Pub Date : 2014-05-27DOI: 10.1109/ITHERM.2014.6892274
M. Mustafa, J. Roberts, J. Suhling, P. Lall
Solder materials in electronic packages are often subjected to thermal cycling, either during their application in products or during accelerated life qualification testing. Cyclic temperatures cause the solder joints to be subjected to cyclic mechanical stresses and strains due to the mismatches in the thermal expansion coefficients of the assembly materials. Such loads lead to thermomechanical fatigue that results in damage accumulation, crack initiation and propagation, and eventual failure. The authors have extensively examined the effects of isothermal aging on solder stress-strain and creep (constitutive) behavior. However, there have been no prior investigations on aging effects on solder fatigue behavior. Aging leads to both grain and phase coarsening, and can cause recrystallization at Sn grain boundaries. Such changes are closely tied to the damage that occurs during cyclic mechanical loading.
{"title":"Effects of aging on shear cyclic stress strain and fatigue behaviors of lead free solder joints","authors":"M. Mustafa, J. Roberts, J. Suhling, P. Lall","doi":"10.1109/ITHERM.2014.6892274","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892274","url":null,"abstract":"Solder materials in electronic packages are often subjected to thermal cycling, either during their application in products or during accelerated life qualification testing. Cyclic temperatures cause the solder joints to be subjected to cyclic mechanical stresses and strains due to the mismatches in the thermal expansion coefficients of the assembly materials. Such loads lead to thermomechanical fatigue that results in damage accumulation, crack initiation and propagation, and eventual failure. The authors have extensively examined the effects of isothermal aging on solder stress-strain and creep (constitutive) behavior. However, there have been no prior investigations on aging effects on solder fatigue behavior. Aging leads to both grain and phase coarsening, and can cause recrystallization at Sn grain boundaries. Such changes are closely tied to the damage that occurs during cyclic mechanical loading.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"15 1","pages":"142-151"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87185413","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892316
Anton Hassebrook, C. Kruse, Chris Wilson, T. Anderson, C. Zuhlke, D. Alexander, G. Gogos, S. Ndao
In this paper, an experimental investigation of the effects of droplet diameters on the Leidenfrost temperature and its shifts has been carried out. Tests were conducted on a 304 stainless steel polished surface and a stainless steel surface which was processed by a femtosecond laser to form Above Surface Growth (ASG) nano/microstructures. To determine the Leidenfrost temperatures, the droplet lifetime method was employed for both the polished and processed surfaces. A precision dropper was used to vary the size of droplets from 1.5 to 4 millimeters. The Leidenfrost temperature was shown to display shifts as high as 85 °C on the processed surface over the range of droplet sizes, as opposed to a 45 °C shift on the polished surface. The difference between the shifts was attributed to the nature of the force balance between dynamic pressure of droplets and vapor pressure of the insulating vapor layer.
{"title":"Effects of droplet diameter on the Leidenfrost temperature of laser processed multiscale structured surfaces","authors":"Anton Hassebrook, C. Kruse, Chris Wilson, T. Anderson, C. Zuhlke, D. Alexander, G. Gogos, S. Ndao","doi":"10.1109/ITHERM.2014.6892316","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892316","url":null,"abstract":"In this paper, an experimental investigation of the effects of droplet diameters on the Leidenfrost temperature and its shifts has been carried out. Tests were conducted on a 304 stainless steel polished surface and a stainless steel surface which was processed by a femtosecond laser to form Above Surface Growth (ASG) nano/microstructures. To determine the Leidenfrost temperatures, the droplet lifetime method was employed for both the polished and processed surfaces. A precision dropper was used to vary the size of droplets from 1.5 to 4 millimeters. The Leidenfrost temperature was shown to display shifts as high as 85 °C on the processed surface over the range of droplet sizes, as opposed to a 45 °C shift on the polished surface. The difference between the shifts was attributed to the nature of the force balance between dynamic pressure of droplets and vapor pressure of the insulating vapor layer.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"7 1","pages":"452-457"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87216084","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892326
Yashwanth Yadavalli, J. Weibel, S. Garimella
Improved-efficiency heat spreaders must address ergonomic- and performance-driven thermal management demands of electronic devices of increasingly thin form factor. Heat pipes offer a potential high-conductance solution, but performance limitations unique to sub-millimeter thickness devices must be characterized. Using a reduced-order, one-dimensional resistance network and a two-dimensional numerical model, the thermal resistance of a flat heat pipe is benchmarked against a solid heat spreader as a function of geometry and power input. The reduced-order model enables a broad parametric study and analytical formulation of performance limitations, while the higher fidelity numerical approach is used to assess the accuracy of the thermal resistance network near these limits. The form factors and operating conditions for which a heat pipe is more effective than a solid heater spreader are identified. Two of the bounding performance limits have been commonly discussed in prior analyses - a capillary wicking limit and an increase in the heat pipe thermal resistance relative to the solid heat spreader at very large thicknesses. A third vapor-phase threshold is observed when the thickness is reduced below a critical limit. At this threshold, the vapor-phase thermal resistance imposed by the saturation pressure/temperature gradient in the heat pipe causes a crossover in the thermal resistance relative to a solid heat spreader. Devices are susceptible to this performance threshold at very low power inputs that would not otherwise induce a capillary limitation. Accurate prediction of this threshold is an important consideration in the selection and design of ultra-thin heat pipes.
