Search for: All records

The ability of traditional room-conditioning systems to accommodate expanding information technology loads is limited in contemporary data centers (DCs), where the storage, storing, and processing of data have grown quickly as a result of evolving technological trends and rising demand for online services, which has led to an increase in the amount of waste heat generated by IT equipment. Through the implementation of hybrid air and liquid cooling technologies, targeted, on-demand cooling is made possible by employing a variety of techniques, which include but are not limited to in-row, overhead, and rear door heat exchanger (HX) cooling systems. One of the most common liquid cooling techniques will be examined in this study based on different conditions for high-power density racks (+50 kW). This paper investigates the cooling performance of a liquid-to-air in-row coolant distribution unit (CDU) in test racks containing seven thermal test vehicles (TTVs) under various conditions, focusing on cooling capacity and HX effectiveness under different supply air temperatures (SAT). This test rig has the necessary instruments to monitor and analyze the experiments on both the liquid coolant and the air sides. Moreover, another experiment is conducted to assess the performance of the CDU that runs under different control fan schemes, as well as how the change of the control type will affect the supply fluid temperature and the TTV case temperatures at 10%, 50%, and 100% of the total power. Finally, suggestions for the best control fan scheme to use for these systems and units are provided at the conclusion of the study. 

Abstract Demand is growing for the dense and high-performing IT computing capacity to support artificial intelligence, deep learning, machine learning, autonomous cars, the Internet of Things, etc. This led to an unprecedented growth in transistor density for high-end CPUs and GPUs, creating thermal design power (TDP) of even more than 700 watts for some of the NVIDIA existing GPUs. Cooling these high TDP chips with air cooling comes with a cost of the higher form factor of servers and noise produced by server fans close to the permissible limit. Direct-to-chip cold plate-based liquid cooling is highly efficient and becoming more reliable as the advancement in technology is taking place. Several components are used in the liquid-cooled data centers for the deployment of cold plate-based direct-to-chip liquid cooling like cooling loops, rack manifolds, CDUs, row manifolds, quick disconnects, flow control valves, etc. Row manifolds used in liquid cooling are used to distribute secondary coolant to the rack manifolds. Characterizing these row manifolds to understand the pressure drops and flow distribution for better data center design and energy efficiency is important. In this paper, the methodology is developed to characterize the row manifolds. Water-based coolant Propylene glycol 25% was used as the coolant for the experiments and experiments were conducted at 21 °C coolant supply temperature. Two, six-port row manifolds' P-Q curves were generated, and the value of supply pressure and the flowrate were measured at each port. The results obtained from the experiments were validated by a technique called flow network modeling (FNM). FNM technique uses the overall flow and thermal characteristics to represent the behavior of individual components. 

Owing to the dramatic increase in IT power density and energy consumption, the data center (DC) sector has started adopting thermally- and energy-efficient liquid cooling methods. This study examines a single-phase direct-to-chip liquid cooling approach for three high-heat-density racks, utilizing two liquid-to-air (L2A) cooled coolant distribution units (CDUs) and a combined total heat load of 128 kW. An experimental setup was developed to test different types of CDUs, cooling loops, and thermal testing vehicles (TTVs) for different operating conditions. IR images and the collected data were used to investigate the effect of air recirculation between cold and hot aisle containments on the CDU’s performance and stability of supply air temperature (SAT). Three different types of cooling loops (X, Y, and Z) were characterized thermally and hydraulically. Results show that Type Y has the lowest cold plate thermal resistance and pressure drop, among others. In a later test that included a single rack at a heat load of 53 kW and a single CDU, the heat capture ratio for fluid was found to be 94%. Experiments show that using blanking panels on the back of the racks limits hot air recirculation and maintains a steady SAT in the cold aisle. Finally, the CDU performance was evaluated at a high heat load for the three racks at 128 kW, and the average cooling capacity of the units is 58.6 kW, and the effectiveness values for CDU 1 and CDU 2 are 0.83 and 0.82, respectively. 

Abstract Due to the increasing computational demand driven by artificial intelligence, machine learning, and the Internet of Things (IoT), there has been an unprecedented growth in transistor density for high-end CPUs and GPUs. This growth has resulted in high thermal dissipation power (TDP) and high heat flux, necessitating the adoption of advanced cooling technologies to minimize thermal resistance and optimize cooling efficiency. Among these technologies, direct-to-chip cold plate-based liquid cooling has emerged as a preferred choice in electronics cooling due to its efficiency and cost-effectiveness. In this context, different types of single-phase liquid coolants, such as propylene glycol (PG), ethylene glycol (EG), DI water, treated water, and nanofluids, have been utilized in the market. These coolants, manufactured by different companies, incorporate various inhibitors and chemicals to enhance long-term performance, prevent biogrowth, and provide corrosion resistance. However, the additives used in these coolants can impact their thermal performance, even when the base coolant is the same. This paper aims to compare these coolant types and evaluate the performance of the same coolant from different vendors. The selection of coolants in this study is based on their performance, compatibility with wetted materials, reliability during extended operation, and environmental impact, following the guidelines set by ASHRAE. To conduct the experiments, a single cold plate-based benchtop setup was constructed, utilizing a thermal test vehicle (TTV), pump, reservoir, flow sensor, pressure sensors, thermocouple, data acquisition units, and heat exchanger. Each coolant was tested using a dedicated cold plate, and thorough cleaning procedures were carried out before each experiment. The experiments were conducted under consistent boundary conditions, with a TTV power of 1000 watts and varying coolant flow rates (ranging from 0.5 lpm to 2 lpm) and supply coolant temperatures (17°C, 25°C, 35°C, and 45°C), simulating warm water cooling. The thermal resistance (Rth) versus flow rate and pressure drop (ΔP) versus flow rate graphs were obtained for each coolant, and the impact of different supply coolant temperatures on pressure drop was characterized. The data collected from this study will be utilized to calculate the Total Cost of Ownership (TCO) in future research, providing insights into the impact of coolant selection at the data center level. There is limited research available on the reliability used in direct-to-chip liquid cooling, and there is currently no standardized methodology for testing their reliability. This study aims to fill this gap by focusing on the reliability of coolants, specifically propylene glycols at concentrations of 25%. To analyze the effectiveness of corrosion inhibitors in these coolants, ASTM standard D1384 apparatus, typically used for testing engine coolant corrosion inhibitors on metal samples in controlled laboratory settings, was employed. The setup involved immersing samples of wetted materials (copper, solder coated brass, brass, steel, cast iron, and cast aluminum) in separate jars containing inhibited propylene glycol solutions from different vendors. This test will determine the reliability difference between the same inhibited solutions from different vendors. 

