L. M. Besley

Thermal cycling apparatus to test germanium thermometer stabilities

Apparatus has been developed for thermal cycling germanium thermometers from ambient temperatures to 20.280 K and evaluating their reproducibility at the lower temperature. Gaseous and liquid helium are drawn from a storage Dewar into refrigerators loosely attached to a shielded and insulated enclosure initially at room temperature. Cooling the enclosure requires 5 h; heating it up to room temperature afterwards, 7 h. The existing apparatus accommodates 30 germanium thermometers and a reference standard platinum resistance thermometer. Because digital voltmeters can be used effectively, most of the data collection is automated and statistical analysis of the data is performed in minutes. Experiments have demonstrated that the standard deviation of 11 series of measurements (no thermal cycling) is the equivalent of 0.04 mK for an average thermometer at 20.280 K. The apparatus is being employed to determine the stability of the 30 thermometers during a 100‐cycle test.

Stability of germanium resistance thermometers at 20 K.

Thirty germanium resistance thermometers have been thermally cycled 100 times between 20 and 300 K, and their stability at 20 K has been evaluated. The results reveal a wide range of stabilities, ranging from 0.1 to 20 mK. Five different modes of behavior have been provisionally classified as stable, drifting, jumping, bimodal, and irregular.

Further stability studies on germanium resistance thermometers at 20 K

Thirty germanium resistance thermometers from three manufacturers have been thermally cycled 90 times between 20 and 300 K. Detailed results are presented on the stability of the thermometers at 20 K.

Properties of industrial‐grade platinum–cobalt resistance thermometers between 1 and 27 K

The stability at 20 K of five industrial‐type platinum–cobalt resistance thermometers on thermal cycling has been studied together with their resistance‐temperature characteristics between 1 and 27 K and their self‐heating. The stability at 20 K on cycling between room temperature and 20 K is of the order of 10 mK and on cycling between 100 and 20 K is a few mK. Differences in the resistance‐temperature relationships of the thermometers are very small, but the self‐heating effects are relatively large, particularly below 4 K.

Stability of some cryogenic carbon resistance thermometers

The stability of 12 commercial carbon resistance thermometers has been tested at three temperatures, 12.6, 20.3, and 30 K, when the sensors were exposed to cycling between 12.6 and 293 K. The resistances of the thermometers changed markedly over 20 such cycles by amounts equivalent to temperature changes of between 0.07 and 0.57 K at 20.3 K. Both abrupt and gradual changes were observed. A few measurements were made at 4.2 K to explore the correlation between changes in resistance at various temperatures. A two‐point recalibration method is shown to be accurate to 1 mK over the range 12.6 to 30 K and can be used to 4.2 K.

Stability studies on Kiev cryogenic germanium resistance thermometers

The stability of 19 germanium resistance thermometers manufactured in Kiev, USSR, has been studied by cycling them 30 times between temperatures of 20 and 300 K. Resistance measurements were made on each thermometer at 20 K on each cycle and also at 15 K on each cycle of the first 20 cycles. All but one of nine four‐lead, standard‐size thermometers were stable within the ±0.2 mK precision of the measurements at 20 K, while the observed changes in resistance at 20 K of 10 miniature thermometers correspond to temperature variations that ranged between 20 and 49 mK.

Measurements of internal heat transfer in germanium resistance thermometers

Heat transfer processes through the solid materials and the filling gas in germanium resistance thermometers from three manufacturers have been investigated experimentally at temperatures from 4.2 to 120 K. The experimental results agree with a published theoretical analysis. The heat transfer coefficient for these thermometers is found to vary by a factor of 50 for different types and different temperatures. In some thermometers, conduction through the gas is greater than that through the solid leads and the influence of contamination of the gas at temperatures between 45 and 65 K is very significant. However, in other thermometers solid conduction is the dominant heat transfer mechanism. It is suggested that solid conduction should be enhanced in the design of germanium resistance thermometers for use in the liquid‐nitrogen temperature range.

Analysis of internal heat transfer in germanium resistance thermometers

A theoretical analysis of internal heat transfer processes and their variation with temperature over the range 4.2–100 K is made for a typical commercial germanium resistance thermometer. It is shown that the proportion of the heat transfer occurring through the solid materials in the thermometer compared to that taking place in the gas trapped in the thermometer capsule varies substantially with temperature. The factors influencing heat transfer include the construction and the size of the thermometer, the thermal properties of materials, and the position of the germanium crystal within its capsule. Where heat transfer through the gas in the capsule is important, any changes in the composition of the gas play a significant role through the consequent changes in its thermal conductivity.

Stability studies on Chinese germanium resistance thermometers designed for use in magnetic fields

The stability of 15 doubly doped germanium resistance thermometers, manufactured in China and designed for use in magnetic fields, has been studied under zero magnetic field conditions while the thermometers were subjected to thermal cycling between 15 K and room temperature and between 85 K and room temperature. Resistance measurements were made on each thermometer at 15 and 20 K on the 15 K cycles and at 85 K on the 85 K cycles. The stability of the thermometers at 20 K was found to range from 1 to 300 mK. Major differences in behavior between thermometers made from n‐ and p‐type doped germanium were observed; many of the n‐type thermometers showed good stability while there were significant resistance changes for all of the p‐type thermometers.

Stability data for germanium resistance thermometers at three temperatures

Twenty‐four germanium resistance thermometers have been cycled 30 times between 15 K and 300 K with resistance measurements being made at 15 K, 20 K and 27 K on each cycle. Stabilities ranging from 0.1 mK to 10 mK at 20 K were observed with both drifts and jumps in the resistance of some thermometers. In most cases, the proportional changes in the resistance of a particular thermometer were the same at each of the three test temperatures. The effect of possible strain introduced by an epoxy heat sink on the thermometer was found to be insignificant.