{"title":"Flat heat pipe performance thresholds at ultra-thin form factors","authors":"Yashwanth Yadavalli, J. Weibel, S. Garimella","doi":"10.1109/ITHERM.2014.6892326","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892326","url":null,"abstract":"Improved-efficiency heat spreaders must address ergonomic- and performance-driven thermal management demands of electronic devices of increasingly thin form factor. Heat pipes offer a potential high-conductance solution, but performance limitations unique to sub-millimeter thickness devices must be characterized. Using a reduced-order, one-dimensional resistance network and a two-dimensional numerical model, the thermal resistance of a flat heat pipe is benchmarked against a solid heat spreader as a function of geometry and power input. The reduced-order model enables a broad parametric study and analytical formulation of performance limitations, while the higher fidelity numerical approach is used to assess the accuracy of the thermal resistance network near these limits. The form factors and operating conditions for which a heat pipe is more effective than a solid heater spreader are identified. Two of the bounding performance limits have been commonly discussed in prior analyses - a capillary wicking limit and an increase in the heat pipe thermal resistance relative to the solid heat spreader at very large thicknesses. A third vapor-phase threshold is observed when the thickness is reduced below a critical limit. At this threshold, the vapor-phase thermal resistance imposed by the saturation pressure/temperature gradient in the heat pipe causes a crossover in the thermal resistance relative to a solid heat spreader. Devices are susceptible to this performance threshold at very low power inputs that would not otherwise induce a capillary limitation. Accurate prediction of this threshold is an important consideration in the selection and design of ultra-thin heat pipes.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"62 1","pages":"527-534"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90403472","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892403
M. David, R. Schmidt
Data centers consume a significant amount of energy in the US and worldwide, much of which is consumed by the cooling infrastructure, particularly the chiller plant and computer room air conditioners and air handlers. To enable energy efficient data center designs, ASHRAE added two new IT environmental classes, A3 and A4, with associated allowable inlet air temperatures of 40C and 45C respectively. IT equipment that meet these new allowable environmental envelopes can operate in data centers with minimal refrigeration cooling and instead rely on ambient free cooling. In this paper we investigate the impact of allowing a data center to operate up to the A3 limit of 40C on total data center energy use for 3 different types of servers in a chiller-less data center located in a variety of locations. The study finds that though facility power reduces as the demand for cold air reduces, the increase in IT power consumption, due to fan speed-up, can offset these savings and in some cases result in an overall increase in data center power. Thus the most energy efficient operating point is dependant on the specific energy use profiles of the infrastructure and the IT equipment. The higher allowable temperature can also result in higher failure rates and an increased risk of equipment or service loss due to data center cooling failures. This paper also presents a study on the potential for chiller elimination and chiller use reduction across the US, Europe and in India by operating in the various ASHRAE envelopes. For wet, water side economized data centers, A2 and A3 equipment is sufficient to almost completely remove the need for chillers in many geographic locations.