More than ever before, data centers must deploy robust thermal solutions to adequately host the high-density and high-performance computing that is in high demand. The newer generation of central processing units (CPUs) and graphics processing units (GPUs) has substantially higher thermal power densities than previous generations. In recent years, more data centers rely on liquid cooling for the high-heat processors inside the servers and air cooling for the remaining low-heat information technology equipment. This hybrid cooling approach creates a smaller and more efficient data center. The deployment of direct-to-chip cold plate liquid cooling is one of the mainstream approaches to providing concentrated cooling to targeted processors. In this study, a processor-level experimental setup was developed to evaluate the cooling performance of a novel computer numerical control (CNC) machined nickel-plated impinging cold plate on a 1 in.  1 in. mock heater that represents a functional processing unit. The pressure drop and thermal resistance performance curves of the electroless nickel-plated cold plate are compared to those of a pure copper cold plate. A temperature uniformity analysis is done using compuational fluid dynamics and compared to the actual test data. Finally, the CNC machined pure copper one is compared to other reported cold plates to demonstrate its superiority of the design with respect to the cooling performance. 

Abstract The practice of commissioning data centers (DCs) is necessary to confirm the compliance of the cooling system to the information technology equipment (ITE) load (design capacity). In a typical DC, there are different types of ITE, each having its physical characteristics. Considering these geometrical and internal differences among ITE, it is infeasible to use the actual ITE as a self-simulator. Hence, a separate device called load bank is employed for that purpose. Load banks create a dummy thermal load to analyze, test, and stress the cooling infrastructure. Available commercial load banks do not accurately replicate a server's airflow patterns and transient heat signatures which are governed by thermal inertia, energy dissipation, flow resistance, and fan system behavior. In this study, a novel prototype of the server called server simulator was designed and built with different components to be used as a server mockup. The server simulator accurately captured air resistance, heat dissipation, and the functionality of actual server behavior. Experimental data showed up to 93% improvement in ITE passive and active flow curves using the designed server simulator compared to the commercial load bank. Furthermore, the experimental results demonstrated a below 5% discrepancy on the critical back pressure and free delivery point between the actual ITE and the designed server simulator. In addition, experimental data indicated that the developed server simulator improved the actual ITE thermal mass by 27% compared to the commercial load bank. 

Miniaturization and 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 thanks 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, electrical discharge machining. When researchers are facing challenges at creating complex geometries in small spaces, 3D-printing can be a solution. In this paper, a 3D-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 top surface area of 1 inch by 1 inch was used to mimic a computer chip. Experimental data has good match with a simulation model which was built using commercial software 6SigmaET. Effects of geometry parameters and operating parameters were investigated. Fin diameter was varied from 0.3 mm to 0.5 mm and 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 L/min to 2 L/min and coolant inlet temperature was varied from 25 to 48 oC. It was observed that for the coolant inlet temperature 25 oC and aluminum cold plate, the junction temperature was kept below 63.2 oC at input power 350 W and pressure drop did not exceed 23 Kpa. Effects of metal materials used in 3D-printing on the thermal performance of the cold plate were also studied in detail. 

This paper focuses on two-phase flow boiling of dielectric coolant HFE 7000 inside a copper multi-microchannel heat sink for high heat flux chip applications. The heat sink is composed of parallel microchannels, 200 μm wide, 2500 μm high, and 20 mm long, with 200-μm-thick fins separating the channels. The copper heat sink consists of almost 100 channels connected by a longitude groove with a nearly trapezoidal cross section. Coolant impinges down to the base at the groove and then goes along the microchannels. A copper block heater arrangement was used to mimic a computer chip with a footprint of 1”x1” (6.45 cm2). The base heat flux was varied from 7.75 W/cm2 to 96.1 W/cm2 and the mass flux from 547.6 to 958.4 kg/m2s, at a nominal saturation temperature of 54 °C. Heat transfer coefficients as high as 57.5 kW/m2K were reached, keeping the base temperature under 66 °C with a maximum of 21.9 kPa of pressure drop, for inlet subcooling of 5 degree and a coolant flow rate of 958.4 kg/m2. Effects of inner diameter of tubing on thermal performance and pressure drop are also discussed. It was observed that an increase of tubing inner diameter by 60 % can result in increase of heat transfer coefficient by 47.8 % and reduction in pressure drop by 63 %.