Limits...
Kiloampere, Variable-Temperature, Critical-Current Measurements of High-Field Superconductors.

Goodrich LF, Cheggour N, Stauffer TC, Filla BJ, Lu XF - J Res Natl Inst Stand Technol (2013)

Bottom Line: Therefore, a significant portion of this review is focused on the reduction of temperature errors to less than ±0.05 K in such measurements.We also calibrated the magnetoresistance effect of resistive thermometers for temperatures from 4 K to 35 K and magnetic fields from 0 T to 16 T.This calibration reduces systematic errors in the variable-temperature data, but it does not affect the liquid/gas comparison since the same thermometers are used in both cases.

View Article: PubMed Central - PubMed

Affiliation: Department of Physics, University of Colorado, Boulder, CO 80309 ; National Institute of Standards and Technology, Boulder, CO 80305.

ABSTRACT
We review variable-temperature, transport critical-current (I c) measurements made on commercial superconductors over a range of critical currents from less than 0.1 A to about 1 kA. We have developed and used a number of systems to make these measurements over the last 15 years. Two exemplary variable-temperature systems with coil sample geometries will be described: a probe that is only variable-temperature and a probe that is variable-temperature and variable-strain. The most significant challenge for these measurements is temperature stability, since large amounts of heat can be generated by the flow of high current through the resistive sample fixture. Therefore, a significant portion of this review is focused on the reduction of temperature errors to less than ±0.05 K in such measurements. A key feature of our system is a pre-regulator that converts a flow of liquid helium to gas and heats the gas to a temperature close to the target sample temperature. The pre-regulator is not in close proximity to the sample and it is controlled independently of the sample temperature. This allows us to independently control the total cooling power, and thereby fine tune the sample cooling power at any sample temperature. The same general temperature-control philosophy is used in all of our variable-temperature systems, but the addition of another variable, such as strain, forces compromises in design and results in some differences in operation and protocol. These aspects are analyzed to assess the extent to which the protocols for our systems might be generalized to other systems at other laboratories. Our approach to variable-temperature measurements is also placed in the general context of measurement-system design, and the perceived advantages and disadvantages of design choices are presented. To verify the accuracy of the variable-temperature measurements, we compared critical-current values obtained on a specimen immersed in liquid helium ("liquid" or I c liq) at 5 K to those measured on the same specimen in flowing helium gas ("gas" or I c gas) at the same temperature. These comparisons indicate the temperature control is effective over the superconducting wire length between the voltage taps, and this condition is valid for all types of sample investigated, including Nb-Ti, Nb3Sn, and MgB2 wires. The liquid/gas comparisons are used to study the variable-temperature measurement protocol that was necessary to obtain the "correct" critical current, which was assumed to be the I c liq. We also calibrated the magnetoresistance effect of resistive thermometers for temperatures from 4 K to 35 K and magnetic fields from 0 T to 16 T. This calibration reduces systematic errors in the variable-temperature data, but it does not affect the liquid/gas comparison since the same thermometers are used in both cases.

No MeSH data available.


Related in: MedlinePlus

Apparent temperature error ΔT versus H2 at 7.5 T and T2 = 5 K where Ic is about 143 A for a Nb-Ti wire soldered to a stainless-steel mandrel on the VTO probe. The T bias (T1−T2) was set to 0, 10, 20, and 30 mK. The weighted temperature Tw = 0.14 T1+0.86 T2 was used to determine the correct Ic, which collapsed the Fig. 38 data onto one line.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4487313&req=5

f39-jres.118.015: Apparent temperature error ΔT versus H2 at 7.5 T and T2 = 5 K where Ic is about 143 A for a Nb-Ti wire soldered to a stainless-steel mandrel on the VTO probe. The T bias (T1−T2) was set to 0, 10, 20, and 30 mK. The weighted temperature Tw = 0.14 T1+0.86 T2 was used to determine the correct Ic, which collapsed the Fig. 38 data onto one line.

Mentions: We measured the apparent temperature error ΔT versus H2 at 7.5 T and T2 = 5 K where Icliq was 143 A with T bias values of 0, 10, 20, and 30 mK as shown in Fig. 38. Since T1 and T2 are different, we must determine an appropriately weighted temperature, Tw. For Fig. 38, we used Tw = 0.5 T1 +0.5 T2, which is the simple average of the two temperatures. The extreme case would have Tw = T2, which, with a T bias of 30 mK, would only shift Tw by 15 mK. Thus this study is focused on a small effect. With equal weighting of the two thermometers, the H2 value where ΔT is zero is 13, 6, −1, and −9 mW for T biases of 0, 10, 20, and 30 mK, respectively. Negative values of H2 where ΔT is zero are not realistic, based on the previous findings. Therefore, we need to determine different temperature weights that gives more consistent results. We found, by iteration to minimize the difference in the H2 value where ΔT is zero, that Tw = 0.14 T1 +0.86 T2 collapses the Fig. 38 data onto essentially one line, as shown in Fig. 39. With this Tw equation, the H2 value where ΔT is zero is 9.6, 10.2, 10.3, and 9.4 mW for T biases of 0, 10, 20, and 30 mK. A plot of ΔT versus H1 for the same data set and Tw equation is shown in Fig. 40. As expected, values of H1 increase for a given H2 with increasing T bias. This demonstrates that the values of T1 and H1 are less influential than values of T2 and H2.