{"title":"Impact of ASHRAE environmental classes on data centers","authors":"M. David, R. Schmidt","doi":"10.1109/ITHERM.2014.6892403","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892403","url":null,"abstract":"Data centers consume a significant amount of energy in the US and worldwide, much of which is consumed by the cooling infrastructure, particularly the chiller plant and computer room air conditioners and air handlers. To enable energy efficient data center designs, ASHRAE added two new IT environmental classes, A3 and A4, with associated allowable inlet air temperatures of 40C and 45C respectively. IT equipment that meet these new allowable environmental envelopes can operate in data centers with minimal refrigeration cooling and instead rely on ambient free cooling. In this paper we investigate the impact of allowing a data center to operate up to the A3 limit of 40C on total data center energy use for 3 different types of servers in a chiller-less data center located in a variety of locations. The study finds that though facility power reduces as the demand for cold air reduces, the increase in IT power consumption, due to fan speed-up, can offset these savings and in some cases result in an overall increase in data center power. Thus the most energy efficient operating point is dependant on the specific energy use profiles of the infrastructure and the IT equipment. The higher allowable temperature can also result in higher failure rates and an increased risk of equipment or service loss due to data center cooling failures. This paper also presents a study on the potential for chiller elimination and chiller use reduction across the US, Europe and in India by operating in the various ASHRAE envelopes. For wet, water side economized data centers, A2 and A3 equipment is sufficient to almost completely remove the need for chillers in many geographic locations.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"11 1","pages":"1092-1099"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75313844","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892410
D. Squiller, E. Mengotti, P. McCluskey
The demand for power electronic systems to operate in harsh environmental conditions has increased over the past 20 years. These environments include those relating to deep oil-well drilling, automotive and aerospace applications. The miniaturization of the power module along with higher power densities have created elevated stress levels on ancillary subsystems, specifically the control circuitry. This study develops first-order methods and models to assess the solder interconnect reliability of critical components on the control circuitry in power electronic systems. Thermal and reliability simulations based upon Physics-of-Failure modeling techniques were conducted on a 2.2 kW variable-frequency drive to evaluate the susceptibility of component level failure mechanisms. CalcePWA, an interconnect reliability modeling software tool, was used as the primary vehicle to conduct these simulation models. A power cycling apparatus was constructed in order to calibrate the reliability models through accelerated testing of the drive.
{"title":"Assessing solder interconnect reliability of control boards in power electronic systems using Physics-of-Failure models","authors":"D. Squiller, E. Mengotti, P. McCluskey","doi":"10.1109/ITHERM.2014.6892410","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892410","url":null,"abstract":"The demand for power electronic systems to operate in harsh environmental conditions has increased over the past 20 years. These environments include those relating to deep oil-well drilling, automotive and aerospace applications. The miniaturization of the power module along with higher power densities have created elevated stress levels on ancillary subsystems, specifically the control circuitry. This study develops first-order methods and models to assess the solder interconnect reliability of critical components on the control circuitry in power electronic systems. Thermal and reliability simulations based upon Physics-of-Failure modeling techniques were conducted on a 2.2 kW variable-frequency drive to evaluate the susceptibility of component level failure mechanisms. CalcePWA, an interconnect reliability modeling software tool, was used as the primary vehicle to conduct these simulation models. A power cycling apparatus was constructed in order to calibrate the reliability models through accelerated testing of the drive.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"35 1","pages":"1154-1163"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75418126","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892367
Van Gils, M. Speetjens, H. Zwart, H. Nijmeijer
Topic is feedback stabilisation of a nonlinear pool-boiling system in three spatial dimensions (3D). Regulation of its unstable (non-uniform) equilibria has great potential for application in micro-electronics cooling and thermal-management systems. Here, as a first step, stabilisation of such 3D equilibria is considered. A control law is designed that regulates the heat supply to the heater as a function of the Fourier-Chebyshev modes of its internal temperature distribution. This admits a controller that is tailored to the system dynamics, as these modes intimately relate to the physical eigenmodes. The internal temperature distribution is, similar to practical situations, estimated from a finite number of measurement positions on the heater surface by an observer. Performance of this output-based modal controller is demonstrated and analysed by simulations of the nonlinear closed-loop system. This provides first proof of principle of the proposed control strategy for the regulation of 3D boiling states.
