The ITER toroidal field model coil project

More documents

Recommendations

Info

198 A. Ulbricht et al. / Fusion Engineering and Design 73 (2005) 189–327 Fig. 3.6. Two pancakes in the reaction moulds in front of the furnace. The radial plates (RP) and covers are made of 316LN stainless steel by forging and machining. Thanks to intermediate heat treatments at 950 ◦ C the flatness of the finally machined radial plates was within ∼0.2 mm distinctly better than originally expected by industry. Fig. 3.7. Insulation of the turns including the installation of the cowound stainless steel tape voltage taps and transfer into the grooves of the radial plate. 3.3.5. Soldering and clamping of the inner joints It turned out that after heat treatment and removal of the fixture all terminations deformed into a bananalike shape, the copper sole being on the convex side. This posed some problems in the assembly of the terminations into the inter-pancake joints, but has been solved by precise machining of the two adjacent copper soles before connecting them. Because the necessity of such a machining was expected, some extra copper thickness was provided. The resulting joints were in agreement with the expectations (see Section 7). The procedure to soft-solder the inner terminations to form the inter-pancake joint caused a 2 mm thermal expansion, which was taken up by the not yet impregnated turn insulation. After soldering, the joints were fitted with rigid clamps as shown schematically in Fig. 3.5. 3.3.6. Laser welding of the covers To give the DP modules the right stiffness, the covers had to be welded with a penetration of 2.5 mm. Nd–YAG laser welding (2 kW/(0.6 m min), automatic tracking) was chosen to keep the heat input at a minimum (Fig. 3.8). Optimum flatness of the DP modules of less than 2 mm was achieved by several turnovers of the plate during the operation. With an eddy current test method developed by ENEA it was possible to check the weld quality and penetration. Fig. 3.8. Laser welding of the covers onto the conductor grooves of the radial plate with automatic tracker system.
3.3.7. Module ground insulation and impregnation The modules were wrapped by 1.3 mm glass/Kapton insulation followed by an impregnation with DGEBA epoxy resin at about 75 ◦ C followed by a curing cycle at about 125 ◦ C. The resin could penetrate through holes in the covers into the turn insulation. Some of the good flatness of the radial plates was lost due to a too low stiffness of the impregnation mould. 3.3.8. Final tests on the DP modules Before shipment, all DP modules passed insulation resistance tests (500 V DC conductor to radial plate), dielectric test (3 kV DC and ACpeak-to-peak conductor to radial plate and 1.5 kV DC radial plate to ground), continuity and mutual insulating test on all voltage taps, gas flow test and a leak test at 30 bar helium internal pressure in a vacuum vessel to a level of 1 × 10 −10 mbar l/s. The test results are compiled in Table 3.6. 3.4. Winding pack manufacture 3.4.1. DP stacking and winding pack impregnation A special stacking device was developed to align and stack the DP’s to form the winding pack, as shown in Fig. 3.9. Glass felt was introduced between the DP modules in order to get good impregnation and bonding between the DP’s and to adjust the height of the winding pack. During impregnation with epoxy resin a further setting of the glass felt took place, namely between the higher compressed lower DP’s. The total height of A. Ulbricht et al. / Fusion Engineering and Design 73 (2005) 189–327 199 Fig. 3.9. Stacking of the DP modules with glass felt inter layer to equalize tolerances and to adjust the height of the winding pack. the winding pack was decreased by about 2 mm. The impregnated winding pack can be seen in Fig. 3.10. 