Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517701
Xianguang Tan, Guofeng Chen, Jiajun Zhang, Chao Liu, N. Ahuja, Jun Zhang
Data center, as backbone of cloud computing, has been rapidly developing and evolving. Server with high reliability, as one of key infrastructure ingredients for computing, storage and networking segments, is obviously foremost footstone to data center robustness. Based on data collection and failure analysis from Baidu infrastructure maintenance group, 78% of server system failure comes from hard drive. So, effectively reducing hard drive failure rate is crucial to server system reliability. Within the academia as well journals not much data exists around Rotational Vibration (RV) performance with hard drive failure rate. Data collected from Baidu data center proves vibration is the major cause of hardware failure. Data collected between general server system and rack server system shows hard drive failure rate in general server system is up to 2.15%; versus 0.71 % in rack server system. This paper addresses optimal design of cooling fans, hard drive, hard drive brackets, chassis and system structure for Rotational Vibration (RV) performance. It then, introduces an advanced rack server system design optimized for Rotational Vibration performance; finally, uses comparison data to prove RV performance of rack server system design is superior to general purpose server system.
{"title":"An advanced rack server system design For Rotational Vibration (RV) performance","authors":"Xianguang Tan, Guofeng Chen, Jiajun Zhang, Chao Liu, N. Ahuja, Jun Zhang","doi":"10.1109/ITHERM.2016.7517701","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517701","url":null,"abstract":"Data center, as backbone of cloud computing, has been rapidly developing and evolving. Server with high reliability, as one of key infrastructure ingredients for computing, storage and networking segments, is obviously foremost footstone to data center robustness. Based on data collection and failure analysis from Baidu infrastructure maintenance group, 78% of server system failure comes from hard drive. So, effectively reducing hard drive failure rate is crucial to server system reliability. Within the academia as well journals not much data exists around Rotational Vibration (RV) performance with hard drive failure rate. Data collected from Baidu data center proves vibration is the major cause of hardware failure. Data collected between general server system and rack server system shows hard drive failure rate in general server system is up to 2.15%; versus 0.71 % in rack server system. This paper addresses optimal design of cooling fans, hard drive, hard drive brackets, chassis and system structure for Rotational Vibration (RV) performance. It then, introduces an advanced rack server system design optimized for Rotational Vibration performance; finally, uses comparison data to prove RV performance of rack server system design is superior to general purpose server system.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132168558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517573
Yangying Zhu, Zhengmao Lu, D. Antao, Hongxia Li, Tiejun Zhang, E. Wang
Capillary-driven thin film evaporation in wick structures is promising for thermal management of high-power electronics because it harnesses the latent heat of evaporation without the use of an external pumping power. The complexities associated with liquid-vapor interface and liquid flow through the wick structures, however, make it challenging to optimize the wick structure geometries to boost the dry-out heat flux. In this work, we developed a numerical model to predict the dry-out heat flux of thin film evaporation from micropillar array wick structures. The model simulates liquid velocity, pressure, meniscus curvature and contact angle along the length of the wick surface through conservation of mass, momentum and energy, based on a finite volume approach. In particular, we captured the three-dimensional meniscus shape, which varies along the wicking direction, by solving the Young-Laplace equation. We determined the dry-out heat flux upon the condition that the minimum contact angle on the micropillar surface reaches the receding contact angle. With this model, we calculated the dry-out heat flux as a function of micropillar structure geometries (diameter, pitch and height), and optimized the geometry to maximize the dry-out heat flux. Our model provides an understanding of the role of the wick structures in capillary-driven thin film evaporation and offers important design guidelines for thermal management of high-performance electronic devices.
