BACKGROUND OF INVENTION1. Field of Invention
Embodiments of the invention relate generally to devices and methods for cooling a room, such as a data center, equipment room or wiring closet. Specifically, aspects of the present invention relate to data centers containing racks and enclosures used to house data processing, networking and telecommunications equipment, and more particularly to cooling systems and methods used to cool equipment housed by such racks and enclosures.
2. Discussion of Related Art
Over the years, a number of different standards have been developed to enable equipment manufacturers to design rack mountable equipment that can be mounted in standard racks manufactured by different manufacturers. A standard rack typically includes front mounting rails to which multiple units of electronic equipment, such as servers and CPUs, are mounted and stacked vertically within the rack. An exemplary industry standard rack is approximately six to six-and-a-half feet high, by about twenty-four inches wide, and about forty inches deep. Such a rack is commonly referred to as a “nineteen inch” rack, as defined by the Electronics Industries Association's EIA-310-D standard.
Nineteen inch racks are used extensively in the data centers and other large facilities described above. With the proliferation of the Internet, it is not uncommon for a data center to contain hundreds of these racks. Further, with the ever decreasing size of computer equipment, and in particular, computer servers and blades, the number of electrical devices mounted in each rack has been increasing, raising concerns about adequately cooling the equipment.
Heat produced by rack-mounted equipment can have adverse effects on the performance, reliability and useful life of the equipment components. In particular, rack-mounted equipment, housed within an enclosure, may be vulnerable to heat build-up and hot spots produced within the confines of the enclosure during operation. The amount of heat generated by a rack of equipment is dependent on the amount of electrical power drawn by equipment in the rack during operation. In addition, users of electronic equipment may add, remove, and rearrange rack-mounted components as their needs change and new needs develop.
Previously, in certain configurations, data centers have been cooled by computer room air conditioner (“CRAC”) units that are typically hard piped, immobile units positioned around the periphery of the data center room. These CRAC units intake air from the fronts of the units and output cooler air upwardly toward the ceiling of the data center room. In other embodiments, the CRAC units intake air from near the ceiling of the data center room and discharge cooler air under a raised floor for delivery to the fronts of the equipment racks. In general, such CRAC units intake room temperature air (at about 72° F.) and discharge cold air (at about 55° F.), which is blown into the data center room and mixed with the room temperature air at or near the equipment racks.
The rack-mounted equipment typically cools itself by drawing air along a front side or air inlet side of a rack, drawing the air through its components, and subsequently exhausting the air from a rear or vent side of the rack. A disadvantage of the CRAC-type air conditioning system is that cool air is mixed with the room temperature air, which is inefficient. Ideally, to make the system as efficient as possible, and to utilize as little energy and floor space as possible, the highest possible temperature air should be drawn into the CRAC units and the outlet air generated by the CRAC should be a few degrees below room temperature. In addition, airflow requirements can vary considerably as a result of different numbers and types of rack-mounted components and different configurations of racks and enclosures.
For large data centers requiring CRAC units at or near the middle or center of the data center room, delivery of coolant to the CRAC units must be located within the raised floor since it is undesirable to secure coolant piping to the ceiling of the data center due to risks involved with the possible failure of the piping joints. Specifically, with traditional CRAC systems, the piping of the units requires significant cutting and hand soldering of pipes. Leaks are common and leaking water or coolant in a data center may result in risk of damage to equipment housed within the equipment racks. For at least these reasons, most data center designers and operators are unwilling to consider overhead piping for cooling a data center.
SUMMARY OF INVENTIONOne aspect of the invention is directed to a cooling unit comprising a first module including a first intake opening and a first exhaust opening formed therein. At least one first air moving device may be configured to draw air along a first flow path from the first intake opening to the first exhaust opening. A first heat exchanger may be positioned in the first flow path. The cooling unit may further comprise a second module including a second intake opening and a second exhaust opening. At least one second air moving device may be configured to draw air along a second flow path from the second intake opening and direct air to the second exhaust opening. A second heat exchanger may be positioned in the second flow path. A conduit system is in fluid communication with the first heat exchanger and the second heat exchanger. The conduit system may be configured to deliver coolant from the first heat exchanger of the first module to the second heat exchanger of the second module and from the second heat exchanger of the second module to the first heat exchanger of the first module.
In one embodiment, the first module and the second module may be contained within a housing, with the first module being positioned over the second module. The cooling unit may further comprise a tray to support the at least one first air moving device of the first module, with the tray being releasably secured to the housing. In a certain embodiment, the tray may be disposed on a generally horizontal plane and supported by at least one support rail configured to secure the tray to the first module. The arrangement is such that the at least one first air moving device of the first module includes a fan, and the tray is configured to receive at one fan. The cooling unit may be configured to comprise a control device coupled to the fan. In one embodiment, the fan is a variable speed fan, and the control device is configured to control the speed of the fan. The first module may comprise a first plenum disposed along a generally vertical axis in fluid communication with the first intake opening, and a second plenum disposed along a generally horizontal axis and configured to direct air from the first plenum to the tray. The first module may further comprise a top having the first intake opening and the first exhaust opening formed therein, a first plenum disposed along a generally vertical axis in fluid communication with the first intake opening, and a second plenum disposed along a generally horizontal axis and configured to direct air taken from the first plenum to the at least one first air moving device of the first module. The second module may further comprise a first side having the second intake opening and a second side, opposite the first side, having the second exhaust opening. The arrangement is such that the first flow path is along a generally horizontal axis. In another embodiment, the first heat exchanger of the first module is disposed at an acute angle with respect to the flow of air from the second plenum to the first exhaust opening of the first module, and the second heat exchanger of the second module is disposed at an acute angle with respect to the flow of air from the first side to the second side of the second module. The first heat exchanger may comprise a condenser unit, and the second heat exchanger may comprise an evaporator unit. The cooling unit may be further configured to comprise a compressor to deliver coolant to the condenser unit and a bypass valve provided in the conduit system. The bypass valve may be configured to divert a portion of coolant from the compressor to the evaporator unit. In a certain embodiment, an expansion valve is disposed between the condenser unit and the evaporator unit.
Another aspect of the invention is directed to a cooling unit comprising a housing having a top with an intake opening and at least one exhaust opening formed therein. At least one air moving device may be configured to be coupled to the housing to draw air along a flow path from the intake opening to the at least one exhaust opening. A heat exchanger may be configured to be coupled to the housing between the at least one air moving device and the at least one exhaust opening within the flow path. A conduit system, in fluid communication with the heat exchanger, may be configured to deliver coolant to the heat exchanger. Coolant flowing through the heat exchanger may be adapted to be cooled by the heat exchanger.
