WO2007139558A1 - Warm cooling for electronics - Google Patents

Abstract

A method of providing cooled air to electronic equipment includes capturing heated air from a volume containing electronic equipment, cooling the heated air by more than fifteen degrees Celsius in an air-to-water heat exchanger, and supplying cooling water to the air-to-water heat exchanger at a temperature above a dew point temperature of the heated air.

Description

Warm Cooling for Electronics

TECHNICAL FIELD

This document relates to systems and methods for providing cooling for areas containing electronic equipment, such as computer server rooms and server racks in computer data centers.

BACKGROUND

Computer users often focus on the speed of computer microprocessors (e.g., megahertz and gigahertz). Many forget that this speed often comes with a cost — higher electrical consumption. For one or two home PCs, this extra power may be negligible when compared to the cost of running the many other electrical appliances in a home. But in data center applications, where thousands of microprocessors maybe operated, electrical power requirements can be very important. Power consumption is also, in effect, a double whammy. Not only must a data center operator pay for electricity to operator its many computers, but the operator must also pay to cool the computers. That is because, by simple laws of physics, all the power has to go somewhere, and that somewhere is, for the most part, conversion into heat. A pair of microprocessors mounted on a single motherboard can draw 200-400 watts or more of power. Multiply that figure by several thousand (or tens of thousands) to account for the many computers in a large data center, and one can readily appreciate the amount of heat that can be generated. It is much like having a room filled with thousands of burning floodlights. Thus, the cost of removing all of the heat can also be a major cost of operating large data centers. That cost typically involves the use of even more energy, in the form of electricity and natural gas, to operate chillers, condensers, pumps, fans, cooling towers, and other related components. Heat removal can also be important because, although microprocessors may not be as sensitive to heat as are people, increases in heat generally can cause great increases in microprocessor errors. In sum, such a system may require electricity to heat the chips, and more electricity to cool the chips.

SUMMARY

This document describes systems and methods that may be employed to remove heat efficiently from areas storing electronic equipment, such as data centers. In certain implementations, cooling of equipment may occur at elevated air temperatures. For example, high temperature rises may be created across a group of heat- generating components by intentionally slowing the flow of cooling air over the components. For example, temperature rises across the components of twenty degree Celsius or more may be maintained, with entering air temperatures of about twenty-five degrees Celsius and exiting temperatures of about forty- five degrees Celsius. The temperatures of any associated cooling water may also be elevated, such as to produce supply temperatures of about twenty degrees Celsius and return temperatures of about forty degrees Celsius. Because of the elevated temperatures, the system may be run under most conditions using only cooling from cooling towers or other free cooling sources, without the need for a chiller or other similar sources that require power for cooling.

In one implementation, a method of providing cooled air to electronic equipment is disclosed. The method comprises capturing heated air from a volume containing electronic equipment, cooling the heated air by more than fifteen degrees Celsius in an air-to-water heat exchanger, and supplying cooling water to the air-to-water heat exchanger at a temperature above a dew point temperature of the heated air. The step of capturing the heated air may comprise capturing substantially only the heated air, and capturing substantially no air that is not from the volume containing the electronic equipment. The cooling water may be returned from the air-to-water heat exchanger at a temperature above 40 degrees Celsius.

In some aspects, the method may further comprise returning water from the air-to-water heat exchanger, wherein the temperature of the supplied water is more than 15 degrees Celsius cooler than the temperature of the returned water. The method may also comprise returning water from the air-to-water-heat exchanger and transferring heat from the returned water to water supplied from a free cooling water source. The free cooling water source may comprise one or more cooling towers. Also, the temperature of the returned water may be more than 10 degrees Celsius warmer than the water supplied from the free cooling water source. In addition, the water supplied from the free cooling water source may define a first temperature delta from the transferring of heat, and the water returned from the air-to-water heat exchanger may define a second temperature delta, wherein the first temperature delta is more than 30 percent less than the second temperature delta.

In another implementation, a method of providing cooled air to electronic equipment is disclosed, and comprises heating air from a workspace by a temperature differential of more than 20 degrees Celsius by passing it over electronic equipment, cooling the heated air by approximately the same temperature as the temperature differential, and returning the air to the workspace. The method may also comprise cooling the air using free cooling. Also, the free cooling may comprise transferring heat from the air to cooling water, and transferring heat from the cooling water to condenser water (i.e., cooling tower water in this application) that undergoes open evaporation.. In some aspects, heating the air from the workspace may comprise heating the air from about 25 degrees Celsius to more than 45 degrees Celsius. The method may further comprise capturing the heated air and controlling the temperature of the captured air by controlling the rate of airflow over the electronic equipment. The controlled temperature may be computed as a factor of a failure temperature for the electronic equipment. The air may be returned directly to the workspace after it is cooled, and the electronic equipment may comprise a large plurality of rack servers in a data center.

In yet another implementation, a system for providing cooled air to electronic equipment is disclosed. The system may include a free cooling water source, a water-to-water heat exchanger in fluid communication with the cooling tower, and an air-to-water heat exchanger in fluid communication with the water-to-water heat exchanger and positioned to received heated air from a group of electronic devices, wherein the water-to-water heat exchanger is configured to supply cooling water to the water-to-air heat exchanger at a temperature above a dew point of air surrounding the water-to-air heat exchange. The free cooling water source may comprise a cooling tower, and the water-to- water heat exchanger may be configured to supply cooling water at 15-35 degrees Celsius.

