Development of a CMRC cooled 10 kA current lead for HTS applications

Many high temperature superconducting (HTS) applications like superconducting cables, power rails, generators, fault current limiter, accelerators and magnets may require several kilowatts of cooling power at an operating temperature range of 50-80 K. The major cooling demand, depending on the application up to 70-90 %, is necessary for the cooling of resistive current leads (CL) that supply electric energy from the ambience at 300 K to the superconducting application at cryogenic temperatures. Therefore, a high thermal integration of the CL with the cooling system is required to provide an energy-efficient and a cost-efficient CL technology. The cryogenic mixed refrigerant cycles (CMRCs) are the optimal solution for the cooling of CLs, due to the possibility to generate cooling power at the required cryogenic temperatures by adjusting the composition of the mixture and cycle parameters to attain a high energy efficiency. The key in this technology is the direct implementation of the resistive current lead into the recuperative heat exchanger of the Linde-Hampson cycle, providing refrigeration power over the temperature range of the heat exchanger. Hereby, a mixed refrigerant is partly condensing and evaporating inside the heat exchanger and absorbing the Joule heat at the temperature it is generated in the current lead. Hereby, the economic efficiency of HTS applications in the energy technology can be considerably improved. The aim of this work is the development of a 10 kA cryogenic mixed refrigerant cooled current lead (CMRC-CL). In order to enable a detailed investigation of the CL design, an existing numerical heat exchanger model is modified and implemented in the resistive current lead model that is developed in the frame of this work. Prior to that implementation, a literature review on the state-of-the-art CL solutions and cooling systems are conducted, numerically investigated and compared with each other. Based on this, a classical multi-tubes-in-tube CMRC-CL and a micro-structured CMRC-CL are developed and investigated numerically in this work. The first developed CMRC-CL design comprises a classical multi-tubes-in-tube heat exchanger which is wounded around a cylindrical CL made of copper. The second CMRC-CL prototype III is made of several micro-structured copper sheets that are connected together by a diffusion bonding process and is developed, patented and manufactured. Further, experimental investigation on the stainless steel micro-structured heat exchanger prototype II are conducted in the CMRC test stand at the Institute of Technical Thermodynamics and Refrigeration (ITTK) and the predictive qualities of the modified heat exchanger model are reviewed. In section 2 five state-of-the-art CL solutions and design approaches are introduced and numerically investigated. The corresponding thermal loads at the cold end, the CL shape factors, the optimal refrigerant mass flows and the theoretical power consumptions of the systems are summarized in the design overview. According to the Wiedemann-Franz-Law, which describes the relation between the specific electrical resistance and thermal conductivity of metals, any metallic material can be used as a current lead. Each material leads to a different optimal geometry of the CL and therefore, the thermophysical properties of relevant resistive CL materials are evaluated and compared with each other. The investigation of the CL types alone is not sufficient to evaluate the efficiency of the whole CL system and therefore, the state-of-the-art cooling systems for CLs are presented and discussed in section 3. Hereby, the total power consumptions P, including the required refrigeration power of the cooling system and the electric power losses of the respective current lead type, are calculated and discussed. A detailed analysis on the Linde-Hampson refrigeration cycle, which is the basis of CMRCs is presented. A numerical model for the calculation of CMRC-CLs is presented in section 4 which comprises the numerical investigations on the classical and the micro-structured CMRC-CLs. In order to identify an optimal refrigerant mixture, several parametric studies are performed with the classical CMRC-CL design. It was found that a refrigerant mixture with a larger composition of the low boiler methane, is an appropriate choice in the current design. In general, the classical CMRC-CL shows a significant reduction of the thermal load at the cold end and of the overall power consumption, compared to other state-of-the-art closed cycle cooling systems. The classical CMRC-CL designs in combination with a cryocooler at the cold stage, yields a specific thermal load at the 80K cold stage of 14 W/kA at an overall power consumption of about 600 W/kA. Compared to a conventional conduction cooled current lead (CCCL) that is cooled by one cryocooler, this is a 67 % reduction of the thermal load at a 50 % reduced overall power consumption. Considerable smaller values are archived only with an optimized self-sufficient vapour cooled current lead (ss-VCCL) with a thermal load of 9 W/kA at a power consumption of 280 W/kA, however, comes with the disadvantage of an open system that requires a continuous supply of LN${}_2$. The classical heat exchanger design leads to several scalability problems that are described in Section 4.8 and the adjustment of this design to larger electric currents is related to relative large CL lengths and diameters. Therefore, a new micro-structured CMRC-CL design is developed, patented and manufactured (in IMVT) in the frame of this work. It consists of several micro-structured copper sheets of 0.5 mm thickness that comprise a certain amount of etched channels for the fluid flow. The total amount of the sheets depends on the electric current and the refrigerant mass flow that is needed to absorb a certain amount of Joule heating, thermal radiation and thermal load due to thermal conduction from the ambience. All sheets are stacked together by a specific stacking pattern, then covered by the top and bottom plates and irreversibly connected in a diffusion bonding process. The new design, allows a simple adjustment of the amount of sheets, the amount of channels per sheet, the CL length and the refrigerant mass flow for a desired electric current. The respective numerical model is developed and CLs for electric currents of 10 kA and 20 kA are designed and numerically investigated. One of the investigated CMRC-CL designs C, yields a thermal load of 6.5 W/kA at a temperature of 85 K. Compared to a CCCL that is working in this temperature range, this is an 85 % reduction of the thermal load and therefore, the largest reduction compared to the state-of-the-art solutions. To follow the concept of the CL comparison that is presented in this work, the CMRC-CL design C is extended by an additional cooling machines (GM-AL60) at the cold end to absorb the remaining thermal load. With a total power consumption of 490 W/kA, the micro-structured CMRC-CL system is the most efficient closed-cycle system. However, it is to denote that it is realistic to develop CMRC-CL systems that do not need an additional cryocooler and temperatures below 85 K are possible by an adjustment of the refrigerant mixture and/or the refrigerant cycle. Further, the developed and manufactured micro-structured CMRC-CL is a solid and important design milestone in the development of future CL types. Experimental results that are obtained with the micro-structured heat exchanger prototype II, which is made of stainless steel, are discussed in section 5. The operating characteristics and the performance of this prototype are evaluated and the measurement data are compared with the results obtained from the numerical model. The conducted experiments showed a temperature decrease to about 85 K with a hydrocarbon based refrigerant mixture. The numerical model showed a good agreement with the measured temperature data along the heat exchanger at temperatures above 170 K, however, a deviation of the temperature gradients is present at lower temperatures and is investigated in this work. Experiments with mixtures containing neon showed no further cool down of the system and the temperature at the cold end increased instead, because of the reduced specific cooling power of this mixture with neon. Further experiments with LRS mixtures, that have a considerable larger specific cooling power, may be performed to overcome this effect. In conclusion, the numerical and experimental results on the micro-structured heat exchanger designs provide evidence that it is possible to develop a CMRC system that can cool down a superconducting application at least to the temperature of liquid nitrogen without the need of additional cryocoolers as the last cold stage. Furthermore, the new micro-structured CMRC-CL design allows a simple adjustment of its geometric and hydraulic parameters for a predefined electric current and it may be possible to design CLs even for the large electric currents of aluminium plants that are typically operated at 200-500 kA.