In a certain embodiment, a tray may be provided to support the at least one air moving device, the tray being releasably secured to the housing. The tray may be disposed on a generally horizontal plane by at least one support rail configured to releasably secure the tray to the housing. The at least one air moving device may include a fan and the tray may be configured to receive at least one fan. A control device may be coupled to the fan. In one embodiment, the fan is a variable speed fan and the control device is configured to detect a speed of each fan. The housing may comprise a first plenum disposed along a generally vertical axis in fluid communication with the intake opening, and a second plenum disposed along a generally horizontal axis and configured to direct air taken from the first plenum to the tray, the first and second plenums defining the flow path. Air that is directed by the at least one air moving device over the heat exchanger is heated. The heat exchanger may be disposed at an acute angle with respect to the flow of air to the at least one exhaust opening.
A further aspect of the invention is directed to a method of cooling a space with a modular cooling unit. The method may comprise: drawing air into a first module of the cooling unit; directing the air over a first heat exchanger of the first module of the cooling unit to cool coolant flowing through the first heat exchanger; exhausting the air out of the first module of the cooling unit; directing the cooled coolant from the first heat exchanger to a second heat exchanger; drawing air into a second module of the cooling unit; directing the air over the second heat exchanger of the second module of the cooling unit to cool the air; and exhausting the cooled air out of the second module of the cooling unit. In one embodiment, the method may further comprise positioning the first module on top of the second module. Drawing air into the first module may include drawing air from a top of the first module, and drawing air into the second module may include drawing air from a side of the second module. The method may further comprise directing coolant from the second heat exchanger to the first heat exchanger and diverting a portion of coolant being directed from the second heat exchanger to the first heat exchanger back to the second heat exchanger.
In yet another embodiment, a cooling unit comprises a housing, a compressor coupled to the housing, a first heat exchanger coupled to the housing, and at least one first air moving device coupled to the housing and configured to direct air along a first flow path. The first heat exchanger may be located along the first flow path. A second heat exchanger may be coupled to the housing. At least one second air moving device may be coupled to the housing and configured to direct air along a second flow path. The second heat exchanger is located along the second flow path. A conduit system may provide fluid communication between the compressor, the first heat exchanger, the second heat exchanger, and the compressor. A bypass valve may be provided in the conduit system. The bypass valve may be configured to divert a portion of coolant from the compressor to the second heat exchanger. In one embodiment, the first heat exchanger comprises a condenser unit and the second heat exchanger comprises an evaporator unit.
BRIEF DESCRIPTION OF DRAWINGSThe accompanying drawings, are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures is represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. In the drawings:
FIG. 1 is a perspective view of a cooling unit of an embodiment of the present invention;
FIG. 2 is a perspective view of components of the cooling unit shown inFIG. 1;
FIG. 3 is a front elevational view of the cooling unit with a front panel of the cooling unit removed;
FIG. 4 is a view similar toFIG. 3 illustrating a bottom air blocking panel removed;
FIG. 5 is a perspective view of the cooling unit shown inFIG. 4 with a fan tray assembly partially removed from the cooling unit;
FIG. 6 is a perspective view similar toFIG. 1 with side panels of the cooling unit removed;
FIG. 7 is a system block diagram of the cooling unit of an embodiment of the invention;
FIG. 8 is a system block diagram of the operational states and modes of the cooling unit;
FIG. 9 is a system block diagram of the condensate state;
FIG. 10 is a system block diagram of the hot gas bypass valve state;
FIG. 11 is a system block diagram of the hot gas bypass valve control;
FIG. 12 is a system block diagram of the condenser fans speed control;
FIG. 13 is a system block diagram of the evaporator fans speed control; and
FIGS. 14-17 are flow charts showing the calculation of cooling capacity using equations of embodiments of the invention.
DETAILED DESCRIPTIONThis invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. Also, the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of “including,” “comprising,” or “having,” “containing”, “involving”, and variations thereof herein, is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.
At least one embodiment of the present invention is directed to a modular cooling system that is selectively configurable to cool electronic equipment housed within equipment enclosures or racks of a data center. As used herein, “enclosures” and “racks” are used to describe apparatus designed to support electronic equipment. Such a cooling system is capable of employing one or more cooling units on an as needed basis to provide localized cooling within the data center. Specifically, multiple cooling units may be interspersed in a row of equipment racks to more efficiently cool the data center. The circulation path of warm air generated by the electronic equipment is greatly reduced, thereby nearly eliminating the mixing of hot and cold air within the data center.
Data centers are typically large rooms designed, in certain instances, to house hundreds of electronic equipment racks arranged in rows within the data center. The rows of equipment racks are arranged in such a manner that there are cold aisles and hot aisles. The cold aisles provide access to the fronts of the enclosures where the electronic equipment is typically accessed. The hot aisles provide access to the backs of the equipment racks. As requirements change, the number of equipment racks may be increased or decreased depending on the functional requirements of the data center. At least one embodiment of the cooling system is modular and scalable, and may take the form of a kit designed to meet these changing needs. Also, although relatively large data centers are discussed as an intended use for such a cooling system, as mentioned above, the system of the present invention is scalable and may be employed in smaller rooms on a smaller scale and for applications other than data center.
In one embodiment, the cooling system may comprise a plurality of cooling units, each cooling unit having a housing adapted to support components of the cooling system. For example, the components of the cooling unit may include first and second heat exchangers coupled to a conduit system configured to deliver coolant to the heat exchangers. Fans may be provided to move air across the heat exchangers. The cooling unit may be disposed within a row of equipment racks and configured to intake the hot air within the data center from a hot aisle, for example, to cool the air to slightly below ambient temperature. This configuration eliminates the inefficiency of mixing hot air with the room temperature air to obtain a warm air mixture. This configuration may also decrease the latent cooling provided by the data center's air conditioning system thereby decreasing the need for humidification.
In certain embodiments, the components of the cooling unit may be provided in kit form so that the person installing the cooling unit does not require specialized tools. The modular nature of the cooling unit allows the user to optimize the location of each cooling unit since each cooling unit includes the ability to sense and display the capacity of the system, the flow rate, coolant and air inlet and outlet temperatures, and pressure differentials obtained from pressure readings taken throughout the cooling unit. Thus, the cooling unit may be employed and redeployed for maximum efficiency and optimal use within the data center.
A typical data center includes a room designed to house a plurality of equipment racks. In one embodiment, each equipment rack may be constructed in accordance with the teachings disclosed in U.S. patent application Ser. No. 10/990,927, entitled EQUIPMENT ENCLOSURE KIT AND ASSEMBLY METHOD, filed on Nov. 17, 2004, which is owned by the assignee of the present invention and is incorporated herein by reference. Further, cabling between the equipment racks may be implemented using cable distribution troughs contained on the roofs of the racks as disclosed in U.S. Pat. No. 6,967,283, which is incorporated herein by reference and assigned to the assignee of the present invention.
