Parametric study of a cold plate for electronic systems cooling

: Experimental and analytical work have been performed to investigate capabilities and thermal performance characteristics of cold plates for electronic equipment cooling. The effort includes air-cooled cold plates, liquid-cooled cold plates, and cold plates provided with heat pipes. Different designs were selected for each of the three categories and thermal tests at different coolant flow and equipment power dissipation rates performed. It has been shown that large amounts of equipment waste heat can be removed by this cooling technique, and thermal performance accurately predicted, particularly with computer-aided analysis.

Metallic TIM Testing and Selection for IC, Power, and RF Semiconductors

Due to a continuous evolving automotive industry, car manufacturers continuously develop new technologies to stay competitive on the market. Most of the modern cars are equipped with electronic devices used for functional or multimedia purposes. Electric cars have additional components used for charging and battery management. All the electronic components used in cars dissipate heat during functioning and all of them need cooling in order to keep a proper functioning temperature. Cold plates are a very efficient cooling solution used in most of the cars. The advantage of the cold plates is that it can dissipate a high amount of heat, have a small volume and can be placed in closed or tight areas. In order to enhance the thermal transfer between the cold plate and the cooling fluid, turbulence enhancement geometries are used inside the cold plate. The purpose of this paper is to use finite element analysis to compare different positions of one protrusion in the cooling channel to enhance the cooling process. The goal is to see if the protrusion position influences the cooling and which placement is more efficient. The second evaluation criteria of the efficiency is to keep a low-pressure drop inside the system.

As energy demands and power electronics density scale concurrently, reliability of such devices is being challenged. Inadequate thermal management can cause system-wide failures due to thermal run-away, thermal expansion induced stresses, interconnect fractures and many more. Conventional techniques used to cool devices consist of heavy, metallic systems such as cold plates and large heat sinks, which can significantly reduce the overall system power density. Moreover, the manufacturing of such components is expensive and often requires custom-made cold plates for improved integration with the electronic system. Although used as a standard practice, these metallic thermal management systems have the potential to intensify electro-magnetic interference (EMI) when coupling with high frequency switching power electronics, and the material density increases the weight of the system, which is detrimental in mobile applications. Lastly, cold plates and heat sinks can create non-uniform cooling profiles in the electronics due to the insufficient management of hot-spots. To combat these drawbacks, a new heat spreader design has been proposed which reduces weight and EMI effects while eliminating hot-spots through localized fluid impingement. This current study describes the methodology and construction of the experimental test setup to characterize the performance of the heat spreading device compared to an off-the-shelf cold plate. Through infrared imagining, the viability of two heated test sections are evaluated in their ability to replicate power module temperature profiles during operation.

Electronics cooling research has been largely focused on high heat flux removal from computer chips in the recent years. However, the equally important field of high-power electronic devices has been experiencing a major paradigm shift from air cooling to liquid cooling over the last decade. For example, multiple 250-W insulated-gate bipolar transistors used in a power drive for a 7000-HP motor used in pumping or in locomotive traction devices would not be sufficiently cooled with air-cooling techniques. Another example is a “hockey puck” SCR of 63 mm diameter used to drive an electric motor that could dissipate over 1500 W and is difficult to cool with air because of the shape of the device. Other devices include radio-frequency generators, industrial battery chargers, printing press thermal and humidity control equipment, traction devices, mining devices, crude oil extraction equipment, magnetic resonance imaging, and railroad engines. This article classifies the cold plates into four types: formed tube cold plate, deep drilled cold plate, machined channel cold plate, and pocketed folded-fin cold plate. The article further discusses selection of cold plate type and channel configuration, and some of the relevant manufacturing issues. It is recommended that the thermal designer be involved in the early stages during the electrical design and layout of the devices.