In yet another implementation, a system for providing cooled air to electronic equipment is disclosed. The system comprises a cooling tower, a water-to-water heat exchanger connected by a first pair of pipes to the cooling tower, and an air-to- water heat exchanger connected by a second pair of pipes to the water-to-water heat exchanger and positioned to received heated air from a group of electronic devices, wherein the first pair of pipes and the second pair of pipes are substantially lacking in insulation or condensation collecting apparatuses.

The details of one or more embodiments of the invention are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the invention will be apparent from the description and drawings, and from the claims.

DESCRIPTION OF DRAWINGS

FIG. 1 is a schematic diagram showing a system for cooling a computer data center.

FIG. 2 is a psychometric chart showing a heating and cooling cycle for air in a data center.

FIG. 3 is a plan view in schematic form of a hydronic cooling system for a data center.

FIG. 4 is a plan view in schematic form of a hydronic cooling system for a data center housed in shipping containers. FIG. 5 shows container-based data centers mounted on a docked container ship and barge.

FIG. 6 is a flowchart showing steps for cooling a data center using elevated temperatures.

Like reference symbols in the various drawings indicate like elements.

DETAILED DESCRIPTION

FIG. 1 is a schematic diagram showing a system 100 for cooling a computer data center 101. The system 100 generally includes an air handling unit (including e.g., fan 110 and cooling coils 112a, 112b) in the data center 101 for transferring heat from the data center's air to cooling water, a heat exchanger 122 for removing heat from the cooling water and passing it to cooling tower water, and a cooling tower 118 to pass the accumulated heat to the ambient air through evaporation and cooling of the cooling tower water. In general operation, the system 100 may be run from the cooling tower/heat exchanger/cooling coil system, though a powered cooling system such as a chiller may be provided for peak loads, such as when the ambient dew point is very high and the cooling tower cannot provide sufficient cooling alone. The temperatures of each portion of the system are selected to be relatively high, so as to permit more efficient operation of the system 100. m particular, the rate of heat transfer between two fluids, such as between the data center 101 air and the cooling water, is directly proportional to the difference in temperature between the fluids. This difference in temperature may be increased substantially by controlling certain parameters of the system. When the difference in temperature is increased substantially, the rate of heat transfer can increase substantially, and energy efficiency of the system 100 may likewise increase substantially.

When the difference in temperatures is not very large, a change of only a few degrees can bring substantial gains in efficiency. For example, by heating are to 45 degrees Celsius rather than 40 degree Celsius, where the cooling water enters at 20 degrees Celsius, the temperature difference is increased from 20 degrees to 25 degrees — which may result in an increase in heat flow of 25 percent. The actual difference will vary slightly, as the entering conditions for air and water are not the only conditions (because the air cools as it passes through a cooling cool, and the water warms); this example, however, indicates how the difference in temperature can affect efficiency of a system.

Use of elevated temperatures in a system may also prevent air in or around the system from falling below its liquid saturation point, i.e., its dew point, and condensing. This may, in certain circumstances, provide benefits both in efficiency and in operations of the system. Efficiency benefits may be obtained because creating condensation requires much more energy than simply cooling air, so that systems creating condensation may use a large amount of electricity or other energy. Improvements in operations of the system may occur because, if pipes in the system carry water that is below the saturation temperature of the air around the pipes, condensation might form on the pipes. That condensation can damage the pipes, and can require the installation of insulation on the pipes (to stop the condensation). hi the system shown in FIG. 1, use of elevated temperatures may substantially reduce, or almost entirely eliminate, the need for energy-intensive cooling components such as chillers and the like, even where the heat load in the data center 101 is very high. As a result, system 100 may be operated at a lower operating cost than could otherwise be achieved. In addition, lower capital costs may be required, because fans, coils, heat exchangers, and cooling towers are relatively basic and inexpensive components. hi addition, those components are often very standardized, so that their acquisition costs are lower, and they are more easily located, particularly in developing countries and remote areas where it may be beneficial to place a data center 101. Use of system 100 in remote areas and other areas with limited access to electrical power is also helped by the fact that system 100 may be operated using less electrical power. As a result, such a system can be located near lower-power electrical sub-stations and the like. As discussed more completely below, lower-powered systems may also be amenable to being implemented as self-powered systems using energy sources such as solar, wind, natural-gas powered turbines, fuel cells, and the like.

Referring now to FIG. 1, there is shown a data center 101, which as shown is a building that houses a large number of computers or similar heat- generating electronic components. A workspace 106 is defined around the computers, which are arranged in a number of parallel rows and mounted in vertical racks, such as racks 102a, 102b. The racks may include pairs of vertical rails to which are attached paired mounting brackets. Trays containing computers, such as standard circuit boards in the form of motherboards, may be placed on the mounting brackets.

In one example, the mounting brackets may be angled rails welded or otherwise adhered to vertical rails in the frame of a rack, and trays may include motherboards that are slid into place on top of the brackets, similar to the manner in which foods trays are slid onto storage racks in a cafeteria. The trays may be spaced closely together to maximize the number of trays in a data center, but sufficiently far apart to contain all the components on the trays and to permit air circulation between the trays. Other arrangements may also be used. For example, trays may be mounted vertically in groups, such as in the form of computer blades. The trays may simply rest in a rack and be electrically connected after they are slid into place, or they may be provide with mechanisms, such as electrical traces along one edge, that create electrical and data connections when they are slid into place.