Specifically, the equipment rack includes a frame or housing adapted to support electronic components, such as data processing, networking and telecommunications equipment. The housing includes front, back, sides, bottom and top. The front of each equipment rack may include a front door so as to enable access into the interior of the equipment rack. A lock may be provided to prevent access into the interior of the equipment rack and the equipment housed by the rack. The sides of the equipment rack may include at least one panel configured to cover a side to enclose the interior region of the rack. The back of the equipment rack may also include at least one panel or a back door to provide access to the interior of the equipment rack from the back of the rack. In certain embodiments, the side and back panels, as well as the front door and the rear door, may be fabricated from perforated sheet metal, for example, to allow air to flow into and out of the interior region of the equipment rack. Otherwise, the panels may be fabricated from solid material.
The equipment racks are modular in construction and configured to be rolled into and out of position, e.g., within a row of the data center. Casters are secured to the bottom of each equipment rack to enable the rack to roll along the floor of the data center. Once positioned, leveling feet may be deployed to securely ground the equipment rack in place within the row. An example of casters and leveling feet employed on such an equipment rack is disclosed in detail in U.S. patent application Ser. No. 10/990,927.
Once in position, electronic equipment may be positioned in the interior region of the equipment rack. For example, the equipment may be placed on shelving secured within the interior region of the equipment rack. Cables providing electrical and data communication may be provided through the top of the equipment rack either through a cover (or “roof” as described in U.S. Pat. No. 6,967,283) at the top of the equipment rack having openings formed therein or through an open top of the equipment rack. In this embodiment, the cables may be strung along the roofs of the rack or be provided in the aforementioned cable distribution trough. In another embodiment, the cables may be disposed within a raised floor and connected to the electronic equipment through the bottom of the equipment rack. With both configurations, power and communication lines are provided to the equipment racks.
As discussed above, data centers are typically configured with rows of equipment racks arranged such that cool air is drawn into the racks from a cool aisle and warm or hot air is exhausted from the racks into a hot aisle. The equipment racks may be arranged in two rows with the fronts of the equipment racks in a near row being arranged in a forward direction and the backs of the equipment racks in a far row being arranged in a rearward direction. However, as stated above, in a typical data center, there are multiple rows of equipment racks wherein the rows may be arranged with the fronts of the equipment racks facing one another to define the cold aisle and with the backs of the equipment racks facing one another to define the hot aisle.
In order to address the heat build-up and hot spots within the data center, and to address climate control issues within the data center in general, a modular cooling unit is provided in one embodiment. As discussed above, due to cooling requirements within the data center or room, a plurality of cooling units may be provided. In one embodiment, the arrangement is such that there is a cooling unit for every two equipment racks provided in the data center. However, it should be understood that a person of ordinary skill in the art, given the benefit of this disclosure, may provide more or less cooling units within the data center based on environmental conditions of the data center. Further in some embodiments, the concentration and locations of cooling units may be adjusted based on the locations of the hottest racks in the data center, or based on information obtained and analyzed by a data center information management system. Additionally, cooling units of embodiments of the invention may be used in combination with other types of cooling systems, such as cooling systems of the types disclosed in U.S. patent application Ser. Nos. 11/335,874, 11/335,856 and 11/335,901, each entitled COOLING SYSTEM AND METHOD and filed on Feb. 10, 2006, which are owned by the assignee of the present invention and incorporated herein by reference.
A cooling unit of embodiments of the invention may be configured to have two sections or modules, which together define a closed loop cooling system that may be configured within a data center, equipment room or wiring closet to cool electrical components housed by equipment storage racks. In one embodiment, a lower module of the cooling unit includes a set of evaporator fans that are configured to draw heated air taken from a “hot” aisle, for example, which is generated by the electrical components. This heated air is drawn through an evaporator having a coil containing a coolant medium (e.g., a refrigerant) to cool the air. The arrangement is such that warm air drawn into the lower module by the evaporator fans flows over the evaporator to cool the air. The cooled air is forced back into the environment through a front of the cooling unit. Based on cooling requirements, other airflow patterns may be provided.
The heat absorbed by the coolant contained in the coil of the evaporator is transported to an upper module, which rests on the lower module. This upper module has a condenser adapted to cool the heated coolant delivered to the upper module from the lower module. In addition to the condenser, the upper module includes a set of condenser fans and a first, generally vertical plenum that is in fluid communication with an opening formed in a top of the upper module to draw relatively cooler air into the upper module. A second, generally horizontal plenum takes air directed by the first plenum to the set of condenser fans, which directs the air over the condenser. The relatively cooler air is heated as it flows over the condenser. Once heated, the air flows through one of two openings formed in the top of the upper module. The liquid coolant is directed back to the evaporator, where the cycle begins again. A compressor, such as a rotary compressor, pumps evaporated coolant from the evaporator to the condenser. The compressor, evaporator fans and condenser fans are all controlled by a controller. In one embodiment, to improve capacity control and efficiency, warmer coolant being delivered by the compressor unit to the condenser may be diverted to the evaporator by a bypass valve.
It should be understood that, in embodiments of the invention, the lower module may be configured to perform the condensing function described above and the lower module may be configured to perform the evaporating function described above.
In one embodiment, the controller is adapted to control the operation of the cooling system based on environmental parameters obtained by the controller. Generally speaking with prior cooling systems, the individual cooling units can not communicate with one another. For example, the controller may embody a plurality of controllers provided in the cooling units that communicate with one another over a controller area network (CAN) Bus. In other embodiments, a master controller may be provided to control the operation of the controllers of the cooling units. Each cooling unit may be provided with a display, which is operably coupled to the controller. The display is adapted to display the environmental conditions of the data room, such as, and not limited to, the temperature and the humidity of the data center at the cooling unit, the temperature of the air entering into and exiting out of the cooling unit, the temperature of coolant entering into and exiting out of the cooling unit, the flow rate of coolant entering the cooling unit, and the cooling capacity of the cooling unit. Suitable monitors and/or gauges may be provided to acquire such information. Alternatively, or in addition to the foregoing embodiment, the environmental conditions may be displayed on a unit provided with an integrated data center control and monitoring system.
Referring now toFIGS. 1-6, and more particularly toFIG. 1, there is generally indicated at10 a cooling unit of an embodiment of the invention. As shown, the coolingunit10 comprises alower module12 and anupper module14 configured to treat air within a room. In one embodiment, the coolingunit10 includes a frame orhousing16, which may be configured in two separate housings forming the frames of the lower and upper modules, or as a single, unitary housing. The coolingunit10 includes a front18, a back20,opposite sides22,24, a bottom26 and a top28. Eachside22,24 of the cooling unit may include at least one panel (not designated) configured to cover the side to enclose the interior region of the cooling unit. The front and the back of the cooling unit may include at least one removable panel or door to provide access to the interior of the cooling unit. In certain embodiments, the front and back panels may be fabricated from perforated sheet metal, for example, to allow air to flow into and out of the interior region of the cooling unit. Casters and leveling feet (both not shown) may be provided to enhance the mobility of the cooling unit and to set the cooling unit in a secure position. Other details of the coolingunit10 shown inFIG. 1 will be discussed in greater detail below as the description of the cooling unit proceeds.