{"title":"Model optimization of dry-out heat flux from micropillar wick structures","authors":"Yangying Zhu, Zhengmao Lu, D. Antao, Hongxia Li, Tiejun Zhang, E. Wang","doi":"10.1109/ITHERM.2016.7517573","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517573","url":null,"abstract":"Capillary-driven thin film evaporation in wick structures is promising for thermal management of high-power electronics because it harnesses the latent heat of evaporation without the use of an external pumping power. The complexities associated with liquid-vapor interface and liquid flow through the wick structures, however, make it challenging to optimize the wick structure geometries to boost the dry-out heat flux. In this work, we developed a numerical model to predict the dry-out heat flux of thin film evaporation from micropillar array wick structures. The model simulates liquid velocity, pressure, meniscus curvature and contact angle along the length of the wick surface through conservation of mass, momentum and energy, based on a finite volume approach. In particular, we captured the three-dimensional meniscus shape, which varies along the wicking direction, by solving the Young-Laplace equation. We determined the dry-out heat flux upon the condition that the minimum contact angle on the micropillar surface reaches the receding contact angle. With this model, we calculated the dry-out heat flux as a function of micropillar structure geometries (diameter, pitch and height), and optimized the geometry to maximize the dry-out heat flux. Our model provides an understanding of the role of the wick structures in capillary-driven thin film evaporation and offers important design guidelines for thermal management of high-performance electronic devices.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128821162","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517653
Tailian Chen
Evaporation of liquid droplets on a heated substrate is an important process in numerous engineering applications, during which the energy transport is dependent upon, among many others, the liquid/substrate wetting characteristics. In this work, transient heat transfer to a liquid droplet deposited on a heated metallic surface with multiple parallel microgrooves were experimentally investigated from distributed temperature measurements underneath the microgrooves. The initial heat conduction following deposition of a liquid droplet causes a sharp decease in the substrate temperature, during which a fleeting but notable temperature plateau for both cases of alcohol and water droplets is likely attributed to a thin layer of vapor formed in between the droplet and the substrate. Depending on the wetting characteristics, the transient heat transfer process is drastically different for the cases of alcohol and water. Deposition of an alcohol droplet is followed by the droplet instantaneous spreading on the microgroove fins and liquid penetration into the microgrooves, leading to continued temperature decrease in the substrate as a result of formation and evaporation of liquid thin films. It takes only nearly half a second for complete evaporation of the deposited alcohol, at which the substrate reaches its lowest temperature in the process. As a water droplet is deposited, it takes about 7 minutes for its complete evaporation, during which the substrate temperature experiences four distinct stages corresponding to evolution of the water droplet on the substrate. The results in this work provide insights into the fundamental physics of heat transfer during evaporation of liquid droplets with different liquid/substrate wetting characteristics.
{"title":"Heat transfer to wetting and non-wetting liquid droplets deposited onto a heated microgroove surface","authors":"Tailian Chen","doi":"10.1109/ITHERM.2016.7517653","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517653","url":null,"abstract":"Evaporation of liquid droplets on a heated substrate is an important process in numerous engineering applications, during which the energy transport is dependent upon, among many others, the liquid/substrate wetting characteristics. In this work, transient heat transfer to a liquid droplet deposited on a heated metallic surface with multiple parallel microgrooves were experimentally investigated from distributed temperature measurements underneath the microgrooves. The initial heat conduction following deposition of a liquid droplet causes a sharp decease in the substrate temperature, during which a fleeting but notable temperature plateau for both cases of alcohol and water droplets is likely attributed to a thin layer of vapor formed in between the droplet and the substrate. Depending on the wetting characteristics, the transient heat transfer process is drastically different for the cases of alcohol and water. Deposition of an alcohol droplet is followed by the droplet instantaneous spreading on the microgroove fins and liquid penetration into the microgrooves, leading to continued temperature decrease in the substrate as a result of formation and evaporation of liquid thin films. It takes only nearly half a second for complete evaporation of the deposited alcohol, at which the substrate reaches its lowest temperature in the process. As a water droplet is deposited, it takes about 7 minutes for its complete evaporation, during which the substrate temperature experiences four distinct stages corresponding to evolution of the water droplet on the substrate. The results in this work provide insights into the fundamental physics of heat transfer during evaporation of liquid droplets with different liquid/substrate wetting characteristics.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"2015 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128902154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517596
P. Lall, A. Abrol, Lee Simpson, J. Glover
Traditional Quartz based oscillators still outnumber their MEMS counterparts in the industry therefore no extensive prior studies exist which provide harsh environment reliability data for Silicon oscillators. MEMS based oscillators serve as clocks which control the timing in electronics, a better clock signal ensures higher performance, more consistent behavior and reliable operation. Harsh environment applications such as under the hood automotive, military, space navigation all make use of MEMS oscillators. None of the previous studies look into the impact of sequential harsh environment operating conditions. Survivability of MEMS oscillators at high relative humidity and high G environments is unknown. The effects of these pre-conditions along with the drop test conditions have been studied and analyzed. Anomalies in the oscillator behavior due to the presence of harsh environments lead to mismatch in the electronic timing of the circuit resulting in a bad consumer product, thus the importance of reliability data. In this paper a test vehicle with a MEMS oscillator, SiT 8103, has been tested under: high relative temperature humidity exposure and then followed by subjection to high-g shock loading environments. The test boards have been subjected to mechanical shocks using the method 2002.5, condition G, under the standard MIL-STD-883H test. The effect of temperature, humidity and shock on the oscillator has been studied. The survivability of SiT 8103 has been demonstrated as a function of change in the output frequency, rise/fall time(s) and duty cycle. Later the deterioration in oscillator output parameters has been characterized using the techniques of Fast Fourier Transform and Principal Component Analysis. The results obtained show that exposure to sequential high relative temperature-humidity and high-g shock affects the working of Silicon MEMS oscillators more than just the high-g shock environment. Rise and fall times, Output frequency and Duty cycle show more deterioration and drift in the 85°C/85%RH cases on comparison with their pristine counterparts. The energy spectrum data obtained after conducting the FFT analysis demonstrate that 85°C/85%RH samples have lower peak amplitudes/signal energy than the pristine samples especially during the first 50 drops.
