Impact of Cabling Pattern, Magnet Field Profile and Joint Properties on Short Sample Qualification Tests of ITER Conductors

Abstract

The voltage signals measured in short sample qualification tests for the ITER conductors often show a linear relationship with the current starting from zero. This behavior is most likely caused by non-uniform current distribution in the joints. To counteract this, two samples have been equipped with extra sets of six voltage tap pairs, connected with regular spacing, as a star configuration around the conductor. The linear relationship still showed up in the average of the extra voltage signals, and were sometimes even seen to be inverse. In some occasions, it was negative for all individual signals.A fair judgment of the approval criterion requires a proper understanding of this behavior. This paper describes the analyses with the aid of the numerical model 'JackPot'. It demonstrates that the described observation is the result of the chosen cabling pattern in combination with the position of the voltage taps. The model also shows that this problem cannot be solved by changing the positions of the voltage taps, but that completely filling the joints with solder can greatly improve the situation.

Results of the TF conductor performance qualification samples for the ITER project

“…The current distribution among the cable strands during a T cs test in the SULTAN facility is mainly determined by the presence of the cable terminations [41,42]. Several investigations performed with numerical codes [16,17], with input data derived from experimental results [38][39][40]54], have led to the conclusion that the solder dipping and solder filling techniques for the joint allow satisfactory current redistribution during the T cs tests, thus reducing the impact of non-uniformities on the assessment of the conductor performance. In one case, namely for the TFJA3 sample [11] (manufactured with a pressure contact joint), the high total joint resistance and the highly dispersed distribution of the joint resistances measured on individual strands were considered to have a negative impact on the DC performance [54,55].…”

“…Because of the space restrictions in the SULTAN magnet system, the bottom joint is only 565 mm from the center of the HFZ. As a result, a set of investigations has been conducted on the effect of the joint technology on the sample performance, both with experiments [38][39][40] and with independent numerical simulations [16,17,41,42]. ITER's SULTAN Working Group (SWG) has settled on the 'solder filled' joint technology, introduced in [14] and first adopted for the ITER conductors of the last generation by the US-DA [9].…”

“…The sample preparation procedures were aimed at avoiding slippage between the cable and the jacket due to their different thermal contraction coefficients by means of crimping rings [13], and at developing a suitable joint technology to improve the current redistribution during the T cs test [9,[14][15][16][17][18][19].…”

Optimization of ITER Nb3Sn CICCs for coupling loss, transverse electromagnetic load and axial thermal contraction

“…The PITSAM CICCs with different cabling patterns show similar tendency with smaller degradation due to lower transverse B × I load as the number of strand layers in the cable is much less than for the TFPRO2. The reduction in T cs is computed with the JackPot model [29][30][31][32], taking into account all strand trajectories, the strand magnet field, temperature and strain scaling, the applied and cable self-field, to assess the maximum possible virgin T cs based on the strand properties and compare it to the SULTAN data. In addition another series of sub-size CICCs, also contained a variation of the twist pitch with respect to the traditional cable pitch sequence of 45 mm × 85 mm × 125 mm but in the direction of shorter pitch lengths 35 mm × 65 mm × 110 mm [33].…”

Qualification tests and facilities for the ITER superconductors

“…According to one proposed interpretation [19,20], the slope versus current originates from the mismatch of contact resistance at the two conductor terminals and is an artefact of the sample joints. Such a hypothesis calls for a highly nonhomogeneous resistance distribution at the joint and for very high inter-strand resistance inside the termination.…”