Turning toFIG. 2, the internal components of the coolingunit10 are illustrated without showing thehousing16 and panels of the cooling unit. The components of the coolingunit10 are suitably secured to and contained within thehousing16 of the cooling unit in the manner shown and described herein. The air flow within the cooling unit will be discussed in greater detail below with reference to other drawings, includingFIG. 6 in particular. One purpose ofFIG. 2 is to illustrate the flow of a coolant medium (e.g., a liquid coolant, such as R134A and R410A coolants) through the working components of the cooling unit.
As shown, acompressor30 is provided for delivering hot gas coolant under pressure to the components of the coolingunit10. The pressurized coolant travels through adischarge pipe32, which connects thecompressor30 to acondenser34. A temperature sensor (not shown) and afirst pressure transducer36 may be provided adjacent to thecondenser34 to measure the temperature and the pressure of the coolant as it enters the condenser. The purpose of the temperature sensor and thepressure transducer36 will be discussed in greater detail below. Ahigh pressure switch38 may be further provided to de-energize the compressor thereby stopping the delivery of coolant to the condenser should the coolant experience an out of tolerance pressure condition that requires power to the compressor to be cut off. Thecondenser34 includes acoil40 having thermally conductive fins (not shown) configured to cool the heated coolant within the coil of the condenser. The air flow over thecondenser coil40 will be discussed in greater detail below with reference to drawings directed to the air flow configuration of the cooling unit10 (e.g.,FIG. 6). Once the coolant is cooled within the condenser34 (e.g., transitioning the coolant from an evaporated state to a condensed state), the coolant travels through anotherliquid pipe42 to anevaporator44. The coolant first travels through a filter drier46 to eliminate impurities and to remove unwanted non-condensables within the coolant. Once through the filter drier46, the coolant travels through athermal expansion valve48 to condition the coolant prior to entering theevaporator44.
Next, the low pressure coolant enters adistributor50 and is distributed to the evaporator by one of several (e.g., three) conduits, each indicated at52 inFIG. 2. As shown, one conduit52 delivers coolant to theevaporator44 near the top of the evaporator. A second conduit52 delivers coolant to a middle of theevaporator44. And finally, a third conduit52 delivers coolant to a bottom of theevaporator44. This configuration ensures that coolant is evenly distributed to theevaporator44, which is designed to include acoil54 in thermal communication with metal fins (not shown) so that heat may be absorbed from relatively warm air flowing over the evaporator. Once heated by warm air passing over theevaporator44, the evaporated coolant travels back to thecompressor30 via a section ofsuction piping56. However, prior to entering thecompressor30, the coolant passes through acompressor suction accumulator58, which ensures that coolant enters into thecompressor30 in an evaporated state. Anothertemperature sensor60 and anotherpressure transducer62 may be provided adjacent to thecompressor30, the purpose of which will be discussed in greater detail below. Acondensate pan35 may be disposed below theevaporator44 to collect condensate generated by the evaporator.
The arrangement is such that high temperature coolant flows from thecompressor30 to thecondenser34. Pressure and temperature readings of the coolant are taken prior to the coolant entering thecondenser34. Thecondenser34 cools the coolant by virtue of relatively cool air passing over thecondenser coil40. Once cooled, the coolant travels to theevaporator44. Abypass valve64 may be provided to divert coolant normally directed to thecondenser34 from thecompressor30 to theevaporator44 via adischarge pipe66. By opening thebypass valve64, a portion of coolant traveling to the condenser is diverted to the evaporator by way ofdistributor50. The operation of thebypass valve64, which may sometimes be referred to as a hot gas bypass valve, may be manipulated to regulate the capacity of the coolingunit10. As will be discussed in greater detail below, by closely monitoring the pressure and/or temperature of the coolant entering into thecondenser34, the efficiency of the coolingunit10 may be optimized by bypassing coolant that travels from the condenser to the evaporator. In one embodiment, the compressor may embody a rotary compressor, such as a 208-230/1/50 or 208-230/1/60 rotary compressor offered by Carrier of Syracuse, N.Y. When employing a rotary compressor, the pressure differential between the evaporating pressure and the condensing pressure, in certain embodiments, must be less than a predetermined pressure difference, such as 7.2 psig, to restart the compressor. To expedite the pressure equalization between evaporating and condensing pressures, the hotgas bypass valve64 may be open until the compressor re-starts.
Referring toFIGS. 3-6, and more particularly toFIG. 6, thelower module12 has a plurality of evaporator fans (sometimes referred to herein as air moving devices), each indicated at68, which are located at the front of the lower module of the coolingunit10. The arrangement is such that air may be drawn from the back of thelower module12 of the cooling unit over theevaporator44 through either an open back or through perforations in a back panel by theevaporator fans68. In one embodiment, there may be three such fans68 (as shown) to draw air through thelower module12 in the manner shown inFIG. 6. However, any number offans68 may be employed, depending on the size of the fans and on how much air is required to be drawn across theevaporator44. In one embodiment, theevaporator fans68 may be 200 mm mixed flow fans provided by EBM Industries of Farmington, CT. Theevaporator fans68 may be configured as part of a fan tray arrangement secured vertically to thefront18 of the coolingunit10, or be secured individually to thehousing16 at thefront18 of the cooling unit. As shown best inFIG. 6, the air drawn through the lower module by the evaporator fans68 (indicated by arrows A) flows over the coil and the fins of theevaporator44 to heat the coolant flowing through the coil. The resultant is that cool air is blown out of theevaporator fans68 at the front of the coolingunit10 to cool the space adjacent the front of the cooling unit.
In one embodiment, one ormore cooling units10 may be positioned so that thebacks20 of the cooling units are adjacent a hot aisle. Depending on the cooling requirements of the data center, more than one coolingunit10 may be provided to cool warm air deposited in the hot aisle by the equipment enclosures.
In a particular configuration, theupper module14 of the coolingunit10 may be configured to include the top28 of the cooling unit, which has three openings formed therein. Specifically, there may be provided anintake opening70 and twoexhaust openings72,74 in the top28 of the coolingunit10. As best shown inFIG. 6, aninterior wall76 and a blockingpanel78 define afirst plenum80 that extends along a generally vertical axis. Thefirst plenum80 is in fluid communication with theintake opening70 to draw air from the intake opening to asecond plenum82 located along a generally horizontal axis at the bottom of theupper module14.
