3D Integrated Water Cooling of a Composite Multilayer Stack of Chips

2010 ◽  
Vol 132 (12) ◽  
Author(s):  
Fabio Alfieri ◽  
Manish K. Tiwari ◽  
Igor Zinovik ◽  
Dimos Poulikakos ◽  
Thomas Brunschwiler ◽  
...  

New generation supercomputers with three dimensional stacked chip architectures pose a major challenge with respect to the removal of dissipated heat, which can reach currently as high as 250 W/cm2 in multilayer chip stacks of less than 0.3 cm3 volume. Interlayer integrated water cooling is a very promising approach for such high heat flux removal due to much larger thermal capacity and conductivity of water compared with air, the traditional cooling fluid. In the current work, a multiscale conjugate heat transfer model is developed for integrated water cooling of chip layers and validated with experimental measurements on an especially designed thermal test vehicle that simulates a four tier chip stack with a footprint of 1 cm2. The cooling heat transfer structure, which consists of microchannels with cylindrical pin-fins, is conceived in such a way that it can be directly integrated with the device layout in multilayer chips. Every composite layer is cooled by water flow in microchannels (height of 100 μm), which are arranged in two port water inlet-outlet configuration. The total power removed in the stack is 390 W at a temperature gradient budget of 60 K from liquid inlet to maximal junction temperature, corresponding to about 1.3 kW/cm3 volumetric heat flow. The computational cost and complexity of detailed computational fluid dynamics (CFD) modeling of heat transfer in stacked chips with integrated cooling can be prohibitive. Therefore, the heat transfer structure is modeled using a porous medium approach, where the model parameters of heat transfer and hydrodynamic resistance are derived from averaging the results of the detailed 3D-CFD simulations of a single streamwise row of fins. The modeling results indicate that an isotropic porous medium model does not accurately predict the measured temperature fields. The variation of material properties due to temperature gradients is found to be large; therefore, variable properties are used in the model. It is also shown that the modeling of the heat transfer in the cooling sublayers requires the implementation of a porous medium approach with a local thermal nonequilibrium, as well as orthotropic heat conduction and hydrodynamic resistance. The improved model reproduces the temperatures measured in the stack within 10%. The model is used to predict the behavior of multilayer stacks mimicking the change of heat fluxes resulting from variations in the computational load of the chips during their operation.

Author(s):  
Fabio Alfieri ◽  
Manish K. Tiwari ◽  
Igor Zinovik ◽  
Dimos Poulikakos ◽  
Thomas Brunschwiler ◽  
...  

New generation supercomputers with three dimensional stacked chip architectures pose a major challenge with respect to removal of dissipated heat which can reach currently as high as 250 W/cm2 in multilayer chip stacks of less than 0.3 cm3 volume. Interlayer integrated water cooling [1] is a very promising approach for such high heat flux removal due to much larger thermal capacity and conductivity of water compared to air, the traditional cooling fluid. In the current work, a multiscale conjugate heat transfer model is developed for integrated water cooling of chip layers and validated with experimental measurements on a specially designed thermal test vehicle that simulates a four tier chip stack with a footprint of 1 cm2. The cooling heat transfer structure, which consists of microchannels with cylindrical pin fins, is conceived in such a way that it can be directly integrated with the device layout in multilayer chips. Every composite layer is cooled by water flow in microchannels (height: 100 μm), which are arranged in 2 port water inlet-outlet configuration. The total power removed in the stack is 390 W at a temperature gradient budget of 60 K from liquid inlet to maximal junction temperature, corresponding to about 1.3 kW/cm3 volumetric heat flow. The computational cost and complexity of detailed CFD modeling of heat transfer in stacked chips with integrated cooling can be prohibitive. Therefore, the heat transfer structure is modeled using a porous medium approach, where the model parameters of heat transfer and hydrodynamic resistance are derived from averaging the results of the detailed 3D-CFD simulations of a single stream-wise row of fins. The modeling results indicate that an isotropic porous medium model does not accurately predict the measured temperature fields. The variation of material properties due to temperature gradients are found to be large, therefore variable properties are used in the model. It is also shown that the modeling of the heat transfer in the cooling sublayers requires the implementation of a porous medium approach with a local thermal non-equilibrium as well as orthotropic heat conduction and hydrodynamic resistance. The improved model reproduces the temperatures measured in the stack within 10%. The model is used to predict the behavior of multilayer stacks mimicking the change of heat fluxes resulting from variations in the computational load of the chips during their operation.


