Experimental setup for evaluating four-stage thermoelectric cooler's performance was designed. Effects of input power, heat dissipation condition and heat load on the temperature difference (ΔT) of four-stage thermoelectric coolers' hot and cold faces were obtained experimentally. The result shows that, the ΔT increases as the input power increases. A linear relationship exists between input current and feedback voltage. In different cooling conditions, the ΔT of thermoelectric cooler (TEC) increases with the temperature of hot face. As the temperature increasing on hot face is 1K, the ΔT increasing of TEC can be about 0.5K. Meanwhile, the power consumption of TEC also increases slightly. Water condensation can be prevented in either dry nitrogen environment or vacuum environment, but the vacuum level has great influence to the ΔT, especially in low operation temperature. The better the vacuum level is, the smaller the convection heat loss has. When the operation temperature of focal plane array (FPA) is lower than 220K, it is prior to use vacuum packaging. Considering the Joule-heat of readout circuit and the heat loss of wire conduction, the minimum working temperature of FPA can reach below 200 K when the temperature of the hot face is 285K. And the coefficient of performance (COP) of TEC can increase sharply from 0.8% to 4% when the controlled operation temperature is 220K rather than 200K.

No abstract

No abstract

The conflict of longevity of satellite's service and limited life of Sterling cooler decides that coolers should work on the intermittent mode in space. As a result, The HgCdTe (MCT) infrared (IR) detectors in satellite are commonly subjected to thousands of repeated thermal cycles from below -173℃ to room temperature (20℃), which brings some new reliability problems. Especially the mismatch of coefficient of thermal expansion (CTE) of different materials may lead to some unfamiliar failure modes with such low temperature and nearly 200℃ span of thermal cycles. In order to study the characteristics of MCT detectors under the stress of thermal cycles, this paper introduced a special automatic system. The system is mainly composed of a sub-container of liquid nitrogen, a heater controlled by the PID hardware, and an object stage on which the MCT detectors to be tested are mounted. Furthermore, the sub-container, the heater and the stage are positioned in a large vacuum tank. In the course of thermal cycles, the object stage moved up and down with MCT detectors is driven by a step motor. When it rises to the bottom of liquid nitrogen sub-container, the stage is to be cooled with detectors, and when declines to the heater, the stage to be heated with detectors, too. At last, two long wavelength MCT detector samples are tested with this equipment, and the resistance, the signal and the noise are measured．It shows that all the pixels' resistance didn't change beyond 5% after 5000 cycles. However, the tested signal of the last pixel of both detectors increased sharply after 1000 cycles, and fell to normal level after 5000 cycles, with its noise altering a little from beginning to end. A deduction is given in this paper for this phenomenon. In accordance, the thermal cycle equipment and the experimental data, would supply some references to the design and fabrication of MCT IR detectors.