Further provided in theupper module14 is atray84 that is releasably secured to the housing of the coolingunit10, the tray having three condenser fans, each indicated at86, secured thereto. The arrangement is such that thetray84 and thecondenser fans86 are disposed along a generally horizontal plane to define an upper wall of thesecond plenum82. Thecondenser fans86 are configured to draw relatively cool air from thefirst plenum80 to thesecond plenum82 and blow the air across thecondenser34 so as to cool coolant running through the condenser coil. Air flows through thecondenser34 and out of the twoexhaust openings72,74 formed in the top28 of theupper module14 of the coolingunit10. The airflow path through theupper module14 is depicted by arrows B inFIG. 6. In one embodiment, theexhaust openings72,74 may be in fluid communication with exhaust ducts (not shown) to transfer the warm air out of the data center or room. In another embodiment, the air may be directed to the top of the data center or room, away from the equipment enclosures. In a further embodiment one exhaust opening or more than two exhaust openings may be provided. It should be understood that one skilled in the art, given the benefit of this disclosure, may configure the cooling unit in any desired manner consistent with the teachings herein.
As shown best inFIG. 6, thecondenser fans86 draw air from theintake opening70 along a first flow path defined by the first andsecond plenums80,82 to theexhaust openings72,74, as indicated by arrows B. As shown in dashed lines, air may also be drawn from a dropped ceiling arrangement. Thecondenser34 is positioned within theupper module14 at an acute angle with respect to the first flow path. Theevaporator fans68 draw air from the intake opening defined by theopen end20 of thelower module12 of the coolingunit10 along a second flow path shown by arrows A to an exhaust opening defined by the oppositeopen end18 of the lower module of the cooling unit. Theevaporator44 is positioned within thelower module12 at an acute angle with respect to the second flow path. In the claims, theupper module14 may be referred to as a first module and thelower module12 may be referred to as a second module. The acute angles of thecondenser34 and theevaporator44 may be selected to maximize the surface areas of the condenser and the evaporator, respectively.
In one embodiment, a controller may be operably coupled to a display unit88 (seeFIG. 1), such as a display unit shown and disclosed in U.S. patent application Ser. Nos. 11/335,874, 11/335,856 and 11/335,901 discussed above. In a certain embodiment, thedisplay unit88 has a liquid crystal display, for example, to display certain environmental conditions, such as temperature and humidity of the data center, the temperature of air entering into and exiting out of the cooling unit, the temperature of coolant entering into and exiting out of the evaporator and condenser of the cooling unit, and the flow rate of coolant within the cooling unit. A plurality of control buttons and status indicators are further provided on thedisplay unit88 to enable the operator to manipulate the operation of the cooling system and to quickly determine the status of a certain condition, respectively. As shown, thedisplay unit88 may be secured to thefront18 of the coolingunit10 within an opening formed in the front of the cooling unit by means of a sealing gasket and a mounting bracket in which screw fasteners may be provided to secure the display assembly to the front panel within the opening.
FIGS. 3-5 illustrate the removal of thetray84 having thecondenser fans86. Specifically,FIG. 3 illustrates the front panel or door of the coolingunit10 removed from the cooling unit. As shown, thedisplay unit88 is secured to thehousing16 of the cooling unit. In a configuration having only one panel, the panel and thedisplay unit88 may be removed either together or separately, depending on the particular design. In the shown embodiment, at least two air blocking panels,90,92 are secured to thehousing16 of the coolingunit10 to contain air within theupper module14 of the cooling unit.FIG. 4 illustrates the lowerair blocking panel90 removed so that thesecond plenum82 of the cooling unit is revealed. Once the lowerair blocking panel90 is removed, thefan tray84 may be removed from thehousing16 of the coolingunit10 by simply pulling the fan tray from the front of the coolingunit10. Thefan tray84 is configured to rest on a pair of support rails94,96, which are secured to thehousing16 within theupper module14 of the coolingunit10. Thecondenser fans86, as well as theevaporator fans68, may be variable speed fans that are independently operable under the control of a controller. The arrangement is such that thefan tray84 may be easily removed from the coolingunit10 to replace or repair a fan, for example.
In other embodiments, as described above, thepanels90 and92 may be combined to create a single panel. With this configuration, thedisplay unit88 must be separately removed so as to access thefan tray84.
As mentioned above, a controller may be configured to control the operation of the coolingunit10 as well as provide communication with external devices. In one embodiment, the controller may be a separately dedicated unit that controls the operation of multiple coolingunits10. In another embodiment, the controller may be provided in one of the coolingunits10, with the cooling unit having the controller functioning as the main cooling unit and the other cooling units functioning as subservient cooling units. In yet another embodiment, the operation of the coolingunit10 may be operated under the control of an integrated data center control and monitoring system with each cooling unit having a controller unit that communicates with the other cooling units over the network. In one such embodiment, the controller may communicate with a data center control system to provide status of the components of the cooling system and to receive control commands for the data center control system. In one particular embodiment each coolingunit10 includes a controller that communicates with the data center controller over a network, such as a CAN Bus network, and in one such embodiment, the data center controller may be implemented using the integrated data center control and monitoring system, such as the InfraStruXure™ data center manager sold by American Power Conversion Corporation of West Kingston, R.I., the assignee of the present invention. Notwithstanding the particular configuration, the controller is adapted to control the flow of coolant from thecompressor30 to thecondenser34 and theevaporator44 depending on the temperature and pressure readings of the cooling unit.
FIG. 7 illustrates a system block diagram of the coolingunit10, showing the major interfaces between the cooling unit and potential external devices. As shown, aheat load98 is applied to thecooling unit10 in which atemperature sensor100 detects and transmits a signal to an embeddedcontroller102 of the cooling unit. In one embodiment, the embeddedcontroller102 may be a Philips XAG49 microprocessor (running at 16 MHz, 512 Kbytes of flash memory, 128 Kbytes of battery backed static RAM, 16 Kbytes of EEPROM, and having a real-time clock). As shown, the embeddedcontroller102 may communicate with anetwork manager104 by means of a CAN, for example. Thenetwork manager104 may communicate with thedisplay unit88, abuilding management system106, if provided, adata center manager108 by means of a local area network110 (LAN), for example, or alocal test port112. In a certain embodiment, the network manager may employ a network management card containing ASIC, 4 Mbytes of static RAM, 16 Kbytes of EEPROM, a real time clock and a CAN controller. In one embodiment, the ASIC includes an Intel 186 microprocessor, running at 50 MHz, and a 10/100 Base-T network interface controller (NIC).
During operation, the coolingunit10 may be configured to function between several states, including, but not limited to, a main operational state, a condensate state and a hot gas bypass valve state. In the main operational state, which is illustrated inFIG. 8, the operation of the cooling unit proceeds as follows: (a) un-powered; (b) start-up delay; (c) off/standby; (d) idle operation; (e) failed operation; (f) warm-up operation; (g) running operation; (h) firmware download; and (i) test. Specifically, once power is provided, the cooling unit operationally moves from an un-powered condition at114 to an initial powered condition at116 in which initialized state variables are set. Once initialized, the cooling unit moves to a start-up delay condition at118 in which no action is taken. After a predetermined time period (as determined by a delay timer, for example) and the synchronization of a stepper motor, the cooling unit transitions to an off/standby condition at120 in which the hot gas bypass valve is fully opened. In this condition, the temperature and discharge pressure threshold alarms are disabled (except when employing a rack containment system, wherein only the temperature threshold alarms are disabled) and the fan speeds (evaporator and condenser) are set to idle (except when employing a rack containment system or when the machine is in proportional spot configuration mode wherein the fans are kept running at a minimum speed). In the off/standby mode120, the coolingunit10 is ready for operation.