{"title":"A study on damage progression in MEMS based Silicon oscillators subjected to high-g harsh environments","authors":"P. Lall, A. Abrol, Lee Simpson, J. Glover","doi":"10.1109/ITHERM.2016.7517596","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517596","url":null,"abstract":"Traditional Quartz based oscillators still outnumber their MEMS counterparts in the industry therefore no extensive prior studies exist which provide harsh environment reliability data for Silicon oscillators. MEMS based oscillators serve as clocks which control the timing in electronics, a better clock signal ensures higher performance, more consistent behavior and reliable operation. Harsh environment applications such as under the hood automotive, military, space navigation all make use of MEMS oscillators. None of the previous studies look into the impact of sequential harsh environment operating conditions. Survivability of MEMS oscillators at high relative humidity and high G environments is unknown. The effects of these pre-conditions along with the drop test conditions have been studied and analyzed. Anomalies in the oscillator behavior due to the presence of harsh environments lead to mismatch in the electronic timing of the circuit resulting in a bad consumer product, thus the importance of reliability data. In this paper a test vehicle with a MEMS oscillator, SiT 8103, has been tested under: high relative temperature humidity exposure and then followed by subjection to high-g shock loading environments. The test boards have been subjected to mechanical shocks using the method 2002.5, condition G, under the standard MIL-STD-883H test. The effect of temperature, humidity and shock on the oscillator has been studied. The survivability of SiT 8103 has been demonstrated as a function of change in the output frequency, rise/fall time(s) and duty cycle. Later the deterioration in oscillator output parameters has been characterized using the techniques of Fast Fourier Transform and Principal Component Analysis. The results obtained show that exposure to sequential high relative temperature-humidity and high-g shock affects the working of Silicon MEMS oscillators more than just the high-g shock environment. Rise and fall times, Output frequency and Duty cycle show more deterioration and drift in the 85°C/85%RH cases on comparison with their pristine counterparts. The energy spectrum data obtained after conducting the FFT analysis demonstrate that 85°C/85%RH samples have lower peak amplitudes/signal energy than the pristine samples especially during the first 50 drops.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114613164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517654
Thomas R. Boziuk, Marc K. Smith, A. Glezer
Two-phase thermal management based on submerged boiling heat transfer has received considerable attention in recent years because of its potential to enable high heat flux using relatively simple hardware and system-level coupling. However, the utility of this attractive heat transfer approach has been hampered by the critical heat flux (CHF) limit on the maximum heat transfer owing to the dynamics of the vapor bubbles that form on the heated surface and the transition to film boiling that results in a large increase in surface temperature. Recent work at Georgia Tech has exploited low-power ultrasonic acoustic forcing to enhance boiling heat transfer and increase the CHF limit by controlling the formation and evolution of the vapor bubbles and inhibiting the instabilities that lead to film boiling. These effects are investigated over both plain and textured (surface-embedded microchannels) boiling heat transfer base surfaces (the transfer of makeup fluid to the boiling sites in the presence of surface microchannels passively decreases surface superheat and increases the CHF). Acoustic actuation has a profound effect on the boiling, and leads to a significant increase in the CHF by limiting the formation of large vapor columns and their collapse into a vapor film. Improvements in the CHF in stagnant bulk fluid exceed 65% for the plain surface (up to 183 W/cm2), and 30% for the textured surface (up to 460 W/cm2 with 7°C r eduction in surface superheat).