In this study, a mechanically pump-assisted capillary-driven two-phase cold plate was fabricated to overcome limitations of conventional two-phase cooling technologies in high-heat-flux electronics applications. The cooling mechanism inside the cold plate was based on thin-film evaporation from wicks. The wicks were continuously supplied with liquid refrigerant R245fa using a mechanical pump. Eight heaters simulated the electronics’ heat load and were located on both sides of the cold plate. Eight wicks were positioned within the cold plate, with the thin film evaporation region located directly across from each heater. Each evaporator was a modulated porous wick structure consisting of thick regions responsible for low-pressure drop coolant supply and a thin layer for low thermal resistance evaporation. A phase separator separated the pumped liquid from the vapor space and, in turn, prevented flooding of the thin film evaporation region. The cold plate demonstrated the removal of heat fluxes exceeding 1.18 kW/cm2 over heater footprint areas less than 0.12 cm2 while operating with a low pumping power of 1.04 W. Highly uniform thermal resistances of less than 0.08 K/(W/cm2) were achieved, making the present cold plate attractive for the thermal management of high-heat-flux electronics. The excellent uniformity of thermal resistance indicates independent heat removal from individual heaters, enabling future electronic devices to operate at higher electrical power beyond the state-of-the-art by mounting more heat sources (i.e., chips) on the cold plate.

The performance of a capillary-driven two-phase cold plate (CP) for thermal management of high heat flux electronic devices was investigated. The CP was a key component of a hybrid two-phase cooling system (HTPCS). The HTPCS operated with the refrigerant R245fa and integrated the benefits of a pumped two-phase cooling with a capillary-driven two-phase cooling. The significance of the present study is to fabricate the CP integrated with evaporator wicks through one single additive manufacturing (AM) process. The CP was a compact enclosure including eight heaters located on both sides of the CP. The evaporator wicks were lattice structures, and individual heater was cooled down by its own evaporator. Compared to the CP without evaporator wicks, the capillary-driven CP (CDCP) led to 39% enhancement in the upper limit heat flux, as well as over 44% improvement in the non-uniformity of the thermal resistance of the CP. The equivalent heat fluxes achieved by the CDCP was 203–212 W/cm2 over areas less than 0.1 cm2. The measured thermal resistances at those ranges of heat fluxes were ∼ 0.16–0.30 K-cm2/W, while operating within a low pumping power below 0.3 W.

Water cold plates are widely used for heat dissipation in electronic devices. The design of water cold plats includes fluid characterization, heat transfer analysis, and fluid-solid coupling analysis, which are mostly carried out by fluid or thermal design software simulation, resulting in low simulation efficiency. This paper proposes a water cold plate simulation method based on digital twins, which realizes the downgrading of the traditional 3D CFD simulation model to a 1D mathematical model (ROM) and greatly shortens the thermal simulation calculation time. Meanwhile, real-time mapping is established between the physical form of the cold plate and the digital twin model to visualize the traditional invisible and unmeasurable parameters. In addition, based on the external real-time measurement data, the digital twin ROM model is driven to perform calculations and output the results, thus realizing the thermal performance evaluation of the cold plate under various working conditions, which greatly improves the efficiency of thermal design and thermal simulation of electronic devices.

Thermal and hydraulic performances of seven water-cooled minichannel cold plates with different internal structures are compared using numerical analysis. Recent increasing demands for high-performance computing have led to serious challenges in the thermal management of electronic devices. In addition to dangerous on-chip temperatures, heterogeneous integration and local regions of elevated temperatures (hotspots) lead to nonuniform chip-level temperature distributions. As a result, the lifespan and reliability of electronic devices are adversely impacted. Due to the limitation of the air-cooled heat sinks, several new methods, such as liquid-cooled microchannel cold plates are developed to remedy these challenges. The objective of this work is to provide a comparative numerical study of the effectiveness of different minichannel cold plate internal structures in the thermal management of a chip with a nonuniform power map and a hotspot. Cold plate thermal resistance, on-chip temperature uniformity, and pump power were the metrics used for this comparison. For four coolant inlet flow rates within the laminar regime, it is seen that increasing the inlet flowrate enhances the thermal resistance of all cold plate designs while creating less uniformity in chip-level temperature distribution relative to the conventional straight microchannels. Concentrating pin fins on the hotspot showed a 7.2% reduction in thermal resistance, despite increasing temperature nonuniformity by about 7.6%. However, it is observed that hotspot-focused pin fins are more effective in lowering the chip's maximum temperature. Obtaining lower chip-level nonuniformity may be possible by modifying the inlet and outlet conditions of the cold plates.