As shown, the mode of operation may transition from either the off/standby condition120 or anidle operation mode122 to a pre-run operation mode at124. The transition occurs if all of the following conditions are met: (1) the delay timer is not running; (2) the device has been commanded upon by the controller; (3) the suction and discharge pressures are equalized; (4) there is no idle requested due to a leak; and (5) the cooling unit inlet temperature, when employing in-row or air containment configurations, or return air temperature, when employing a spot cooling configuration, exceeds a predetermined cool set point and a dead band (i.e., a tolerance that prevents the unwanted transition to back to off/standby or idle modes). The transition topre-run operation mode124 may also occur when the forgoing transition does not occur, and the device has been commanded upon the controller. When inidle operation mode122, the transition may also occur when (1) the delay timer is not running, (2) the suction pressure is above a predetermined threshold, e.g., 92 psig, (3) the condensate pan is not full, (4) there is no idle requested due to a leak, (5) the line pressure has equalized, and (6) the cooling unit inlet temperature (for in-row or containment configurations) or return air temperature (for spot cooling configurations) exceeds the predetermined cool set point and dead band.
Duringpre-run mode124, the hot gas bypass valve is fully closed to clear any events that are no longer active. The temperature and discharge pressure threshold alarms are enabled and the evaporator and condenser fans are operated at full (maximum) speed. A delay timer is set for a predetermined time period, e.g., twenty seconds. When warming up at126, the cooling unit is providing environmental control functionality in which the compressor is running. In this state, the evaporator and condenser fans are run at full (maximum) speed and the bypass valve is closed to allow the system to warm up and stabilize prior to attempting to control the system. Once warmed up, the cooling unit may be operated at128 to provide the cooling operation described above. If failure occurs, which is indicated at130 inFIG. 8, at either at the pre-run124, warm-up126 or running128 modes, thepre-run routine124 may be started again. The cooling unit may be further configured to conduct firmware download operations at132 and manufacturing testing at134, either during operation, or while powering up.
Transition toidle mode122 may occur when upon one of the following events: (1) the condensate pan is full; (2) if there is an idle requested due to leak; (3) when employing a spot cooling configuration, the return air temperature is less than or equal to the cool set point; (4) when employing an in-row or containment system configurations, the cooling unit inlet temperature is below a cool set point, e.g., 90° F.-sec; (5) if high head pressure input is asserted (and not the third such event in thirty minutes); or (6) suction pressure is below a predetermined threshold, e.g., 92 psig (and not the third such event in thirty minutes). Transition from either warm-upmode126 or runningmode128 tofailure mode130 may occur when the cooling unit is commanded upon by the controller and an analog sensor has failed, or there were three high head pressure events in thirty minutes, for example, or there were three low suction pressure events in thirty minutes, for example.
In one embodiment, transition fromfailure mode130 topre-run mode124 may occur when all of the following conditions are met: (1) the cooling unit is commanded upon by the controller; (2) the delay timer is not running; (3) the condensate pan is not full; (4) no analog sensors have failed; (5) the three high head pressure in thirty minutes event is cleared, for example; and (6) the three low suction pressure in thirty minutes event is clear, for example. Transition back to off/standby mode120 may occur when the unit is commanded upon by the controller. Transition from warm-upmode126 toidle mode122 may occur upon one of the following events: (1) if the condensate pan is full; (2) there is an idle requested due to leak; (3) when in discrete mode, the return air temperature is less than or equal to the cool set point, or, when in proportional mode, the rack inlet temperature is below the cool set point for 90° F.-sec, for example; or (4) if the high head pressure input is asserted and it is not the third such event in thirty minutes.
Referring toFIGS. 9 and 10, the condensate state mode is illustrated inFIG. 9 and the hot gas bypass valve state mode is illustrated inFIG. 10. With particular reference toFIG. 9, when referencing the condensate state, the cooling unit transitions from an un-powered condition at140 to a powered condition at142. Once powered on, a normal condensate level is indicated at144. When the condensate pan becomes full, the condition of the cooling unit transitions from thenormal mode144 to afull mode146. When infull mode146, a signal may be transmitted to the controller or some other visual or audible alarm to discharge the condensate pan. When a predetermined period of time expires, e.g., ten minutes, the state of the cooling unit transitions to a full timeout mode at148. All other failures are indicated at150 inFIG. 9. For example, failmode150 may be triggered when a sensor, e.g., a float sensor, fails to deliver a signal to the controller. As withmode148, upon the expiration of a predetermined period of time, e.g., ten minutes, the state of the cooling unit transitions to a full timeout mode at152. The failures indicated at146,148,150 and152 may be cured in which the cooling unit returns to its normal state at144.
With reference toFIG. 10, when referencing the hot gas bypass valve state, the cooling unit transitions from an un-powered condition at160 to a powered condition at162. Once powered on, a command is issued (e.g., by the controller or network manager) to the bypass valve to a synchronized closed position at164 in which a delay timer is set. Once synchronized, indicated at166 inFIG. 10, the cooling unit transitions to a post-synchronized mode at168 in which the bypass valve is reset to a position where it was before synchronization, if commanded by the controller or network manager. Upon a predetermined delay as determined by the delay timer, for example, the mode transitions from168 to a ready mode at170 in which the bypass valve position is recorded and re-commanded to synchronize. From this position, the cooling unit transitions back tomode166 in which the bypass valve receives a command to synchronize. This command may include how long to keep the bypass valve closed and whether to return the valve to the previous position after synchronization.
While operating, the coolingunit10 is configured to provide the system operator with the ability to set the operating environment to one of the following: spot cooling within a space; hot aisle cooling within a row of equipment racks; or rack air containment cooling, which is described in U.S. patent application Ser. Nos. 11/335,874, 11/335,856 and 11/335,901 discussed above. When configured for spot cooling, the controller may be selected to one of the following control methods: proportional control, which modulates the bypass valve and fan speeds at certain set points and dead bands; or discrete control (e.g., energy saving), which closes the bypass valve and the compressor runs when the return or remote air temperature at the cooling unit exceeds a certain threshold. When employing a remote air sensor in in-row configurations, the sensor is located adjacent the front of the rack. The compressor is turned off when the temperature at the cooling unit is less than or equal to the threshold temperature. When configured for hot aisle and cold aisle environments, the proportional control method may be employed. Similarly, when configured for rack air containment, the proportional control method may be used. When in operation, the cooling unit may be configured to protect and maximize the life of the compressor.