{"title":"Enhanced boiling heat transfer on micromachined surfaces using acoustic actuation","authors":"Thomas R. Boziuk, Marc K. Smith, A. Glezer","doi":"10.1109/ITHERM.2016.7517654","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517654","url":null,"abstract":"Two-phase thermal management based on submerged boiling heat transfer has received considerable attention in recent years because of its potential to enable high heat flux using relatively simple hardware and system-level coupling. However, the utility of this attractive heat transfer approach has been hampered by the critical heat flux (CHF) limit on the maximum heat transfer owing to the dynamics of the vapor bubbles that form on the heated surface and the transition to film boiling that results in a large increase in surface temperature. Recent work at Georgia Tech has exploited low-power ultrasonic acoustic forcing to enhance boiling heat transfer and increase the CHF limit by controlling the formation and evolution of the vapor bubbles and inhibiting the instabilities that lead to film boiling. These effects are investigated over both plain and textured (surface-embedded microchannels) boiling heat transfer base surfaces (the transfer of makeup fluid to the boiling sites in the presence of surface microchannels passively decreases surface superheat and increases the CHF). Acoustic actuation has a profound effect on the boiling, and leads to a significant increase in the CHF by limiting the formation of large vapor columns and their collapse into a vapor film. Improvements in the CHF in stagnant bulk fluid exceed 65% for the plain surface (up to 183 W/cm2), and 30% for the textured surface (up to 460 W/cm2 with 7°C r eduction in surface superheat).","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116406796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517594
Chia-Cheng Chang, Hung-Te Yang, Yen-Fu Su, Yu-Ting Hong, K. Chiang
This paper discusses about the residual stress and packaging effect of three-axis micro-electro-mechanical system (MEMS) accelerometer. The 3D FEM model with modal analysis method is adopted for the resonance frequency estimation. This paper also presents a simple compensation model for trimming the offset of capacitance differentiation using the measured resonance frequency. This trimmming methodology can be realized by adjusting circuit gain in real product.
{"title":"A method to compensate packaging effects on three-axis MEMS accelerometer","authors":"Chia-Cheng Chang, Hung-Te Yang, Yen-Fu Su, Yu-Ting Hong, K. Chiang","doi":"10.1109/ITHERM.2016.7517594","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517594","url":null,"abstract":"This paper discusses about the residual stress and packaging effect of three-axis micro-electro-mechanical system (MEMS) accelerometer. The 3D FEM model with modal analysis method is adopted for the resonance frequency estimation. This paper also presents a simple compensation model for trimming the offset of capacitance differentiation using the measured resonance frequency. This trimmming methodology can be realized by adjusting circuit gain in real product.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117149077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517691
G. Pavlidis, David Mele, T. Cheng, F. Medjdoub, S. Graham
The ability to fabricate AlGaN/GaN high electron mobility transistors (HEMTs) on Si substrates has enabled the production of low cost high power electronics. To further enhance the performance of GaN electronics for high power conversion, the ability to maintain high off-state breakdown voltages with large electron densities is necessary. The use of Si substrates, however, limits the device's capabilities due to its weak electrical field strength. This limitation has been identified as the main cause for breakdown in HEMTs when the high electric field reaches the silicon substrate underneath the region between the gate and drain. To overcome this obstacle, removal of the Si substrate between the gate and drain region has shown to increase the device's breakdown voltage up to 3000 V. While removing the Si substrate extends the capabilities of GaN HEMTs for high voltage applications, the effects of the Si removal on the thermal performance during operation has not yet been investigated. Raman Thermometry, a well-developed technique, is used to compare the maximum temperature rise between a Local Substrate Removed (LSR) device and a non-LSR device. The application of nanoparticles (TiO2 and ZnO) for measuring surface temperatures via Raman spectroscopy is also investigated and applied to determine a more accurate temperature of the gate junction temperature. The LSR device was found to have a much higher thermal resistance than its non-LSR device counterpart limiting the maximum power dissipation the LSR device can achieve before severe degradation. Volumetric averaged residual stress mapping was also measured via Raman Spectroscopy and suggests the removal of the Si relaxes the stress in the GaN buffer layer and AlGaN barrier which can be exploited in designs to improve reliability. Methods to improve the thermal reliability of LSR devices are key to implementing such devices as future power switches.
