Search

We studied the transient local surface temperature response that occurs when jet impinging of nitrogen gas is applied to the surface to which a pulsating heat flux is applied. After manufacturing a micro jet impinging device, which was composed of three parts—a silicon wafer, a vinyl sticker, and a Pyrex wafer—nitrogen gas was used as working fluid and the velocity was 344.8 m/s, corresponding jet Reynolds was 1467. And the amplitudes of heat fluxes varied from 5.67 to 16.17 W/cm2, the frequency also varied from 0.1 to 1000 Hz, then temperature response of RTDs was acquired. For example, when a heat flux, 8.5 W/cm2, was applied with a frequency of 0.1 Hz, the highest and lowest temperatures of RTD located on the center of the heater and under indirect jets were 30.2 °C and 20.7 °C. From the acquired results, the time constant of the device was estimated between 17 ms and 32 ms, and this result shows the conjugated mode of conduction and convection of heat transfer. To understand the dominant heat transfer, we decoupled the conduction mode and the convective heat transfer mode through analytical calculations and confirmed that the time constants were 11 µs and 128 ms, respectively. In addition, the heat transfer coefficient was inversely calculated through numerical calculation by simulating the RTD manufactured in the device, and through this, it was confirmed that the dimensionless time constant was related to the Nu number and the correlation was developed as τ/t = 3.51Nu−1.067 for f ≤ 1 Hz. [ABSTRACT FROM AUTHOR]

We studied the transient local surface temperature response that occurs when jet impinging of nitrogen gas is applied to the surface to which a pulsating heat flux is applied. After manufacturing a micro jet impinging device, which was composed of three parts—a silicon wafer, a vinyl sticker, and a Pyrex wafer—nitrogen gas was used as working fluid and the velocity was 344.8 m/s, corresponding jet Reynolds was 1467. And the amplitudes of heat fluxes varied from 5.67 to 16.17 W/cm2, the frequency also varied from 0.1 to 1000 Hz, then temperature response of RTDs was acquired. For example, when a heat flux, 8.5 W/cm2, was applied with a frequency of 0.1 Hz, the highest and lowest temperatures of RTD located on the center of the heater and under indirect jets were 30.2 °C and 20.7 °C. From the acquired results, the time constant of the device was estimated between 17 ms and 32 ms, and this result shows the conjugated mode of conduction and convection of heat transfer. To understand the dominant heat transfer, we decoupled the conduction mode and the convective heat transfer mode through analytical calculations and confirmed that the time constants were 11 µs and 128 ms, respectively. In addition, the heat transfer coefficient was inversely calculated through numerical calculation by simulating the RTD manufactured in the device, and through this, it was confirmed that the dimensionless time constant was related to the Nu number and the correlation was developed as τ/t = 3.51Nu−1.067 for f ≤ 1 Hz. [ABSTRACT FROM AUTHOR]

Micro heat exchangers have been widely used in both of academic and industrial fields for controlling heat for decades. Especially, the printed circuit heat exchanger (PCHE) is applicable to high pressure and high temperature cases since the manufacturing process, diffusion bonding, gives high durability. However, flow does not occur uniformly in multiple channels, which causes a heat transfer difference between the channels that reduces the effectiveness of the system. Also, it makes the total pressure drop larger than when it is uniformly distributed. Flow distribution has been studied intensively for decades, but the mechanism has not been clarified yet, since flow in the header is much too complicated. As a step to reveal the mechanism, we focus on maldistribution and quantitatively investigate the phenomena. We prepared experimental devices, including a set of parallel micro-channels, inlet and outlet. We used water as a working fluid, and water soluble red ink for visualization. We measured the interface location to derive the flow speed at each channel. We focused on the dependency of channel length and flow speed. In the theoretical part, we derived the flow speed profile of the entire channels that shows good agreement with the experimental results. This study provides a theoretical basis on resolving the maldistribution problem. [ABSTRACT FROM AUTHOR]