Air may circulate from workspace 106 across the trays and into warm-air plenums 104a, 104b behind the trays. The air may be drawn into the trays by fans mounted at the back of the trays. The fans may be programmed or otherwise configured to maintain a set exhaust temperature for the air into the warm air plenum, and may also be programmed or otherwise configured to maintain a particular temperature rise across the trays. Where the temperature of the air in the work space 106 is known, controlling the exhaust temperature also indirectly controls the temperature rise.

The temperature rise can be large. For example, the work space 106 temperature may be about 25 degrees Celsius and the exhaust temperature may be set to 45 degrees Celsius, for a 20 degree rise in temperature. The temperature rise may also be as much as 100 degrees Celsius where the heat generating equipment can operate at such elevated temperature. For example, the temperature of the air exiting the equipment and entering the warm-air plenum may be 48, 50, 54, 58, 62, 66, 70, 74, 78, 82, 86, 90, 94, or 98 degrees Celsius. Such a large rise in temperature generally runs contrary to teachings that cooling of heat-generating electronic equipment is best conducted by washing the equipment with large amounts of fast-moving, cool air. Such a cool-air approach does cool the equipment, but it also uses lots of energy. Cooling of particular electronic equipment, such as microprocessors, may be improved even where the flow of air across the trays is slow, by attaching impingement fans to the tops of the microprocessors or other particularly warm components, or by providing heat pipes and related heat exchangers for such components.

The heated air may be routed upward into a ceiling area, or attic 105, and may be gathered there by air handling unit 110, which may include, for example,

5 one or more fans, such as centrifugal fans appropriately sized for the task. The air handling unit 110 may then deliver the air back into a plenum 108 located in the workspace 106. The plenum 108 maybe simply a bay-sized area in the middle of a row of racks, that has been left empty of racks, and that has been isolated from any warm-air plenums on either side of it. o Cooling coils 112a, 112b may be located on opposed sides of the plenum approximately flush with the fronts of the racks. The coils may have a large surface area and be very thin so as to present a low pressure drop to the system 100. In this way, slower, smaller, and quieter fans may be used to drive air through the system. Protective structures such as louvers may be placed in front 5 of the coils 112a, 112b to prevent them from being damaged.

In operation, fan 110 pushes air down into plenum 108, causing increased pressure in plenum 108 to push air out through cooling coils 112a, 112b. As the air passes through the coils 112a, 112b, its heat is transferred into the water in the coils 112a, 112b, and the air is cooled. 0 The speed of the fan 110 and/or the flow rate or temperature of cooling water flowing in the cooling coils 112a, 112b may be controlled in response to measured values. For example, the pumps driving the cooling liquid may be variable speed pumps that are controlled to maintain a particular temperature in work space 106. Such control mechanisms may be used to maintain a constant 5 temperature in workspace 106 or plenums 104a, 104b and attic 105.

The workspace 106 air may then be drawn into racks 102a, 102b such as by fans mounted at the back edge of the many trays that are mounted in racks 102a, 102b. This air may be heated as it passes over the trays and through power supplies running the computers on the trays, and may then enter the warm-air 0 plenums 104a, 104b. Each tray may have its own power supply and fan, with the power supply at the back edge of the tray, and the fan attached to the back of the power supply. AU of the fans may be configured or programmed to deliver air at a single common temperature, such as at a set 45 degrees Celsius. The process may then start again as fan 110 captures and delivers the warm air.

Additional items may also be cooled using system 100. For example, room 116 is provided with a self-contained fan-coil unit 114 which contains a fan and a cooling coil. The unit 114 may operate, for example, in response to a thermostat provided in room 116. Room 116 may be, for example, an office or other workspace ancillary to the main portions of the data center 101.

In addition, as shown in more detail below with respect to FIG. 3, supplemental cooling may also be provided if necessary. For example, a standard roof-top or similar air-conditioning unit may be installed to provide particular cooling needs on a spot basis. As one example, system 100 may be designed to deliver 78 degree Fahrenheit supply air, and workers may prefer to have an office that is cooler. Thus, a dedicated air-conditioning unit may be provided for the office. This unit may be operated relatively efficiently, however, where its coverage is limited to a relatively small area of a building or a relatively small part of the heat load from a building.

Fresh air may be provided to the workspace 106 by various mechanisms. As will be seen in later figures, a supplemental air-conditioning unit, such as a standard roof-top unit may be provided to supply necessary exchanges of outside air. Also, such a unit may also serve to dehumidify the work space for the limited latent loads in the system 100, such as human perspiration. Alternatively, louvers may be provided from the outside environment to the system 100, such as powered louvers to connect to the warm air plenum 104b. System 100 may be controlled to draw air through the plenums when environmental ambient humidity and temperature are sufficiently low to permit free cooling with outside air. Such louvers may also be ducted to fan 110, and warm air in plenums 104a, 104b may simply be exhausted to atmosphere, so that the outside air does not mix with, and get diluted by, the warm air from the computers. Appropriate filtration may also be provided in the system, particularly where outside air is used.

Also, the workspace 106 may include heat loads other than the trays, such as from people in the space and lighting. Where the volume of air passing through the various racks is very high and picks up a very large thermal load from multiple computers, the small additional load from other sources may be negligible, apart from perhaps a small latent heat load caused by workers, which may be removed by a smaller auxiliary air handling unit as described above.

Cooling water may be provided from a cooling water circuit powered by pump 124. The cooling water circuit may be formed as a direct-return, or indirect-return, circuit, and may generally be a closed-loop system. Pump 124 may take any appropriate form, such as a standard centrifugal pump. Heat exchanger 122 may remove heat from the cooling water in the circuit. Heat exchanger 122 may take any appropriate form, such as a plate-and-frame heat exchanger or a shell-and-tube heat exchanger.