Data centers are witnessing an unprecedented increase in processing and data storage, resulting in an exponential increase in the servers’ power density and heat generation. Data center operators are looking for green energy efficient cooling technologies with low power consumption and high thermal performance. Typical air-cooled data centers must maintain safe operating temperatures to accommodate cooling for high power consuming server components such as CPUs and GPUs. Thus, making air-cooling inefficient with regards to heat transfer and energy consumption for applications such as high-performance computing, AI, cryptocurrency, and cloud computing, thereby forcing the data centers to switch to liquid cooling. Additionally, air-cooling has a higher OPEX to account for higher server fan power. Liquid Immersion Cooling (LIC) is an affordable and sustainable cooling technology that addresses many of the challenges that come with air cooling technology. LIC is becoming a viable and reliable cooling technology for many high-power demanding applications, leading to reduced maintenance costs, lower water utilization, and lower power consumption. In terms of environmental effect, single-phase immersion cooling outperforms two-phase immersion cooling. There are two types of single-phase immersion cooling methods namely, forced and natural convection. Here, forced convection has a higher overall heat transfer coefficient which makes it advantageous for cooling high-powered electronic devices. Obviously, with natural convection, it is possible to simplify cooling components including elimination of pump. There is, however, some advantages to forced convection and especially low velocity flow where the pumping power is relatively negligible. This study provides a comparison between a baseline forced convection single phase immersion cooled server run for three different inlet temperatures and four different natural convection configurations that utilize different server powers and cold plates. Since the buoyancy effect of the hot fluid is leveraged to generate a natural flow in natural convection, cold plates are designed to remove heat from the server. For performance comparison, a natural convection model with cold plates is designed where water is the flowing fluid in the cold plate. A high-density server is modeled on the Ansys Icepak, with a total server heat load of 3.76 kW. The server is made up of two CPUs and eight GPUs with each chip having its own thermal design power (TDPs). For both heat transfer conditions, the fluid used in the investigation is EC-110, and it is operated at input temperatures of 30°C, 40°C, and 50°C. The coolant flow rate in forced convection is 5 GPM, whereas the flow rate in natural convection cold plates is varied. CFD simulations are used to reduce chip case temperatures through the utilization of both forced and natural convection. Pressure drop and pumping power of operation are also evaluated on the server for the given intake temperature range, and the best-operating parameters are established. The numerical study shows that forced convection systems can maintain much lower component temperatures in comparison to natural convection systems even when the natural convection systems are modeled with enhanced cooling characteristics.

Miniaturization high heat flux of power electronic devices have posed a colossal challenge for adequate thermal management. Conventional air-cooling solutions are inadequate for high-performance electronics. Liquid cooling is an alternative solution due to the higher specific heat and latent heat associated with the coolants. Liquid-cooled cold plates are typically manufactured by different approaches, such as skived, forged, extrusion, and electrical discharge machining. When researchers are facing challenges in creating complex geometries in small spaces, 3-D-printing can be a solution. In this article, a 3-D-printed cold plate was designed and characterized with water coolant. The printed metal fin structures were strong enough to undergo pressure from the fluid flow even at high flow rates and small fin structures. A copper block with the top surface area of 1 in <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times \,\,1$ </tex-math></inline-formula> in was used to mimic a computer chip. Experimental data have good match with a simulation model, which was built using commercial software 6SigmaET. Effects of geometry parameters and operating parameters were investigated. The fin diameter was varied from 0.3 to 0.5 mm and the fin height was maintained at 2 mm. A special manifold was designed to maximize the surface contact area between coolant and metal surface and therefore minimize thermal resistance. The flow rate was varied from 0.75 to 2 L/min and the coolant inlet temperature was varied from 25 °C to 48 °C. It was observed that for the coolant inlet temperature 25 °C and aluminum cold plate, the junction temperature was kept below 63.2 °C at an input power of 350 W and the pressure drop did not exceed 23 Kpa. Effects of metal materials used in 3-D-printing on the thermal performance of the cold plate were also studied in detail.