Kiloampere, Variable-Temperature, Critical-Current Measurements of High-Field Superconductors.

Goodrich LF, Cheggour N, Stauffer TC, Filla BJ, Lu XF - J Res Natl Inst Stand Technol (2013)

Apparent temperature error ΔT versus H2 at 7.5 T and T2 = 5 K where Ic is about 143 A for a Nb-Ti wire soldered to a stainless-steel mandrel on the VTO probe. The T bias (T1−T2) was set to 0, 10, 20, and 30 mK. The weighted temperature Tw = 0.14 T1+0.86 T2 was used to determine the correct Ic, which collapsed the Fig. 38 data onto one line.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4487313&req=5

f39-jres.118.015: Apparent temperature error ΔT versus H2 at 7.5 T and T2 = 5 K where Ic is about 143 A for a Nb-Ti wire soldered to a stainless-steel mandrel on the VTO probe. The T bias (T1−T2) was set to 0, 10, 20, and 30 mK. The weighted temperature Tw = 0.14 T1+0.86 T2 was used to determine the correct Ic, which collapsed the Fig. 38 data onto one line.
Mentions: We measured the apparent temperature error ΔT versus H2 at 7.5 T and T2 = 5 K where Icliq was 143 A with T bias values of 0, 10, 20, and 30 mK as shown in Fig. 38. Since T1 and T2 are different, we must determine an appropriately weighted temperature, Tw. For Fig. 38, we used Tw = 0.5 T1 +0.5 T2, which is the simple average of the two temperatures. The extreme case would have Tw = T2, which, with a T bias of 30 mK, would only shift Tw by 15 mK. Thus this study is focused on a small effect. With equal weighting of the two thermometers, the H2 value where ΔT is zero is 13, 6, −1, and −9 mW for T biases of 0, 10, 20, and 30 mK, respectively. Negative values of H2 where ΔT is zero are not realistic, based on the previous findings. Therefore, we need to determine different temperature weights that gives more consistent results. We found, by iteration to minimize the difference in the H2 value where ΔT is zero, that Tw = 0.14 T1 +0.86 T2 collapses the Fig. 38 data onto essentially one line, as shown in Fig. 39. With this Tw equation, the H2 value where ΔT is zero is 9.6, 10.2, 10.3, and 9.4 mW for T biases of 0, 10, 20, and 30 mK. A plot of ΔT versus H1 for the same data set and Tw equation is shown in Fig. 40. As expected, values of H1 increase for a given H2 with increasing T bias. This demonstrates that the values of T1 and H1 are less influential than values of T2 and H2.

Bottom Line: Therefore, a significant portion of this review is focused on the reduction of temperature errors to less than ±0.05 K in such measurements.We also calibrated the magnetoresistance effect of resistive thermometers for temperatures from 4 K to 35 K and magnetic fields from 0 T to 16 T.This calibration reduces systematic errors in the variable-temperature data, but it does not affect the liquid/gas comparison since the same thermometers are used in both cases.

View Article: PubMed Central - PubMed

Affiliation: Department of Physics, University of Colorado, Boulder, CO 80309 ; National Institute of Standards and Technology, Boulder, CO 80305.

ABSTRACT
We review variable-temperature, transport critical-current (I c) measurements made on commercial superconductors over a range of critical currents from less than 0.1 A to about 1 kA. We have developed and used a number of systems to make these measurements over the last 15 years. Two exemplary variable-temperature systems with coil sample geometries will be described: a probe that is only variable-temperature and a probe that is variable-temperature and variable-strain. The most significant challenge for these measurements is temperature stability, since large amounts of heat can be generated by the flow of high current through the resistive sample fixture. Therefore, a significant portion of this review is focused on the reduction of temperature errors to less than ±0.05 K in such measurements. A key feature of our system is a pre-regulator that converts a flow of liquid helium to gas and heats the gas to a temperature close to the target sample temperature. The pre-regulator is not in close proximity to the sample and it is controlled independently of the sample temperature. This allows us to independently control the total cooling power, and thereby fine tune the sample cooling power at any sample temperature. The same general temperature-control philosophy is used in all of our variable-temperature systems, but the addition of another variable, such as strain, forces compromises in design and results in some differences in operation and protocol. These aspects are analyzed to assess the extent to which the protocols for our systems might be generalized to other systems at other laboratories. Our approach to variable-temperature measurements is also placed in the general context of measurement-system design, and the perceived advantages and disadvantages of design choices are presented. To verify the accuracy of the variable-temperature measurements, we compared critical-current values obtained on a specimen immersed in liquid helium ("liquid" or I c liq) at 5 K to those measured on the same specimen in flowing helium gas ("gas" or I c gas) at the same temperature. These comparisons indicate the temperature control is effective over the superconducting wire length between the voltage taps, and this condition is valid for all types of sample investigated, including Nb-Ti, Nb3Sn, and MgB2 wires. The liquid/gas comparisons are used to study the variable-temperature measurement protocol that was necessary to obtain the "correct" critical current, which was assumed to be the I c liq. We also calibrated the magnetoresistance effect of resistive thermometers for temperatures from 4 K to 35 K and magnetic fields from 0 T to 16 T. This calibration reduces systematic errors in the variable-temperature data, but it does not affect the liquid/gas comparison since the same thermometers are used in both cases.

No MeSH data available.


Related in: MedlinePlus