Heat may be passed from the cooling water circuit to a condenser water circuit that includes heat exchanger 122, pump 120, and cooling tower 118. Pump 120 may also take any appropriate form, such as a centrifugal pump. Cooling tower 118 may be, for example, a forced draft tower or an induced draft tower. The cooling tower may be considered a free cooling source, because it requires power only for movement of the water in the system and the powering of a fan to cause evaporation; it does not require operation of a compressor in a chiller or similar structure.

As shown, the fluid circuits create an indirect water-side economizer arrangement. This arrangement may be relatively energy efficient, in that the only energy needed to power it is the energy for operating several pumps and fans. In addition, this system may be relatively inexpensive to implement, because pumps, fans, cooling towers, and heat exchangers are relatively technologically simple structures that are widely available in many forms. In addition, because the structures are relatively simple, repairs and maintenance may be less expensive and easier to complete. Such repairs may be possible without the need for technicians with highly specialized knowledge.

Alternatively, direct free cooling may be employed, such as by eliminating heat exchanger 122, and routing cooling tower water (condenser water) directly to cooling coils 112a, 112b. Such an implementation may be more efficient, as it removes one heat exchanging step. However, such an implementation also causes water from the cooling tower 118 to be introduced into what would otherwise be a closed system. As a result, the system in such an implementation may be filled with water that may contain bacteria, and may also be filled with other contaminants in the water.

Control valve 126 is provided in the condenser water circuit to supply make-up water to the circuit. Make-up water may generally be needed because cooling tower 118 operates by evaporating large amounts of water from the circuit. The control valve 126 may be tied to a water level sensor in cooling tower 118, or to a basin shared by multiple cooling towers. When the water falls below a predetermined level, control valve 126 may be caused to open and supply additional makeup water to the circuit. Optionally, a separate chiller circuit may be provided. Operation of system 100 may switch partially or entirely to this circuit during times of extreme atmospheric ambient (i.e., hot and humid) conditions or times of high heat load in the data center 101. Controlled mixing valves 134 are provided for electronically switching to the chiller circuit, or for blending cooling from the chiller circuit with cooling from the condenser circuit. Pump 128 may supply condenser water to chiller 130, and pump 132 may supply chilled water, or cooling water, from chiller 132 to the remainder of system 100. Chiller 130 may take any appropriate form, such as a centrifugal, reciprocating, or screw chiller, or an absorption chiller. The chiller circuit may be controlled to provide various appropriate temperatures for cooling water. In some implementations, the chilled water may be supplied exclusively to a cooling coil, while in others, the chilled water may be mixed, or blended, with water from heat exchanger 122, with common return water from a cooling coil to both structures. The chilled water may be supplied from chiller 130 at temperatures elevated from typical chilled water temperatures. For example, the chilled water may be supplied at temperatures of 13 degrees Celsius (55 degrees Fahrenheit) to 18 to 21 degrees Celsius (65 or 70 degrees Fahrenheit). The water may then be returned at temperatures like those discussed below, such as 15 to 80 degrees Celsius, hi this approach that uses sources in addition to, or as an alternative to, free cooling, increases in the supply temperature of the chilled water can also result in substantial efficiency improvements for the system 100. W

Pumps 120, 124, 128, 132, may be provided with variable speed drives. Such drives may be electronically controlled by a central control system to change the amount of water pumped by each pump in response to changing set points or changing conditions in the system 100. For example, pump 124 may be controlled to maintain a particular temperature in workspace 106, such as in response to signals from a thermostat or other sensor in workspace 106.

In operation, system 100 may respond to signals from various sensors placed in the system 100. The sensors may include, for example, thermostats, humidistats, flowmeters, and other similar sensors, hi a simple implementation, one or more thermostats may be provided in warm air plenums 104a, 104b, and one or more thermostats may be placed in workspace 106. In addition, air pressure sensors may be located in workspace 106, and in warm air plenums 104a, 104b. The thermostats may be used to control the speed of associated pumps, so that if temperature begins to rise, the pumps turn faster to provide additional cooling waters. Thermostats may also be used to control the speed of various items such as fan 110 to maintain a set pressure differential between two spaces, such as attic 105 and workspace 106, and to thereby maintain a consistent airflow rate. Where mechanisms for increasing cooling, such as speeding the operation of pumps, are no longer capable of keeping up with increasing loads, a control system may activate chiller 130 and associated pumps 128, 132, and may modulate control valves 134 accordingly to provide additional cooling.

Various values for temperature of the fluids in system 100 may be used in the operation of system 100. hi one exemplary implementation, the temperature in warm air plenums 104a, 104b may be selected to be at or near a maximum temperature for trays in racks 102a, 102b. This maximum temperature may be selected, for example, to be a known failure temperature for components in the trays, or may be a sufficient temperature below such a known failure temperature, hi certain implementations, a temperature of 45°C may be selected. In other implementations, temperatures of 25⁰C to 125⁰C may be selected.

Higher temperatures may be particularly appropriate where alternative materials are used in the components of the computers in the data center, such as high temperature gate oxides and the like. In one implementation, supply temperatures for cooling water may be 20°C, while return temperatures may be 40°C. In other implementations, temperatures of 10°C to 29°C may be selected for supply water, and 15°C to 80°C for return water. Chilled water temperatures may be produced at much lower levels according to the specifications for the particular selected chiller. Cooling tower water supply temperatures may be generally about the wet bulb temperature under ambient atmospheric conditions, while cooling tower return water temperatures will depend on the operation of the system 100.

