Transient surface heat flux measurements were performed at several locations on the cylinder head of a divided-chamber diesel engine. The local heat flux histories were found to be significantly different. These differences are attributed to the spatial nonuniformity of the fluid motion and combustion. Both local time-averaged and local peak heat fluxes decreased with decreasing speed and load. Retarding the combustion timing beyond TDC decreased the peak heat flux in the antechamber but increased the peak heat flux in the main chamber. This is attributed to the relative increase in the portion of fuel that burns in the main chamber with retarded combustion timing.
To address the requirement for prediction and understanding of airflow in forced convection cooled electronic systems, a detailed experimental investigation of the outlet flow of typical axial cooling fans has been performed. The flow is shown to be complex over much of the fans operational range, with significant radial and tangential velocities and regions with little or no flow. The effect of partially blocking a fan and running it at elevated temperatures are both shown to be significant. The effect of attaching a fan to an electronic system is then investigated. Flow drawn through a system is shown to be simple and well predicted by a standard CFD package. Flow blown into a system is far more complex, with large areas of recirculating flow, and less accuracy in the prediction. The paper gives valuable and novel design insight into forced cooling flows in electronic systems and shows that the industry is still some way from a reliable design method.
The thermal resistance of electronic components is known to often differ considerably between standard test conditions and those found in service. One way to correct for this is to use multi-parameter thermal resistances. Another, presented here, is to adjust the junction-to-ambient thermal resistance to account for operational conditions. For forced convection applications, two factors are proposed; the first accounts for any upstream aerodynamic disturbance and the second addresses purely thermal interaction. Thus if an upstream powered component interacts with a downstream component, the two factors are combined. It is shown that both factors may be quantified in terms of readily measured temperatures and then used as coefficients to adjust the standard thermal resistance data for operational conditions. To overcome the misconception that the currently published single-value thermal resistances are solely a property of the electronic package, thermal resistance is redefined to include both the resistance of the package and the part of the printed circuit board (PCB) covered by the component thermal footprint. This approach is applied to a symmetrical array of board mounted 160-lead devices and data showing how the factors vary with component position, nondimensional power distribution and Reynolds number is presented. Based on data a new method of generating operational component thermal resistances is proposed. [S1043-7398(00)00603-4]
Electronic package manufacturers publish thermal characteristics of components, which are measured using standard tests, measuring a thermal resistance value for a single component on a standard test printed circuit board (PCB). This limits the applicability of the characterization, as it does not show what aerodynamic or thermal interaction each package will have in a real system. This paper presents a new board-level electronics system test vehicle consisting of an array of ball grid components on three different effective thermal conductivity multi-layer PCB’s. Aerodynamic and thermal measurements are presented. It appears that PCB’s populated with low profile electronic packages behave like flat plates, leading to the proposition that component temperatures can be calculated using flat plate predictions. It is shown how both the airflow and the board conductivity can have a critical effect on the junction temperature, and a simple design rule is suggested, in terms of influence factors, to take account of these effects. These will lead to better estimates of electronic system reliability in the early part of the design cycle.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.