfaces, and the DAs and their suppliers. Designs of the IWS and VV supports have progressed including detailed electromagnetic, thermal and structural analyses. The ITER fabrication technology has progressed based on results of the L-3 project and additional R&D [4–7] by the Domestic Agencies. The fabricated sectors and ports will be transported to the ITER site in 2015–2016 for assembly. References [1] K. Ioki, P. Barabaschi, V. Barabash, W. Daenner, F. Elio, et al., Design improvements and R&D achievements for VV and in-vessel components towards ITER construction, Nuclear Fusion 43 (2003) 268–273. [2] K. Ioki, V. Barabash, J. Cordier, M. Enoeda, G. Federici, et al., ITER Vacuum Vessel, in-vessel components and plasma facing material, Fusion Engineering and Design 83 (2008) 787–794. [3] Design and Construction Rules for Mechanical Components of Nuclear Installation, RCC-MR, French Association for the Design, Construction and Operating Supervision of the Equipment for Electro-Nuclear boilers (AFCEN), 2007.
K. Ioki et al. / Fusion Engineering and Design 87 (2012) 828–835 [4] K. Koizumi, M. Nakahira, Y. Itou, E. Tada, G. Johnson, et al., Design and development of the ITER vacuum vessel, Fusion Engineering and Design 41 (1998) 299. [5] L. Jones, J.-F. Arbogast, A. Bayon, S. Galvan, B. Giraud, et al., European preparations for the ITER VV sectors manufacture, Fusion Engineering and Design 86 (2011) 2096–2100.
835
[6] L. Jones, J. Duhovnik, M. Ginola, J. Huttunen, K. Ioki, et al., Results from iter vacuum vessel sector manufacturing development in Europe, Fusion Engineering and Design 82 (2007) 1942. [7] B.Y. Kim, H.J. Ahn, M.S. Ha, Y.K. Kim, H.S. Kim, et al., Design analysis of the hinge support for the ITER vacuum vessel, Fusion Engineering and Design 86 (2011) 2003–2007.