2020 ◽  
Author(s):  
Jakob Sablowski ◽  
Simon Unz ◽  
Michael Beckmann

In recent years, established heat transfer models for dropwise condensation (DWC) have been refined to consider the influence of wetting behavior, surface structure and nucleation dynamics on the heat transfer rate in more detail. Despite these efforts to develop more sophisticated models, uncertainties of the model parameters still lead to a high variation of the calculated heat transfer rate. In this study, we apply quantitative sensitivity analysis to a pure steam DWC heat transfer model in order to attribute the variation of the model result to its input parameters. Four scenarios with different variations of the model parameters are discussed and sensitivity coefficients for each parameter are calculated. Our results show that the contact angle and the nucleation site density have the greatest influence on the model result if no additional coating layer is considered. The influence of the nucleation site density is mainly due to the large uncertainties associated with this parameter. In scenarios with an additional coating layer, the heat flux is mainly governed by the thickness of the coating layer, underlining the need for thin conformal coatings to effectively promote DWC. Furthermore, trends within the heat transfer model are discussed and beneficial conditions for a high heat flux are identified for each scenario. The results underline that the wetting properties of functional surfaces should be tailored towards medium contact angles in the range of 70° to 130° and low contact angle hystereses below 10° to achieve high heat flux DWC.


2015 ◽  
Vol 9 (3) ◽  
pp. 161-166
Author(s):  
Aneta Bohojło-Wiśniewska

Summary This paper presents an example of humid air flow around a single head of Chinese cabbage under conditions of complex heat transfer. This kind of numerical simulation allows us to create a heat and humidity transfer model between the Chinese cabbage and the flowing humid air. The calculations utilize the heat transfer model in porous medium, which includes the temperature difference between the solid (vegetable tissue) and fluid (air) phases of the porous medium. Modelling and calculations were performed in ANSYS Fluent 14.5 software.


Author(s):  
Kazuhisa Yuki ◽  
Masahiro Uemura ◽  
Koichi Suzuki ◽  
Ken-ichi Sunamoto

Two-phase flow loop system using a metal porous heat sink is proposed as a cooling system of the future power electronic devices with a heat load exceeding 300W/cm2. In this paper, as the first step, the heat transfer performance of the porous heat sink is evaluated under high heat flux conditions and the applicability and some engineering issues are discussed. The porous medium, which is fabricated by sintering copper particles, has a functional structure with several sub-channels inside it to enhance phase-change as well as discharge of generated vapor outside the porous medium. This porous heat sink is attached onto a heating chip and removes the heat by evaporating cooling liquid passing through the porous medium against the heat flow. Experiments using 30 kW of heating system show that the heat transfer performance of a copper-particles-sintered porous medium with the sub-channels exceeds 800W/cm2 in both high and low subcooling cases and achieves 300W/cm2 at a wall temperature of 150 °C (Tin = 70 °C) and 130 °C (Tin = 70 °C). These results prove that this porous heat sink is applicable enough for cooling 300 W/cm2 class of power electronic devices.


2021 ◽  
Vol 2021 ◽  
pp. 1-14
Author(s):  
Saba Javaid ◽  
Asim Aziz

The present work covers the flow and heat transfer model for the power-law nanofluid in the presence of a porous medium over the penetrable plate. The flow is caused by the impulsive movement of the plate embedded in Darcy’s type porous medium. The flow and heat transfer model has been examined with the effect of linear thermal radiation and the internal heat source or sink in the flow regime. The Rosseland approximation is utilized for the optically thick nanofluid. To form the closed-form solutions for the governing partial differential equations of conservation of mass, momentum, and energy, the Lie symmetry analysis is used to get the reductions of governing equations and to find the group invariants. These invariants are then utilized to obtain the exact solution for all three cases, i.e., shear thinning fluid, Newtonian fluid, and shear thickening fluid. In the end, all solutions are plotted for the cu -water nanofluid and discussed briefly for the different emerging flow and heat transfer parameters.