FIG. 2 is a psychrometric chart showing a heating and cooling cycle for air in a data center. A psychrometric chart graphically represents the thermodynamic properties of moist air (which is air containing any appreciable moisture, and not merely air that would feel moist to a person). The chart is from ASHRAE Psychrometric Chart No.l, which defines properties of air for sea level applications. See 1997 ASHRAE Handbook - Fundamentals, at page 6.15. Other charts may also be used, and the chart shown here is merely used to exemplify certain aspects of the concepts discussed in the present document. The psychrometric chart is criss-crossed with a number of lines that represent various properties of air. Cooling and heating processes on air can be analyzed by identifying a point on the chart that represents air at a particular condition (e.g., temperature and humidity), and then locating another point that represents the air at another condition. A line between those points, generally drawn as a straight line, maybe fairly assumed to represent the conditions of the air as it moves from the first condition to the second, such as by a cooling process. Several properties will be discussed here. First, saturation temperature is an arc along the left of the graph and represents the temperature at which air becomes saturated and moisture begins coming out of the air as a liquid — also known popularly as the "dew point." When the temperature of air is taken below the dew point, more and more water comes out of the air because the cooler air is capable of holding less water.

The dry bulb temperature of air is listed along the bottom of the graph and represents what is popularly viewed as temperature, i.e., the temperature returned by a typical mercury thermometer. The chart shows two numbers relating to the humidity of air. The first is the humidity ratio, listed along the right edge of the graph, and is simply the weight of moisture per each unit weight of dry air. Thus, the humidity ratio will stay constant at various temperatures of air, until moisture is removed from the air, such as by pushing the air temperature down to its dew point (e.g., the moisture comes out of the air and ends up on the grass in the morning) or by putting moisture into the air (e.g., by atomizing water into such as fine mist in a humidifier that the mist can be supported by the natural motion of the air molecules). Thus, when graphing processes that involve simple changes in air temperature, the point that represents the state of the air will move straight left and right along the graph at a constant humidity ratio. That is because the dry- bulb temperature will go up and down, but the humidity ratio will stay constant.

The second humidity parameter is the so-called relative humidity. Unlike the humidity ratio, which measures the absolute amount of moisture in the air, the relative humidity measures the amount of moisture in the air as a percent of the total moisture the air could possibly hold at its current temperature. Warmer air can hold more moisture than can colder air, because the molecules in warmer air are moving more rapidly. Thus, for an equal amount of moisture in the air (i.e., an equal humidity ratio), the relative humidity will be lower at a high temperature than at a low temperature.

As one example, on a summer day when the overnight low was 55 degrees Fahrenheit and there is dew on the ground, and the daytime high is around 75 degrees Fahrenheit, the relative humidity in the morning will be about 100% (the dew point), but the relative humidity in the afternoon will be a very comfortable 50%, even if one assumes that the identical amount of water is in the air at both times. This exemplary process is shown in FIG. 2 by the points marked A and B, with point B showing saturated air at 55 degrees Fahrenheit (the overnight low), and point A representing that same air warmed to 75 degrees Fahrenheit (the daytime temperature) . Commercial air handling systems take advantage of this same process in providing conditioned air in a building. Specifically, systems may gather air from an office space at 75 degrees Fahrenheit and a relative humidity of 60 percent. The systems pass the air through a cooling coil that looks like an automobile radiator to cool the air to 55 degrees Fahrenheit, which will typically push the air down to its dew point. This will make moisture pour out of the air as it passes through the cooling coil. The moisture can be captured in drains below the cooling coil and then be removed from the building. The air can then be returned to the work space, and when it warms back to 75 degrees, it will be a very comfortable 50 percent relative humidity.

This common cooling process is shown by points A, B, C, and D on the chart of FIG. 2. Point A shows the 75 degree air at 70 percent relative humidity. Point B shows the air cooled to its dew point, which it hits at a temperature (dry bulb) of about 65 degree Fahrenheit. Further cooling of the air to 55 degrees