{"title":"The thermal effects of substrate removal on GaN HEMTs using Raman Thermometry","authors":"G. Pavlidis, David Mele, T. Cheng, F. Medjdoub, S. Graham","doi":"10.1109/ITHERM.2016.7517691","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517691","url":null,"abstract":"The ability to fabricate AlGaN/GaN high electron mobility transistors (HEMTs) on Si substrates has enabled the production of low cost high power electronics. To further enhance the performance of GaN electronics for high power conversion, the ability to maintain high off-state breakdown voltages with large electron densities is necessary. The use of Si substrates, however, limits the device's capabilities due to its weak electrical field strength. This limitation has been identified as the main cause for breakdown in HEMTs when the high electric field reaches the silicon substrate underneath the region between the gate and drain. To overcome this obstacle, removal of the Si substrate between the gate and drain region has shown to increase the device's breakdown voltage up to 3000 V. While removing the Si substrate extends the capabilities of GaN HEMTs for high voltage applications, the effects of the Si removal on the thermal performance during operation has not yet been investigated. Raman Thermometry, a well-developed technique, is used to compare the maximum temperature rise between a Local Substrate Removed (LSR) device and a non-LSR device. The application of nanoparticles (TiO2 and ZnO) for measuring surface temperatures via Raman spectroscopy is also investigated and applied to determine a more accurate temperature of the gate junction temperature. The LSR device was found to have a much higher thermal resistance than its non-LSR device counterpart limiting the maximum power dissipation the LSR device can achieve before severe degradation. Volumetric averaged residual stress mapping was also measured via Raman Spectroscopy and suggests the removal of the Si relaxes the stress in the GaN buffer layer and AlGaN barrier which can be exploited in designs to improve reliability. Methods to improve the thermal reliability of LSR devices are key to implementing such devices as future power switches.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125874804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517564
K. Nolan, A. Agarwal, S. Lei, E. Dalton
The flow of shear-thinning viscoelastic fluids is investigated experimentally in a serpentine microchannel at very large Weissenberg numbers (Wi > 104) undergoing elastic instability. The effects of geometric curvature on local flow instability and the consequent heat transfer enhancement are reported. Unlike previous studies where fluids with large zero-shear viscosities (up to 300 mPa.s) were used, we employ a working fluid with a lower viscosity (η0 = 9 mPa.s) more suited to microfluidic heat transfer applications while exhibiting viscoelastic characteristics. This results in Elasticity number (EI = Wi/Re) flows an order of magnitude larger than previously reported in the literature with apparent viscosities close to the solvent viscosity under flow conditions. Detailed Micro Particle Image Velocimetry (μPIV) measurements reveal the local enhancements due to instantaneous flow structures which result in vigorous local mixing at sub-critical Reynolds numbers. In addition the pressure drop increase is moderate as mixing occurs locally and the flow is maintained undisturbed elsewhere throughout the flow path.
{"title":"Mixing enhancement due to viscoelastic instability in serpentine microchannels at very large Weissenberg numbers","authors":"K. Nolan, A. Agarwal, S. Lei, E. Dalton","doi":"10.1109/ITHERM.2016.7517564","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517564","url":null,"abstract":"The flow of shear-thinning viscoelastic fluids is investigated experimentally in a serpentine microchannel at very large Weissenberg numbers (Wi > 104) undergoing elastic instability. The effects of geometric curvature on local flow instability and the consequent heat transfer enhancement are reported. Unlike previous studies where fluids with large zero-shear viscosities (up to 300 mPa.s) were used, we employ a working fluid with a lower viscosity (η0 = 9 mPa.s) more suited to microfluidic heat transfer applications while exhibiting viscoelastic characteristics. This results in Elasticity number (EI = Wi/Re) flows an order of magnitude larger than previously reported in the literature with apparent viscosities close to the solvent viscosity under flow conditions. Detailed Micro Particle Image Velocimetry (μPIV) measurements reveal the local enhancements due to instantaneous flow structures which result in vigorous local mixing at sub-critical Reynolds numbers. In addition the pressure drop increase is moderate as mixing occurs locally and the flow is maintained undisturbed elsewhere throughout the flow path.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127350246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517681
Tianyi Gao, E. Kumar, M. Sahini, Charles Ingalz, A. Heydari, Wendy Lu, Xiaogang Sun