Author(s):  
Mahdi Abbasi ◽  
Mohammad Ahmadi ◽  
Alireza Kazemi ◽  
Mohammad Sharifi

Global warming and reducing fossil fuel resources have increased the interest in using renewable resources such as geothermal energy. In this paper, in the first step, heat transfer equations have been presented for reservoir during water (steam) injection by considering heat loss to adjacent formations. According to radius of thermal front, the reservoir is partitioned into two regions with different fluid physical properties. The heat transfer model is coupled with a fluid flow model which is used to calculate the reservoir pressure or fluid flow rates. Then by calculating outer radius of heated region and using radial composite reservoir model, the fluid flow equations in porous media are solved. Using pressure derivative plot in regions with different thermal conductivity coefficients, a type curve plot is presented. The reservoir and adjacent formation thermal conductivity coefficients can be calculated by matching the observed pressure data on the thermal composite type curve. Additionally, the interference test in composite geothermal reservoir is discussed. In the composite reservoir model, parameters such as diffusivity coefficient, conductivity ratio and the distance to the radial discontinuity are considered. New type curves are provided to introduce the effect of diffusivity/conductivity contrast ratios on temperature behavior. Improving interpretations, and performing fast computations and fast sensitivity analysis are the benefits of the presented solutions.


Author(s):  
Gurveer Singh ◽  
Vishwa Deepak Kumar ◽  
Laltu Chandra ◽  
R. Shekhar ◽  
P. S. Ghoshdastidar

Abstract The open volumetric air receiver (OVAR)-based central solar thermal systems provide air at a temperature > 1000 K. Such a receiver is comprised of porous absorbers, which are exposed to a high heat-flux > 800 Suns (1 Sun = 1 kW/m2). A reliable assessment of heat transfer in an OVAR is necessary to operate such a receiver under transient conditions. Based on a literature review, the need for developing a comprehensive, unsteady, heat transfer model is realized. In this paper, a seven-equations based, one-dimensional, zonal model is deduced. This includes heat transfer in porous absorber, primary-air, return-air, receiver casing, and their detailed interaction. The zonal model is validated with an inhouse experiment showing its predictive capability, for unsteady and steady conditions, within the reported uncertainty of ±7%. The validated model is used for investigating the effect of operating conditions and absorber geometry on the thermal performance of an absorber. Some of the salient observations are (a) the maximum absorber porosity of 70–90% may be preferred for non-volumetric and volumetric-heating conditions, (b) the minimum air-return ratio should be 0.7, and (c) the smallest gap to absorber-length ratio of 0.2 should suffice. Finally, suggestions are provided for extending the model.


2001 ◽  
Author(s):  
Marc Sulfridge ◽  
Taher Saif ◽  
Norman Miller ◽  
Keith O’Hara

Abstract This paper presents experimental evidence that MEMS devices may be manipulated using beams of light. Light possesses momentum, and hence it imparts a force equal to 2W/c when perfectly reflected by a surface. Here W is the total power of the reflected light, and c is the speed of light. The radiation pressure of light can be quite significant to MEMS devices. This actuation method is demonstrated, both in air and in vacuum, by switching the state of a bi-stable MEMS device. The associated heat transfer model is also presented.


2021 ◽  
Vol 2116 (1) ◽  
pp. 012012
Author(s):  
Jakob Sablowski ◽  
Simon Unz ◽  
Michael Beckmann

Abstract Established heat transfer models for dropwise condensation (DWC) consider wetting behavior, surface structure and nucleation dynamics to calculate the heat flux. However, model results often deviate from experiments, in part due to uncertainties of the model input parameters. In this study, we apply quantitative sensitivity analysis to a pure steam DWC heat transfer model in order to attribute the variation of the model result to its input parameters. Four scenarios with different variations of the model parameters are discussed and sensitivity coefficients for each parameter are calculated. Our results show a high sensitivity of the model result towards the coating thickness, the contact angle and the nucleation site density, underlining the need to accurately determine these parameters in DWC experiments.


2011 ◽  
Vol 366 ◽  
pp. 229-233
Author(s):  
Chang Gui Cheng ◽  
Le Yu ◽  
Wen Cheng Wan ◽  
Zhong Tian Liu

The paper has established a two-dimensional non-steady-state heat transfer model for 2024 aluminum alloy solidification process under the static magnetic field. According to the measured temperature and the inverse heat transfer problem, the boundary conditions of model have been determined, the results show that the solidification rate and heat flux acting on the ingot surface increase with increasing of the static magnetic filed strength; when the static magnetic filed strength become stronger, the isotherm location will move towards the liquid pool center, and the temperature gradient in the liquid metal pool will increase.


Sign in / Sign up

Export Citation Format

Share Document