Fahrenheit (to point C) rides along the saturation curve, and water will come out of the air during that portion of the cooling. Finally, the state of the air moves to point D as the air is mixed in with other room air and reaches 75 degrees Fahrenheit again. At this point, the relative humidity will be 50 percent (assuming it does not pick up additional moisture from the room or the existing room air) rather than the original 60 percent because the cooling process has dehumidified the air by pulling moisture out of the air in the cooling coil. If the room air contains more moisture than does the cooled air, point D will be slightly above its position shown in FIG. 2, but still below point A. Such a common process brings with it a number of challenges. First, to cool the air to 55 degrees Fahrenheit, the system must provide cooling water in the cooling coil that can absorb all the heat. Such water would need to be at least cooler than 55 degrees Fahrenheit. It may be expensive to create such cool water — requiring systems such as chillers and other energy-intensive systems. In addition, the area immediately around the pipes that supply the cooling water will be cooler than 55 degrees Fahrenheit, i.e., cooler than the dew point of the air if the pipes run through the workspace or through air having the same state as the air in the workspace. As a result, moisture from the air may condense on the pipe because the air has fallen to its dew point. Thus, insulation may be required around the cool pipes to prevent such condensation, and condensation might occur in any event, and cause rusting or even mold or other problems. Finally, as any high school physic student likely knows, it takes a lot of energy to wring moisture out of water, i.e., to change the water from one state to another. The warm air cooling features discussed with respect to FIG. 1 above may, in certain implementations, avoid one or more of these challenges. An exemplary warm air cooling process is shown on the graph of FIG. 2 by points E and F. Point E shows a room-air condition in a workspace that is near the top of, but within, common guidelines for comfort levels for people dressed in summer clothing. See 1997 ASHRAE Handbook - Fundamentals, at page 8.12. That condition is 75 degrees Fahrenheit (24 degrees Celsius), and a relative humidity of about 70 percent. Point F shows heating of that air without the addition of moisture, such as by passing the air over heat-generating computer components in a rack-mounted server system. The heat rise is 36 degrees Fahrenheit (20 degree Celsius) to bring the air to a state of 111 degrees Fahrenheit (44 degrees Celsius) at about 23 percent relative humidity. The air may then be cooled to its original temperature of 75 degrees Fahrenheit in a cooling coil before it is reintroduced to the work space. Points G and H on the graph represent the condition of air in the space immediately surrounding cooling pipes. It is assumed for this example that the cooling supply water is 68 degrees Fahrenheit (20 degrees Celsius) and the return temperature is 104 degrees Fahrenheit (40 degrees Celsius). It is also assumed that the air near the pipes will contain the same moisture level as the rest of the air in the space, and that the air immediately surrounding the pipe takes on the same temperature as the water inside the pipe. As can be seen, this air associate with the cooling water also stays above the saturation point, so that there should be no condensation on the cooling water pipes, and thus no need for insulation on the pipes. It can be seen by this process that the air never becomes saturated. As a result, the system need not provide energy to create a phase change in the air. In addition, the system need not provide liquid recovery structures at the cooling coil, or pipe insulation for anywhere. Other similar temperatures, and in many implementations warmer temperatures, may be used. The particular temperatures discussed here are meant to be exemplary only.

FIG. 3 is a plan view in schematic form of a hydronic cooling system for a data center 300. The data center 300 includes an existing plant 302, and a future plant 304. As will be seen, the system is highly modular, and can therefore be scaled from a very small system to a very large system by adding additional components and subsystems. Although the existing and future systems are shown as separate systems, extensions of the existing system may also be employed for future expansion. For example, oversized piping may be installed in some implementations and may be tapped for future additions of added components when the data center 300 is expanded. Also, other forms of cooling loops, or additional or alternative cooling loops may be employed.

The cooling system for existing plant 302 is similar to that shown in FIG. 1. Cold air is supplied to a workspace 306 by air handling unit 310. Air handling unit 310 may include a supply fan (not shown) that draws warm air out of a plenum, such as in an above-ceiling attic space, and pushes the air through cooling coils 312, 314. The warm air may be produced by passing cooled air over computers, such as computer 318, mounted on trays in a rack system. Air passing over the computers may be collected in warm-air plenums 316, and may be routed to the above-ceiling space or another appropriate space. hi alternative implementations, multiple air handling units that include a fan and a cooling coil may be placed immediately above or below the warm air plenums 316. Such air handling units may take the form of long banks of small centrifugal fans and long heat exchange units mounted to the fans. The fans may pull air or push air through the coils. Pushing air may have the advantage of being quieter, as the coils may block out a certain amount of the fan noise. Also pushing of air may be more efficient. Pulling of air may provide a benefit of allowing a limited number of fans to operate on a much larger bank of coils, as all the pulling fans can be connected to a plenum, and may create a relative vacuum behind the coils to pull air through. In such an arrangement, if one of the fans breaks down, the others can more easily provide support across the entire coil length.

Cooling water may be provided by pumps 326 to cooling coils 312, 314 to cool the air. These pumps 326 may draw cooled water from heat exchangers 324, and drive the water into existing plant 302, through cooling coils 312, 314, and back through heat exchanger 324, where the heat acquired from coils 312, 314, may be removed from the water. Heat in the cooling water may be removed from heat exchangers 324 by cooled condenser water supplied through pump 322 from cooling towers 320. The system is shown for exemplary purposes as having two parallel circuits for cooling water and one circuit for condenser water from the cooling towers 320. However other appropriate arrangements for the piping system may also be appropriate. For example, a single heat exchanger may be used, as may a single cooling tower. Other sources of free cooling may also be used, such as ground water or deep lake cooling.

Also, additional space may be provided for extra cooling towers and future heat exchangers, and piping may be sized to accommodate future expansion. For example, cooling towers may be provided in an extendable stacked arrangement, so that a third tower will be piped in parallel with the first two towers 320 when the additional cooling is needed.

Existing plant 302 includes an additional workspace 308. This space may be for example, an office space, control room, or storage room. The space may be designed to be inhabited extensively by people, and may require lower temperatures than does workspace 306. As a result, workspace 308 is provided with air-conditioning unit 330, which may be a standard rooftop air conditioning unit. Although air-conditioning unit 330 may require additional electricity for providing higher degrees of cooling, including latent heat removal, the unit 330 generally serves an area and a thermal load that is much smaller than the entire system. As a result, unit 330 may be much smaller than what would otherwise be required to provide similar low-temperature cooling for the rest of the system. Although not shown, one or more chillers and chillers circuits may be provided to operate the system, in addition to the cooling water circuit, whether together or alternatively. The piping of the chiller circuit may take a form similar to that shown schematically in FIG. 1.

As noted above, future plant 304 may be provided in a variety of ways. In general, a modular approach may be used, so that additional heat exchangers are added in a size commensurate with the additional loads that are expected. Certain components may be piped as entirely separate circuits, while others may be extensions of the original system. Sharing components between old and new systems may cause construction-based interruptions in the existing system, but may also provide for better utilization of the components in the full system. For example, condenser water can be shared through the system, so that if each half of the system requires cooling from 1.5 cooling tower, only three towers will be operated. If the two systems are separate, each system would have to run two towers, and thereby use additional energy.