The controller and/or system manager of the coolingunit10 may be configured to monitor and control other aspects of the cooling unit. For example, the controller and/or system manager may be configured to monitor and control power management, event management, user interfaces, testing requirements, condensate monitoring, leak monitoring, run hour counters, maintenance alarms, fault alarms and user inputs and outputs.
Specifically, with certain prior art cooling systems, the compressor has a tendency to cycle on and off during operation, thereby creating a situation in which hot air flows from the hot aisle to the cold aisle since the evaporator fans are operating when the compressor is turned off. As discussed above, in order to address unwanted cycling of the compressor, the coolingunit10 includes thebypass valve64 andevaporator fans68. In a certain embodiment, thebypass valve64 is an electronic expansion valve offered by Sporlan Division of Parker-Hannifin Corporation of Washington, MO. The temperature of air entering into thelower module12 may be monitored and used as a set temperature. Cooling capacity control may be achieved in one of three modes. They are: (1) spot-proportional mode (e.g., in a data closet); (2) in-row mode (e.g., in hot aisle/cold aisle applications within a data center); and (3) rack air containment systems (RACS; e.g., in enclosed systems having at least one equipment rack and at least one cooling unit).
In circumstances where the cooling load is minimal, the temperature of the air entering the coolingunit10 may drop so that the temperature approximates the unit's set temperature. In circumstances where the air temperature entering into thelower module12 approaches the set temperature, the cooling unit's cooling capacity is reduced by bypassing hot gas back to theevaporator44 with thebypass valve64 viaconduit66. There are two methods used to reduce (or regulate) the cooling unit's cooling capacity. They are: (1) use of the hot gas bypass valve; and (2) use of the variable evaporator fan speed and the hot gas bypass valve together.
In another embodiment, by reducing the speed of the evaporator fans, the capacity of the coolingunit10 is reduced. This results in the reduction of supply air temperatures, with the bypass valve regulating supply air temperature in proportional mode. Specifically, in a particular embodiment, the default supply air set point in spot-proportional modes may be 57° F. In another particular embodiment, the default supply air set point in in-row and containment air system mode may be 68° F. As discussed above, the temperature sensor e.g.,sensor60, installed within the system adjacent the input of thecompressor30 may monitor the return gas temperature of coolant entering into the compressor. To keep the return coolant less than a predetermined temperature (e.g., 68° F.) so as to ensure the protection of thecompressor30, thebypass valve64 may be throttled and the evaporator fan speed may be reduced, even when there is a demand for coolant bypass.
Under certain circumstances, while thebypass valve64 bypasses hot gas coolant, coolant velocities within the pipes are reduced. This reduction of coolant velocities may limit the ability of compressor oil contained within the coolant from returning to thecompressor30. As is well known, a predetermined amount of oil may be provided in the coolant to protect the working components of thecompressor30. In a certain embodiment, thebypass valve64 may be configured to close for a short period of time, e.g., approximately ten seconds, every twenty minutes or so when the bypass valve is operating in an open position. By closing thebypass valve64 periodically for a short duration, relatively high coolant velocities may be achieved within the pipes so that any oil trapped within the pipes is forced back to thecompressor30.
During operation, low evaporating temperatures may be encountered in systems in which the filter drier46 or an air filter become clogged, thus reducing the cooling capacity of the coolingunit10. Low evaporating temperatures may reach a critical condition and cause damage to thecompressor30 by virtue of liquid coolant entering into the compressor. In one embodiment, to alleviate this potential concern, theevaporator fans68 of the coolingunit10, as discussed above, may embody multiple (e.g., three) variable speed, DC fans that work with thepressure transducer62 located adjacent theaccumulator58. Thepressure transducer62 is configured to obtain an evaporating pressure reading of the coolant entering into thecompressor30. In one example, when the evaporating temperature drops below a predetermined temperature, e.g., 40° F., for example, as detected by thepressure transducer62, a saturation temperature of the coolant is determined from a look up table embedded in the firmware of the controller. Thetemperature sensor60 measures return gas temperature flowing to thecompressor30. The coolingunit10 will increase evaporator air flow rate by increasing the speed of theevaporator fans68 via a program control loop (e.g., a PID control loop) written in the firmware of the controller. If increasing the evaporator air flow rate (by increasing evaporator fan speed) does not result in the increase of the evaporating coolant temperature, and the temperature drops below a second predetermined temperature, 37° F., for example, thebypass valve64 will bypass coolant hot gas to increase the evaporating temperature. The controller may be configured to trigger an alarm if the evaporating temperature drops below a third predetermined temperature, e.g., 34° F.
Conversely, the coolingunit10 of embodiments of the invention may also be configured to address the issue of high return coolant gas temperatures entering into thecompressor30 as a result of high heat loads in the data center or equipment room. Such high temperatures may reduce the life of thecompressor30 or cause compressor damage if prolonged. In one embodiment, the coolingunit10 may be configured so that thetemperature sensor60 adjacent thecompressor30 detects coolant temperature entering the compressor. When the coolant gas temperature reaches a predetermined temperature, e.g., 68° F., the controller may reduce the evaporator air flow by decreasing the speed of theevaporator fan speed68 via a control loop (e.g., a PID loop) written in the firmware of the controller.
In certain environments, it may be difficult to control the air temperature entering the condenser through theintake opening70 when the data center or equipment room includes a dropped ceiling or when hoses or ducts are used to move air to ambient since there is no temperature control within a dropped ceiling configuration. This restriction may result in low air temperatures being exerted on thecondenser34 and thus low coolant temperatures being generated by the condenser. In certain conditions, extremely low condensing temperatures may cause the coolant to flood back to thecompressor30 in liquid condition and damage the compressor. In one embodiment, thecondenser fans86 may embody three variable speed DC condenser fans that operate in cooperation with thepressure transducer36 adjacent thecondenser34. In other embodiments, the condenser fans may be AC with VFD or SC PSC with a sine wave chopper. The condensing temperature may be calculated from readings taken from thepressure transducer36 or from the temperature sensor. The controller may be configured to maintain the temperature above a predetermined temperature, e.g., 95° F., for example, by a control loop (e.g., a PID loop) written in the firmware of the controller. In another embodiment, the discharge pressure of coolant discharged by thecondenser34 throughdischarge pipe42 may be monitored to control thecondenser fans86 speed. Specifically, a discharge pressure set point, e.g., 420 psig, may be achieved by regulating the fan speed by a control loop (e.g., a PID loop).
