No abstract

No abstract

The electrical interconnection of Nb 3 Sn/Cu strands is a key issue for the construction of Nb 3 Sn based damping ring wigglers and insertion devices for third generation light sources. We compare the electrical resistance of Nb 3 Sn/Cu splices manufactured by solid state welding using Electromagnetic Pulse Technology (EMPT) with that of splices produced by soft soldering with two different solders. The resistance of splices produced by soft soldering depends strongly on the resistivity of the solder alloy at the operating temperature. By solid state welding splice resistances below 10 n can be achieved with 1 cm strand overlap length only, which is about 4 times lower than the resistance of Sn96Ag4 soldered splices with the same overlap length. The comparison of experimental results with Finite Element simulations shows that the electrical resistance of EMPT welded splices is determined by the resistance of the stabilizing copper between the superconducting filaments and confirms that welding of the strand matrix is indeed achieved. EMPT allows interconnecting the ductile, unreacted strands, which reduces the risk of damaging the brittle reacted Nb 3 Sn strands.To be published in Superconductor Science and Technology Geneva, Switzerland CERN-ATS-2011-230 AbstractThe electrical interconnection of Nb 3 Sn/Cu strands is a key issue for the construction of Nb 3 Sn based damping ring wigglers and insertion devices for third generation light sources. We compare the electrical resistance of Nb 3 Sn/Cu splices manufactured by solid state welding using Electromagnetic Pulse Technology (EMPT) with that of splices produced by soft soldering with two different solders. The resistance of splices produced by soft soldering depends strongly on the resistivity of the solder alloy at the operating temperature. By solid state welding splice resistances below 10 nΩ can be achieved with 1 cm strand overlap length only, which is about 4 times lower than the resistance of Sn96Ag4 soldered splices with the same overlap length. The comparison of experimental results with Finite Element simulations shows that the electrical resistance of EMPT welded splices is determined by the resistance of the stabilizing copper between the superconducting filaments and confirms that welding of the strand matrix is indeed achieved. EMPT allows interconnecting the ductile, unreacted strands, which reduces the risk of damaging the brittle reacted Nb 3 Sn strands.

The Large Hadron Collider (LHC) main interconnection splices consist of Rutherford-type cable splice and busbar stabilizer splices. Busbar stabilizer splices have been consolidated during the first long LHC shutdown by soldering additional Cu shunts. In view of the large number of quality controls (QCs) that were integrated in the splice consolidation process, efficient and unambiguous QC procedures needed to be developed. Directcurrent electrical resistance measurements have been selected for the control of the busbar splices and the individual shunts. About 400 000 resistance measurements performed at room temperature before and after each consolidation step have been analyzed. The resistance of the consolidated splices is comparable with the resistance of continuous busbars without splice. Resistance changes during the consolidation process correspond to those calculated from the changes in Cu cross-sectional area.

For the consolidation of the LHC 13 kA main interconnection splices, shunts will be soldered onto each of the 10170 splices. The solder alloy selected for this purpose is Sn60Pb40. In this context the electrical resistance of shunt to busbar lap splices has been measured in the temperature range from RT to 20 K. A cryocooler set-up has been adapted such that a test current of 150 A could be injected for accurate resistance measurements in the low nΩ range. To study the influence of the solder bulk resistivity on the overall splice resistance, connections produced with Sn96Ag4 and Sn77.2In20Ag2.8 have been studied as well. The influence of the Sn60Pb40 solder resistance is negligible when measuring the splice resistance in a longitudinal configuration over a length of 6 cm. In a transverse measurement configuration the splice resistance is significantly influenced by the solder. The connections prepared with Sn77.2In20Ag2.8 show significantly higher resistance values, as expected from the relatively high solder resistivity at cryogenic temperatures. The experimental results are complemented by simulations with Comsol multiphysics, showing the contribution of each splice component on the overall splice resistance. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. 3LPA-09