One important motivation of data center mechanical system R&D is to improve the energy efficiency and reliability. Many new cooling solutions have been successfully used in production data centers, such as hybrid/liquid cooling systems and free cooling systems, and a better Power Usage Effectiveness (PUE) has been achieved when compared with traditional air cooling data centers. Liquid cooling can be assorted in different categories such as server liquid cooling, rack liquid cooling and pod liquid cooling. In terms of rack liquid cooling, there are several mature technologies such as a rear door heat exchanger, an in row cooler, an overhead heat exchanger, a water cooled cabinet and so on. The hybrid cooling solution can be understood as a rack liquid cooling solution operated in a hybrid environment with CRAH/CRAC units in either a raised or a non-raised floor data center. This paper proposes and investigates a new rack liquid cooling design which the cooling unit is located at the bottom of a customized server rack. The bottom cooling unit consists of an air duct and a heat exchanger. The rack is front door and back door contained, and air is moved by a fan wall installed on the back of the rack recirculating within the cabinet, passing through the cooling unit and cooling the IT. First of all, a description of the customized rack and the concept of the novel rack cooling solution is provided. Then, a thermal feasibility analysis of this proposed rack cooling solution is conducted using a combination of analytical and computational modeling. Several modeling cases are designed to characterize the sensitivities of some major design and operating parameters. The results and corresponding analyses will be used to guide the prototype development. The height of the rack cooling unit is one of the key design parameters: with a minimal height required by the cooling coil, the loss of node space on the rack can be reduced. Therefore the design and selection of the heat exchanger is of paramount importance. On one hand, the design should provide adequate cooling capacity and sufficient heat transfer area; on the other hand, the height should be minimized. The effects of the heat exchanger design on the cooling performance and air side pressure drop are modeled and analyzed quantitatively in this work. In addition, another two important design parameters namely the front door and back door containment sizes are parametrically modeled. Furthermore, the operating conditions including the chilled water supply temperature, water flow rate, fan operating duty circle are investigated and results are reported. An expected mechanical PUE of this novel rack design is proposed.
{"title":"Innovative server rack design with bottom located cooling unit","authors":"Tianyi Gao, E. Kumar, M. Sahini, Charles Ingalz, A. Heydari, Wendy Lu, Xiaogang Sun","doi":"10.1109/ITHERM.2016.7517681","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517681","url":null,"abstract":"One important motivation of data center mechanical system R&D is to improve the energy efficiency and reliability. Many new cooling solutions have been successfully used in production data centers, such as hybrid/liquid cooling systems and free cooling systems, and a better Power Usage Effectiveness (PUE) has been achieved when compared with traditional air cooling data centers. Liquid cooling can be assorted in different categories such as server liquid cooling, rack liquid cooling and pod liquid cooling. In terms of rack liquid cooling, there are several mature technologies such as a rear door heat exchanger, an in row cooler, an overhead heat exchanger, a water cooled cabinet and so on. The hybrid cooling solution can be understood as a rack liquid cooling solution operated in a hybrid environment with CRAH/CRAC units in either a raised or a non-raised floor data center. This paper proposes and investigates a new rack liquid cooling design which the cooling unit is located at the bottom of a customized server rack. The bottom cooling unit consists of an air duct and a heat exchanger. The rack is front door and back door contained, and air is moved by a fan wall installed on the back of the rack recirculating within the cabinet, passing through the cooling unit and cooling the IT. First of all, a description of the customized rack and the concept of the novel rack cooling solution is provided. Then, a thermal feasibility analysis of this proposed rack cooling solution is conducted using a combination of analytical and computational modeling. Several modeling cases are designed to characterize the sensitivities of some major design and operating parameters. The results and corresponding analyses will be used to guide the prototype development. The height of the rack cooling unit is one of the key design parameters: with a minimal height required by the cooling coil, the loss of node space on the rack can be reduced. Therefore the design and selection of the heat exchanger is of paramount importance. On one hand, the design should provide adequate cooling capacity and sufficient heat transfer area; on the other hand, the height should be minimized. The effects of the heat exchanger design on the cooling performance and air side pressure drop are modeled and analyzed quantitatively in this work. In addition, another two important design parameters namely the front door and back door containment sizes are parametrically modeled. Furthermore, the operating conditions including the chilled water supply temperature, water flow rate, fan operating duty circle are investigated and results are reported. An expected mechanical PUE of this novel rack design is proposed.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129082828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1109/ITHERM.2016.7517602