FIG. 4 is a plan view in schematic form of a hydronic cooling system for a data center housed in shipping containers 400 and 404. The system is arranged in a manner similar to the system in FIG. 3. Here, however, cooling coils are mounted along the length of both of shipping containers 400, 404, such as below or above the computer-holding racks in the system. For example, elevated false floor may be provided down the center of the shipping containers 400, 404, and squirrel cage fans may blow warm air through cooling coils into the space under the grate. The cooled air may then pass up through the floor to be drawn by fans associated with each tray in the racks, into warm-air plenums along the sides of the shipping containers, and may be routed back through the squirrel cage fans. Using exterior space along the outer walls for the warm-air plenum may have the advantage of minimizing the amount of atmospheric heat that passes into each container.

In this implementation, the various benefits of using little energy, of modularity, and of standardization on low complexity components may be appreciated. Specifically, shipping containers 400, 404 may be outfitted in a central location using standardized techniques and specialized, well-trained labor — a type of prefab construction approach. The shipping containers 400, 404 may then be delivered to a site, such as a site having adequate electrical, water, and data services, and may be easily connected to such services, and then begin their operation.

The other components of the system may be provided separately to the site. For example heat exchangers and pumps may be provided on preassembled racks, dropped into the place, and connected to other components. Cooling towers may also be delivered to a site and connected relatively quickly. Also, because the system may operate without the need for components such as chillers that demand high power levels, the system may be installed in more areas, such as remote areas having lower levels of electrical service. As shown, the system is also highly modular, so that additional resources may be added relatively quickly and inexpensively as they are needed. Containers 402 are shown as part of an existing system, and containers 404 (with associated equipment) are shown as part of a future addition to the system. Other modules may also be added as the need arises, until the level of available utilities is tapped out. Each container may contain hundreds or thousands of trays, so that a complete installation could include hundreds of thousands of computers.

Systems like those shown in this document may be deployed in a number of different and flexible manners. For example, locations may be selected according to the availability of resources. For example, cooling water may be obtained from ground water or surface water for free cooling processes. In such an environment, cooling towers may be eliminated from the system.

In addition, systems may be deployed far from suitable construction labor (and may use labor having a lower skill level) if they are assembled centrally and shipped to a site.

Such systems may also be self-contained. For example, fuel cells or gas- powered turbines may be deployed with a system to generate electricity, and may be supplemented by wind or solar power. In addition, generators, such as combustion turbines may be selected for less than the full load of a system, but may be deployed to permit peak-shaving of the power required by a system. Such peak-shaving may permit a system to operate in an area that cannot provide all of the power needed for a system. In addition, peak shaving may allow an operator of the system to negotiate energy rates that are lower than would otherwise be available, because the system would not be drawing high levels off the grid when the electrical provider does not have such extra capacity. Moreover, heat from the turbines may be used to help power equipment such as an ammonium-based absorption chiller. In addition, Peltier-like devices may be used to convert heat into electrical energy. In addition, excess pressure from make-up water applied to the cooling towers may also be used to drive small turbines for limited electrical production.

Such systems may also be co-located with other systems that may have use for the hot water generated through the cooling of electronic components. For example, a data center may be located near a hospital or other high density residence to supply heating water, such as for space heating or water heating. In addition, the hot water may be used in common with various processes, such as manure digestion, yogurt culture fabrication, and ethanol production. Moreover, the hot water may be used to heat swimming pools or other buildings, or may be run through ice and snow melt piping, such as in airport runways. Likewise, heat from the system may be used to control temperatures in a co-located fish farm. Other implementations may also be employed to serve particular needs for data services, particularly where distribution of processing power closer to end users is needed.

FIG. 5 shows container-based data centers mounted on a docked container ship and a barge. These are simplified diagrams showing one particular implementation of a container-based data center. Specifically, data centers may be constructed modularly in shipping containers at a central location, and then may be transported to loading docks in a conventional manner. In some implementations, the shipping containers may be moved off the boats, and may be deployed onto land. The cooling systems and other systems to support the data centers may be shipped with the systems, or may be prepared on land while the containers are being prepared and shipped. In addition, such portable systems (i.e., capable of being loaded onto a truck or train, and transported to another location for use) are also capable of being redeployed to another fixed location relatively easily. For example, if a particular location faces unexpected data loads, and another location has lower- than-expected loads, containers may be moved from the second location to the first.

As shown in FIG. 5, however, the systems are being used on the boats. Specifically, the boats are docked, with utility lines running to the docks. The utility lines may include electrical, data, and cooling water lines, such as lines for make-up water if cooling towers are located on the boats. Cooling towers or other cooling mechanisms may also be located near the dock, as may heat exchangers, and condenser or cooling water may be piped to the boats.

Alternatively, cooling water may be obtained out of the body of water in which the boats sit, such as when free cooling using deep lake cooling. In addition, generators on the boat or other sorts of power plants on the boats, may generate electricity for the data centers. In such situations, a boat may be anchored offshore, and may attach to an offshore data connection, such as an optical fiber connection, hi this manner, computing power may be easily projected geographically on short notice, and may be alternatively used at sea and on land as the need arises. For example, where a large event is expected to occur, or a large number of computer users is expected to congregate, near a port, such a data center may be supplied at the port.