{"title":"Modal control of unstable boiling states in three-dimensional nonlinear pool-boiling","authors":"Van Gils, M. Speetjens, H. Zwart, H. Nijmeijer","doi":"10.1109/ITHERM.2014.6892367","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892367","url":null,"abstract":"Topic is feedback stabilisation of a nonlinear pool-boiling system in three spatial dimensions (3D). Regulation of its unstable (non-uniform) equilibria has great potential for application in micro-electronics cooling and thermal-management systems. Here, as a first step, stabilisation of such 3D equilibria is considered. A control law is designed that regulates the heat supply to the heater as a function of the Fourier-Chebyshev modes of its internal temperature distribution. This admits a controller that is tailored to the system dynamics, as these modes intimately relate to the physical eigenmodes. The internal temperature distribution is, similar to practical situations, estimated from a finite number of measurement positions on the heater surface by an observer. Performance of this output-based modal controller is demonstrated and analysed by simulations of the nonlinear closed-loop system. This provides first proof of principle of the proposed control strategy for the regulation of 3D boiling states.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"255 1","pages":"833-839"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78986159","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892331
Xiaojin Wei, G. Goth, P. Kelly, R. Zoodsma, A. VanDeventer
Air-water hybrid cooling offers flexible design choices for computer systems with components of different thermal management needs. On one hand, water cooling enables the continuous growth of CPU performance and increasing packaging density. High performance cold plates such as microchannels have been successfully implemented for water cooling in previous high-end systems. When coupled with an air-water heat exchanger or radiator, the water loop becomes a closed one with no need for facility chilled water. This significantly reduces the complexity to deploy the server in the data center. On the other hand, for components with less thermal demand, traditional air-cooling technology is adequate with low cost, high availability and better serviceability. For the computer system as a whole, an air-water hybrid cooling system may be optimized. Such a hybrid system typically requires pumps to drive the water loops, air-movers to drive air through the radiator and blowers or fans to drive the air flow for component cooling. It is the focus of this paper to study the optimum allocation of energy between the pumps and air-movers for a given total cooling energy budget and overall load. The goals are to achieve better overall thermal performance and to reduce the cooling energy consumption. To this end models for each cooling block are established based on test data. These include the air-water heat exchanger, pumps, blowers, and cold plates. These models are linked together to predict the overall thermal system operating points for different application scenarios. A parametric study is then conducted to define the near optimum allocation of cooling energy for these scenarios that meets the thermal design objectives. Additionally, sub-threshold leakage for the CPU is taken into account to enhance the model since temperature provides positive feedback. It is shown through modeling that additional performance enhancement is possible with judicious allocation of cooling energy for a given overall energy budget. It is argued in this paper that overall energy efficiency can be improved significantly through intelligent data driven energy allocation.
{"title":"Air-water hybrid cooling for computer servers: A case study for optimum cooling energy allocation","authors":"Xiaojin Wei, G. Goth, P. Kelly, R. Zoodsma, A. VanDeventer","doi":"10.1109/ITHERM.2014.6892331","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892331","url":null,"abstract":"Air-water hybrid cooling offers flexible design choices for computer systems with components of different thermal management needs. On one hand, water cooling enables the continuous growth of CPU performance and increasing packaging density. High performance cold plates such as microchannels have been successfully implemented for water cooling in previous high-end systems. When coupled with an air-water heat exchanger or radiator, the water loop becomes a closed one with no need for facility chilled water. This significantly reduces the complexity to deploy the server in the data center. On the other hand, for components with less thermal demand, traditional air-cooling technology is adequate with low cost, high availability and better serviceability. For the computer system as a whole, an air-water hybrid cooling system may be optimized. Such a hybrid system typically requires pumps to drive the water loops, air-movers to drive air through the radiator and blowers or fans to drive the air flow for component cooling. It is the focus of this paper to study the optimum allocation of energy between the pumps and air-movers for a given total cooling energy budget and overall load. The goals are to achieve better overall thermal performance and to reduce the cooling energy consumption. To this end models for each cooling block are established based on test data. These include the air-water heat exchanger, pumps, blowers, and cold plates. These models are linked together to predict the overall thermal system operating points for different application scenarios. A parametric study is then conducted to define the near optimum allocation of cooling energy for these scenarios that meets the thermal design objectives. Additionally, sub-threshold leakage for the CPU is taken into account to enhance the model since temperature provides positive feedback. It is shown through modeling that additional performance enhancement is possible with judicious allocation of cooling energy for a given overall energy budget. It is argued in this paper that overall energy efficiency can be improved significantly through intelligent data driven energy allocation.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"33 1","pages":"568-573"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81533493","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892371
P. Coman, C. Veje
This paper presents a dynamic model for simulating the heat dissipation and the impact of Phase Change Materials (PCMs) on the peak temperature in Lithium-ion batteries during discharging operation of a hybrid truck under different ambient temperatures.