N. Lamaison, J. Marcinichen, C. L. Ong, J. Thome
This paper is the fourth part of the present study on two-phase mini-thermosyphon cooling. As mentioned in the first three parts, gravity-driven cooling systems using microchannel flow boiling can become a long-term scalable solution for cooling of datacenter servers. Indeed, the enhancement of thermal performance and the drastic reduction of power consumption together with the possibility of energy reuse and the inherent passive nature of the system offer a wide range of solutions to thermal designers. While Part 1 presented the first-of-a-kind low-height microchannel two-phase thermosyphon test results and Parts 2 and 3 showed the system scale steady and dynamic modeling and simulation results associated with this design using our inhouse simulator, Part 4 deals here with an end-user application, i.e. the cooling of a 2U server. The dynamic code of Part 3 is used to model the behavior of a mini-thermosyphon that would fit within the height of a 2U server (8.9cm high), while respecting the other geometric constraints (positions of the processors, distance of the processors to the back of the blade, etc.). Thus, the simulated system consists of two parallel multi-microchannel evaporator cold plates on the top of two chips of about 11cm2, a riser, a common water-cooled micro-condenser at the back of the blade, a liquid accumulator and a downcomer (including the piping branches to/from the two cold plates). First, an analysis of the steady-state operation highlights multiple solutions from which one is stable and one is unstable. Then, the influences of few parameters such as refrigerants, piping diameters, water coolant inlet temperature and flow rates, filling ratio and heat flux are evaluated. Simulations with unbalanced heat loads on the two chips being cooled in parallel then show the desirable flow distribution obtained in such gravity-driven systems. Finally, temporal heat load and water coolant flow rate disturbances are simulated and discussed. Noting all of these numerous influences on optimal mini-thermosyphon operation, the need for a accurate and detailed simulation code, benchmarked versus actual system tests, is seen to be imperative for attaining a good, reliable, robust design.
{"title":"Two-phase mini-thermosyphon electronics cooling, Part 4: Application to 2U servers","authors":"N. Lamaison, J. Marcinichen, C. L. Ong, J. Thome","doi":"10.1109/ITHERM.2016.7517602","DOIUrl":"https://doi.org/10.1109/ITHERM.2016.7517602","url":null,"abstract":"This paper is the fourth part of the present study on two-phase mini-thermosyphon cooling. As mentioned in the first three parts, gravity-driven cooling systems using microchannel flow boiling can become a long-term scalable solution for cooling of datacenter servers. Indeed, the enhancement of thermal performance and the drastic reduction of power consumption together with the possibility of energy reuse and the inherent passive nature of the system offer a wide range of solutions to thermal designers. While Part 1 presented the first-of-a-kind low-height microchannel two-phase thermosyphon test results and Parts 2 and 3 showed the system scale steady and dynamic modeling and simulation results associated with this design using our inhouse simulator, Part 4 deals here with an end-user application, i.e. the cooling of a 2U server. The dynamic code of Part 3 is used to model the behavior of a mini-thermosyphon that would fit within the height of a 2U server (8.9cm high), while respecting the other geometric constraints (positions of the processors, distance of the processors to the back of the blade, etc.). Thus, the simulated system consists of two parallel multi-microchannel evaporator cold plates on the top of two chips of about 11cm2, a riser, a common water-cooled micro-condenser at the back of the blade, a liquid accumulator and a downcomer (including the piping branches to/from the two cold plates). First, an analysis of the steady-state operation highlights multiple solutions from which one is stable and one is unstable. Then, the influences of few parameters such as refrigerants, piping diameters, water coolant inlet temperature and flow rates, filling ratio and heat flux are evaluated. Simulations with unbalanced heat loads on the two chips being cooled in parallel then show the desirable flow distribution obtained in such gravity-driven systems. Finally, temporal heat load and water coolant flow rate disturbances are simulated and discussed. Noting all of these numerous influences on optimal mini-thermosyphon operation, the need for a accurate and detailed simulation code, benchmarked versus actual system tests, is seen to be imperative for attaining a good, reliable, robust design.","PeriodicalId":426908,"journal":{"name":"2016 15th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"232 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131885310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：