The containers may be networked for electricity, data, and cooling, in an appropriate manner. In addition, access ladders or other structures may be provided where the containers are stacked on top of each other.

FIG. 6 is a flowchart 600 showing steps for cooling a data center using elevated temperatures. The method is exemplary only; other steps may be added, steps may be removed, and the steps may be performed in different orders than those shown, as appropriate.

At box 602, the exhaust temperatures of various trays in a computer rack system are maintained at a computed value, such as a maximum value for maintaining the trays in reliable operation. The trays may each be self-throttling, in that a fan and related temperature sensor may be associated with each tray. At the same time, other components of the system may maintain a workspace temperature in front of the racks at a particular temperature, such as a maximum allowed temperature. The maximum allowed temperature may be, for example, a maximum temperature at which the workers in the workspace are willing to operate. It may also be a temperature set by law or regulation, such as a maximum 40 degree Celsius limit. Federal OSHA and California OSHA guidelines may provide limitations on such temperatures. The set temperature may also change at appropriate times; for example, the temperature may be higher during times when people are not expected to be in the workspace.

At box 606, the workspace cooling is changed in response to changes in the system. Such changes may occur, for example, if the set temperature for the space changes. The changes may also occur, for example, if the temperature remains constant but the load on the system changes — such as if an additional rack of computers is powered up, or if the computers are loaded more heavily. The cooling may change, for example, by increasing the rate of pumping cooling fluid through cooling coils. Alternatively, additional coils may be brought online, or a chiller may be started to lower the temperature of cooling water.

At box 608, the exhaust temperature of air from the computer racks is maintained at the maximum computer value. Specifically, although the workspace temperature has changed, and other parameters in the system may have changed, the exhaust temperature may remain unaffected, as each fan associated with a tray remains set on one exhaust temperature.

A number of embodiments of the invention have been described. Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of the invention. For example, the steps of the exemplary flow chart on FIG. 6 may be performed in other orders, some steps may be removed, and other steps may be added. Accordingly, other embodiments are within the scope of the following claims.

Claims

WHAT IS CLAIMED IS:

1. A method of providing cooled air to electronic equipment, comprising: capturing heated air from a volume containing electronic equipment, cooling the heated air by more than fifteen degrees Celsius in an air-to- water heat exchanger; and supplying cooling water to the air-to-water heat exchanger at a temperature above a dew point temperature of the heated air.

2. The method of claim 1, wherein the step of capturing the heated air comprises capturing substantially only the heated air, and capturing substantially no air that is not from the volume containing the electronic equipment.

3. The method of claim 1 , wherein the cooling water is returned from the air-to- water heat exchanger at a temperature above 40 degrees Celsius.

4. The method of claim 1 , further comprising returning water from the air-to- water heat exchanger, wherein the temperature of the supplied water is more than 15 degrees Celsius cooler than the temperature of the returned water.

5. The method of claim 1 , further comprising returning water from the air-to- water-heat exchanger and transferring heat from the returned water to water supplied from a free cooling water source.

6. The method of claim 5, wherein the free cooling water source comprises one or more cooling towers.

7. The method of claim 5, wherein the temperature of the returned water is more than 10 degrees Celsius warmer than the water supplied from the free cooling water source.

8. The method of claim 5, wherein the water supplied from the free cooling water source defines a first temperature delta from the transferring of heat, and the water returned from the air-to-water heat exchanger defines a second temperature delta, and wherein the first temperature delta is more than 30 percent less than the second temperature delta.

9. A method of providing cooled air to electronic equipment, comprising: heating air from a workspace by a temperature differential of more than 20 degrees Celsius by passing it over electronic equipment, cooling the heated air by approximately the same temperature as the temperature differential; and returning the air to the workspace.

10. The method of claim 9, further comprising cooling the air using free cooling.

11. The method of claim 10, wherein the free cooling comprises transferring heat from the air to cooling water, and transferring heat from the cooling water to condenser water that undergoes open evaporation.

12. The method of claim 10, wherein heating the air from the workspace comprises heating the air from about 25 degrees Celsius to more than 45 degrees Celsius.

13. The method of claim 12, further comprising capturing the heated air and controlling the temperature of the captured air by controlling the rate of airflow over the electronic equipment.

14. The method of claim 13, wherein the controlled temperature is at or near a maximum failure preventative temperature for the electronic equipment.

15. The method of claim 9, wherein the air is returned directly to the workspace after it is cooled.

16. The method of claim 9, wherein the electronic equipment comprises a large plurality of rack servers in a data center.

17. A system for providing cooled air to electronic equipment, comprising: a free cooling water source; a water-to-water heat exchanger in fluid communication with the cooling tower; and an air-to-water heat exchanger in fluid communication with the water-to- water heat exchanger and positioned to received heated air from a group of electronic devices, wherein the water-to-water heat exchanger is configured to supply cooling water to the water-to-air heat exchanger at a temperature above a dew point of air surrounding the water-to-air heat exchange.

18. The system of claim 17, wherein the free cooling water source comprises a cooling tower.

19. The system of claim 17, wherein the water-to-water heat exchanger is configured to supply cooling water at 15-35 degrees Celsius.

20. A system for providing cooled air to electronic equipment, comprising: a cooling tower; a water-to-water heat exchanger connected by a first pair of pipes to the cooling tower; and an air-to-water heat exchanger connected by a second pair of pipes to the water-to-water heat exchanger and positioned to received heated air from a group of electronic devices, wherein the first pair of pipes and the second pair of pipes are substantially lacking in insulation or condensation collecting apparatuses.