3.4.2. Outer joint manufacturing The gap between adjacent terminations had to be bridged by copper shims to secure a good contact. Unlike the inner joints, it was not possible to connect the terminations of two adjacent double pancakes by soldering because of the 2 mm thermal expansion during heating. This would have caused unacceptably high stresses in the jacket. These two technical constraints led the supplier to a solution using copper pins as shims (9–10 mm diameter) that were introduced into holes drilled between two Table 3.6 Data of final tests of the DP modules at Ansaldo works Flow test Pressure drop [bar] DP1 [N m3 /h] DP2 [N m3 /h] DP3 [N m3 /h] DP4 [N m3 /h] DP5 [N m3 /h] N2 mass flow through pancake DPX.1 (X = 1–5) N2 mass flow through pancake DPX.2 (X = 1–5) High voltage tests 3 kV DC between turns and RP, 1 min [G�] 3kVpeak–peak AC between turns and RP, 1 min [mA] 1.5 kV DC between RP and ground, 1 min [G�] 2 11.5 9.7 10.0 9.4 11.5 4 21.0 18.3 18.7 17.5 21.0 2 11.5 9.0 10.2 9.5 11.0 4 21.0 16.5 18.9 17.0 20.0 13 9.2 8.35 7.6 7.75 91 99 102 101 92 9.7 9.5 3 10 10.9
Page 1 and 2: Fusion Engineering and Design 73 (2
Page 3 and 4: The application and continuation of
Page 5 and 6: 3.2. Conductor manufacture 3.2.1. S
Page 7 and 8: A. Ulbricht et al. / Fusion Enginee
Page 9: Table 3.5 TFMC double pancake data
Page 13 and 14: impregnation of the coil in the cas
Page 17 and 18: trial consortium AGAN and the Europ
Page 23 and 24: ˙Q ICS [W] ˙Q case [W] ˙Q bus ba
Page 29 and 30: 4.3.2.4. Standby operation over nig
Page 35 and 36: 2 weeks (JAERI method); for joint 1
Page 37 and 38: Table 4.5 Main operation parameters
Page 45 and 46: Fig. 4.35. Computed TFMC windings c
Page 49 and 50: the quench detection system. The lo
Page 55 and 56: ply current regulation. This situat
Page 57 and 58: Fig. 5.6. (a) V-Tin characteristics
Page 61 and 62:
known, which is partly true (see Se
Page 63 and 64:
Fig. 5.18. Summary of performance a
Page 65 and 66:
filaments in the TFMC (and similar)
Page 67 and 68:
Symbols Explanation C00 Coefficient
Page 69 and 70:
A. Ulbricht et al. / Fusion Enginee
Page 71 and 72:
the outer joint outlet. This heat e
Page 73 and 74:
Fig. 6.9. Loss energy of the TFMC w
Page 75 and 76:
Fig. 6.13. Evolution of the couplin
Page 77 and 78:
and during Phase 2 (within 6%). The
Page 79 and 80:
other hand, if using the central ch
Page 81 and 82:
Fig. 6.20. Top: quench propagation
Page 83 and 84:
Fig. 6.23. Measured (multi star) an
Page 85 and 86:
fB Friction factor of the bundle re
Page 87 and 88:
surement with voltage taps located
Page 89 and 90:
Table 7.3 TFMC pancake (joint) resi
Page 91 and 92:
Table 7.8 Values of TFMC and FSJS j
Page 93 and 94:
Fig. 7.7. The insulated bus bar typ
Page 95 and 96:
Rneg, 1[nΩ] = 1.77 [n�] + 3.75
Page 97 and 98:
law actually is good (better than
Page 99 and 100:
Fig. 7.16. Current in the petals re
Page 101 and 102:
ence may exceed the inaccuracy of t
Page 103 and 104:
Page 105 and 106:
provided support to the coil and th
Page 107 and 108:
(c) The best fit in relation to the
Page 109 and 110:
Page 111 and 112:
Page 113 and 114:
8.5. Comparison to ITER TF coil str
Page 115 and 116:
Page 117 and 118:
Page 119 and 120:
Page 121 and 122:
Page 123 and 124:
Fig. 9.7. Partial discharge activit
Page 125 and 126:
Page 127 and 128:
circular cross-section and central
Page 129 and 130:
two different sets of Hall probe ar
Page 131 and 132:
equation: dx = Ax + Bu dt y = Cx +
Page 133 and 134:
[34] E. Salpietro, R. Maix, G. Bevi
Page 135 and 136:
[86] R. Heller, D. Ciazynski, J.L.
Page 137 and 138:
ple for ITER, IEEE Trans. Appl. Sup
Page 139:
PSB: Power System Blockset is a MAT
show all

The ITER toroidal field model coil project

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?