{"title":"Modeling temperature development of Li-Ion battery packs in hybrid refuse truck operating at different ambient conditions","authors":"P. Coman, C. Veje","doi":"10.1109/ITHERM.2014.6892371","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892371","url":null,"abstract":"This paper presents a dynamic model for simulating the heat dissipation and the impact of Phase Change Materials (PCMs) on the peak temperature in Lithium-ion batteries during discharging operation of a hybrid truck under different ambient temperatures.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"18 1","pages":"862-869"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81838652","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892332
Bharath Nagendran, S. Nagaraj, J. Fernandes, R. Eiland, D. Agonafer, V. Mulay
As a common practice in the data center industry, chassis fans are used to direct air flow independent from neighboring servers. In general, smaller fans are less efficient compared to geometrically similar larger fans. In this study, a novel approach is proposed whereby chassis enclosed fans are replaced with a smaller number of larger fans installed behind a stacked array of servers which share airflow. As a baseline study, a CPU dominated 1.5U Open Compute server with four 60mm fans installed within its chassis is characterized experimentally for its flow impedance, fan speed dependent flow rate, effect on die temperature and power consumption at various compute utilization levels. Larger fans with a square frame size of 80mm and 120mm are selected and individually characterized for their air moving capacity and power consumption. Primary emphasis is placed on the 80mm fan case, with discussion of the 120mm fans included. CFD is used to simulate a system of stacked servers serviced by larger fans to obtain its flow characteristics and operating points. The fan power consumption of the larger fans is determined experimentally at these operating points replicated in an air flow bench. Comparing with the base line experiments, replacing smaller fans with larger units results in a significant decrease in fan power consumption without conceding flow rate and static pressure requirements.
作为数据中心行业的一种常见做法，机箱风扇用于引导空气流动，而不依赖于相邻的服务器。一般来说，与几何形状相似的大型风扇相比，较小的风扇效率较低。在本研究中，提出了一种新颖的方法，即将机箱封闭的风扇替换为安装在堆叠服务器阵列后面的数量较少的较大风扇，从而共享气流。作为基线研究，以CPU为主的1.5U Open Compute服务器为例，在机箱内安装4个60mm风扇，实验表征了不同计算利用率下的流阻抗、风扇转速相关的流量、对芯片温度和功耗的影响。选择方形框架尺寸为80mm和120mm的大型风扇，并根据其空气移动能力和功耗进行单独表征。主要重点放在80mm风扇机箱上，包括对120mm风扇的讨论。利用CFD模拟了一个由大型风扇服务的堆叠服务器系统，以获得其流动特性和工作点。在空气流动实验台上复制这些工作点，实验确定了较大风扇的风扇功耗。与基线实验相比，在不降低流量和静压要求的情况下，用大风机代替小风机可以显著降低风机功耗。
{"title":"Improving cooling efficiency of servers by replacing smaller chassis enclosed fans with larger rack-mount fans","authors":"Bharath Nagendran, S. Nagaraj, J. Fernandes, R. Eiland, D. Agonafer, V. Mulay","doi":"10.1109/ITHERM.2014.6892332","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892332","url":null,"abstract":"As a common practice in the data center industry, chassis fans are used to direct air flow independent from neighboring servers. In general, smaller fans are less efficient compared to geometrically similar larger fans. In this study, a novel approach is proposed whereby chassis enclosed fans are replaced with a smaller number of larger fans installed behind a stacked array of servers which share airflow. As a baseline study, a CPU dominated 1.5U Open Compute server with four 60mm fans installed within its chassis is characterized experimentally for its flow impedance, fan speed dependent flow rate, effect on die temperature and power consumption at various compute utilization levels. Larger fans with a square frame size of 80mm and 120mm are selected and individually characterized for their air moving capacity and power consumption. Primary emphasis is placed on the 80mm fan case, with discussion of the 120mm fans included. CFD is used to simulate a system of stacked servers serviced by larger fans to obtain its flow characteristics and operating points. The fan power consumption of the larger fans is determined experimentally at these operating points replicated in an air flow bench. Comparing with the base line experiments, replacing smaller fans with larger units results in a significant decrease in fan power consumption without conceding flow rate and static pressure requirements.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"21 1","pages":"574-582"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82868650","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 : 2014-05-27DOI: 10.1109/ITHERM.2014.6892263