A numerical analysis of the effects of water-cooled cold plate designs and the coolant inlet velocity on the thermal management of computer chips with hotspots is presented. With the increasing demand for computational capabilities of highperformance computing, non-uniform temperature distribution across the chip becomes a significant thermal management problem. Localized high temperature regions on the chip surface, known as hotspots, contribute to this non-uniform temperature distribution and may increase the chip temperatures to dangerous levels. Also, the resulting temperature gradient has detrimental impact on various reliability mechanisms of electronic devices. Available research suggests that conventional methods of thermal management, such as air-cooled heat sinks, have reached their optimal limit. Therefore, novel and more efficient thermal management of the computer chips is needed to improve their reliability and performance. Thanks to the advancement of additive manufacturing, it is possible to investigate the heat-removal efficiency of various cold plate designs that are not achievable using conventional manufacturing. In this study, several water-cooled cold plate designs involving different pin fin and mini channel configurations are analyzed for a typical CPU with a hotspot using COMSOL, commercial finite element analysis program. Conjugate heat transfer physics corresponding to heat transfer in solids and laminar flow was used. Various velocities of inlet water are studied as well, while maintaining a laminar flow inside the cold plate. The inlet temperature of water was chosen to be room temperature. Chip temperature uniformity, thermal resistance of the cold plate, maximum and minimum temperatures at the chip level, pressure drop, and pump power were calculated. These response parameters were used to compare the thermal performance of each of the cold plate designs and the inlet velocities. The results from this study suggest that a cold plate with pin fins located on the hotspot performs better than one with only channels in mitigating the hotspot. It is also observed that a quasi-uniform temperature distribution may be achieved if overcooling of the background region is avoided.

Given the characteristics of some missile-borne electronic equipment, such as narrow space, high heat flux, limited power supply, and high reliability in a short time, the idea of phase change heat dissipation is proposed, absorbing a large amount of heat from electronic components through the vaporization phase change of working fluid, to ensure its working performance. The heat dissipation device is a cold plate for micro-channel phase change heat dissipation. Because it is difficult to study phase-change heat dissipation by simulation calculation, the experimental scheme is designed. A set of experimental systems that can not only simulate the working mode of the cold plate in high altitude and low-pressure environments but also carry out multi-parameter exploration is built with conventional and simple devices and instruments. Through extensive and repeated experiments, it has been concluded that: for the same mass of phase change working medium, the smaller the pressure difference between the working medium in the cold plate and the cold plate outlet is, the longer the temperature control time of the cold plate is. When the surface heat load of the cold plate is 400 W, the mass of the phase change working medium is 270 g, and the cold plate temperature control time is up to 10 min. When the working medium of phase transformation is 35% ethanol solution, the cooling and temperature equalization effect of a cold plate are the best, and the maximum temperature of the cold plate surface is stable between 70±1°C for a long time.

Contemporary electronic systems are currently constrained by the high heat fluxes in which they generate at component level. It is evident that heat fluxes are currently approaching the limits of forced air cooling, and that liquid cooling is now under consideration. In this paper five commercially-available and one custom-made cold plates were characterised experimentally. The six cold plates utilized different geometries which included: an array of jets impinging onto a pin matrix; a fin structure; a pin fin structure; a large serpentine channel structure; a slot jet impinging onto wave shaped fins; and the custom cold plate having no significant geometry associated with it, as it was used as a bench mark. The bench mark is anticipated to be the minimum cost solution. The pressure drop, thermal resistance and hydrodynamic power consumption were determined for each solution as a function of flow rate. The results showed that there was a variety of operational power consumption costs coupled with a range of performance levels reached by the six cold plates. This emphasizes the need of a optimum cooling package for a specific application. A relationship of thermal resistance as a function of hydrodynamic power consumed was formulated, thus facilitating the selection of a cold plate for a practical application.

Water cold plates for high power converters: A software tool for easy optimized design