K. Geisler
Solid state light sources, such as light emitting diodes (LEDs), provide many inherent benefits and will dominate the lighting industry in the coming decades. While much of the industry is currently focused on packaging LED technology in standardized 19th century form factors to address the massive installed base of traditional fixtures, the unique characteristics of solid state devices can and will be exploited to produce new luminaire types and new paradigms in lighting design for the 21st century and beyond. Since operating temperatures directly impact energy efficiency, output spectrum, and product lifetime, thermal management is a key linkage in the interdependence of application requirements, design trade-offs, and performance characteristics. This paper details the design of a high-power, high-density LED-based light source for large-scale lighting applications. In particular, a low-profile folded-fin copper heat sink was designed for forced-convection cooling by an array of 38 mm × 38 mm fans. Heat sink dimensions, including fin thickness, fin spacing, heat sink length, and heat sink base thickness to fin height ratio, were varied within form factor constraints and manufacturing limits to produce a suitable thermal solution for a 60,000+ lumen, 50.8 mm × 50.8 mm LED array dissipating 600 W of heat. Results of numerical, analytical, and experimental investigations demonstrate that LED junction temperatures can be maintained below maximum operating limits in a 45°C ambient.
固态光源，如发光二极管(led)，提供了许多固有的好处，并将在未来几十年主导照明行业。虽然大多数行业目前都专注于将LED技术封装在标准化的19世纪形状因素中，以解决传统灯具的大量安装基础，但固态器件的独特特性可以并且将被利用来生产21世纪及以后的新型灯具类型和照明设计的新范例。由于工作温度直接影响能源效率、输出频谱和产品寿命，因此热管理是应用需求、设计权衡和性能特征相互依赖的关键环节。本文详细介绍了用于大规模照明应用的大功率高密度led光源的设计。特别地，设计了一个低轮廓的折叠翅片铜散热器，用于38 mm × 38 mm风扇阵列的强制对流冷却。散热片的尺寸，包括散热片厚度、散热片间距、散热片长度和散热片基座厚度与散热片高度的比例，在外形因素限制和制造限制的范围内进行了调整，以生产出适合60000流明、50.8 mm × 50.8 mm LED阵列的散热方案，散热功率为600w。数值、分析和实验研究的结果表明，LED结温可以在45°C的环境中保持在最大工作极限以下。
{"title":"Parametric design of a low-profile, forced convection heat sink for high-power, high-density LED arrays","authors":"K. Geisler","doi":"10.1109/ITHERM.2014.6892263","DOIUrl":"https://doi.org/10.1109/ITHERM.2014.6892263","url":null,"abstract":"Solid state light sources, such as light emitting diodes (LEDs), provide many inherent benefits and will dominate the lighting industry in the coming decades. While much of the industry is currently focused on packaging LED technology in standardized 19th century form factors to address the massive installed base of traditional fixtures, the unique characteristics of solid state devices can and will be exploited to produce new luminaire types and new paradigms in lighting design for the 21st century and beyond. Since operating temperatures directly impact energy efficiency, output spectrum, and product lifetime, thermal management is a key linkage in the interdependence of application requirements, design trade-offs, and performance characteristics. This paper details the design of a high-power, high-density LED-based light source for large-scale lighting applications. In particular, a low-profile folded-fin copper heat sink was designed for forced-convection cooling by an array of 38 mm × 38 mm fans. Heat sink dimensions, including fin thickness, fin spacing, heat sink length, and heat sink base thickness to fin height ratio, were varied within form factor constraints and manufacturing limits to produce a suitable thermal solution for a 60,000+ lumen, 50.8 mm × 50.8 mm LED array dissipating 600 W of heat. Results of numerical, analytical, and experimental investigations demonstrate that LED junction temperatures can be maintained below maximum operating limits in a 45°C ambient.","PeriodicalId":12453,"journal":{"name":"Fourteenth Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)","volume":"7 1","pages":"47-58"},"PeriodicalIF":0.0,"publicationDate":"2014-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87047368","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：