Conversely, when the air temperature in the dropped ceiling (or in situation in which a flexible hose supplies air to the condenser viaintake opening70, which may reduce the flow rate of air to the condenser) is higher than a predetermined temperature, e.g., above 100° F., for example, a resulting elevated condensing pressure may occur. This may result in a highpressure cutout switch38 cutting off power to thecompressor30, thus causing total loss of cooling. In one embodiment, the condensing pressure may be measured by thepressure transducer36. For example, the coolingunit10 may be configured so that the highpressure cutout switch38 cuts off the power to thecompressor30 when the condensing pressure reaches a predetermined pressure, e.g., 550 psig. In one method (in spot-discrete mode), if the condensing pressure reaches a predetermined pressure, e.g., 520 psig, thebypass valve64 opens and bypasses the coolant to theevaporator44 viaconduit66. The reduced coolant mass flow rate to thecondenser34 reduces the heat rejection and the condensing pressure is prevented from reaching the predetermined cutoff pressure. While a slight pressure drop in the system may result, the coolingunit10 is still operating to provide the requisite cooling. In another method (in spot-proportional, in-row and rack air containment applications), if the condensing pressure reaches a predetermined pressure, e.g., 520 psig, the speed of theevaporator fans68 is reduced to reduce the cooling capacity and heat rejection in the condenser which will assist in reducing the discharge pressure. Thebypass valve64 may also be manipulated to increase or decrease the discharge pressure. Reducing evaporator air flow will reduce the supply air temperature, which is regulated by the bypass valve.
As discussed above, in certain embodiments, the following control strategies may be employed when in running mode, e.g.,mode128 inFIG. 8. When controlling the bypass valve in discrete capacity control mode, the bypass valve is normally held closed. If the discharge pressure exceeds a certain threshold, e.g., 520 psig, the bypass valve is opened linearly to a maximum of fifty percent at an elevated predetermined pressure, e.g., 550 psig. With reference toFIG. 11, when controlling the bypass valve in proportional capacity control mode, the bypass valve is regulated by a PID controller, for example, to (1) maintain the evaporating temperature within a safe range, e.g., between 38° F. and 56° F., and (2) maintain the supply air temperature at the predetermined, user-configurable supply air set point, e.g., between 52° F. and 73° F.
In certain other embodiments, and with reference toFIG. 12, in all modes (discrete capacity and proportional capacity modes), the condenser fans speed may be regulated by a PID controller to maintain a predetermined discharge pressure, e.g., 425 psig. Similarly, with reference toFIG. 13, the evaporator fans speed, in discrete capacity mode, the evaporator fans are normally run at a constant speed. The user may specify the evaporator fans speed by setting the preferences within the controller user interface, e.g., the display unit. Specifically, in one embodiment, the user may select one of five fan speeds when in spot-discrete mode, e.g., high, medium-high, medium, medium-low and low. In other embodiments, the fan may be configured to operate in any number of fan speeds. In addition, if the suction discharge temperature exceeds a predetermined limit, e.g., 68° F., the evaporator fans speed may be lowered linearly to a minimum of fifty percent of their normal speed at a predetermined temperature, e.g., 75° F., for example. In proportional capacity control mode, which may be employed in spot cooling or in-row configurations, the evaporator fans speed may be regulated by a PID controller to maintain the cooling rack inlet temperature at a predetermined user-configurable set point. Additionally, the evaporator fans speed may be scaled down to alleviate certain abnormal conditions. Specifically, the evaporator fans speed may be post-scaled under the following conditions. When detecting high head pressure, the evaporator fans speed may be scaled one hundred percent up to a predetermined discharge pressure, e.g., 520 psig, down to a minimum of sixty-five percent at an elevated predetermined temperature, e.g., 550 psig, for example. The rate of the scaling may be configured to be linear. When detecting high suction temperature, the evaporator fans speed may be scaled one hundred percent up to a predetermined suction temperature, e.g., 68° F., down to a minimum of fifty percent at an elevated predetermined temperature, e.g., 75° F., for example. As with high head pressure detection, the rate of the scaling factor may be configured to be linear. Should high head pressure and high suction temperature abnormalities be detected simultaneously, the numerically lowest scaling factor may be used to scale the evaporator fans speed.
In a particular embodiment having a proportional capacity control configuration or in an air containment configuration, the evaporator fans speed may be controlled by employing the following equation:
TR=((Qactual*3415)/(CFMactual*1.08))+TS (1)
where
- TR—theoretical return air temperature;
- Qactual—actual power output in Watts;
- CFMactual—airflow through the evaporator in cubic feet per minute;
- TS—supply air temperature in ° F.;
- 3415—converts kW to BTU per hour; and
- 1.08—power constant.
Next, the theoretical air return temperature may be averaged with the actual air return temperature by using the following equation:
TR=TR+TRactual (2)
Next, the power demand for the air containment system is calculated by using the following equation:
QDMD=(ΔT*CFMactual*1.08)/3415 (3)
where
- QDMD—power demand;
- ΔT=TR−Tsetp;
- Tsetp—supply air set point in ° F.; and
- CFMactual—airflow through the evaporator in cubic feet per minute.
Next, the air containment system airflow demand is calculated by employing the following equation:
CFMDMD=(QDMD*3415)/(ΔTset*1.08) (4)
where
- CFMDMD—air containment system airflow power demand; and
- ΔTset—desired temperature delta across the cooling unit.
And finally, the desired evaporator fans speed may be calculated by employing the following equation:
Speedpercent=(CFMDMD/1200)*100 (5)
where
- Speedpercent—evaporator fans speed; and
- 1200—maximum airflow in cubic feet per minute.
Based on the foregoing manipulation of the condenser fans and the evaporator fans, the coolingunit10 may be configured to optimize its cooling capacity. Specifically, cooling capacity may be calculated by the following equation:
The controller and the components of the coolingunit10 may be configured, by employing a polynomial equation having coefficients unique to thecompressor30, coolant evaporating temperature and coolant condensing temperature. Specifically, for each coolingunit10 used in the data center or equipment room, thecompressor30 has a set of coefficients to calculate cooling output, mass flow rate, energy efficiency rating, and current draw. Each set may consist of up to ten coefficients, and as a result, the compressor cooling capacity may be calculated by firmware provided in the controller. The coolant pressures are measured by the pressure transducers, and evaporating and condensing temperatures may be calculated from the coolant evaporating and condensing pressures and/or by temperature sensors.
To determine cooling capacity based on the fundamental equation (6), one of four equations may be used. In one embodiment, the cooling capacity may be determined by employing the following equation:
Pc=(Qcomp−Qcomp loss−1052.6*CR*0.2928104−Pf)/1000 (7)
where
- Pc—net sensible cooling capacity;
- Qcomp—compressor performance;
- Qcomp loss—compressor heat loss;
- 1052.6—amount of energy required to condense one pound of water;
- CR—condensate production rate;
- 0.2928104—converts BTU/hour to Watts; and
- Pf—fan power.
Equation (7) relies on industry-standard ARI-540 compressor cooling coefficients. Specifically, and with reference toFIG. 14, when employing equation (7), the coolant pressure at the pressure transducers is measured in psig at200 and manipulated to calculate the absolute coolant suction and discharge pressures (by adding 14.7 psi to the gauge pressures) at202. The pressure measurements are next converted from I-P units to SI units at204. At206, the coolant evaporating and condensing temperatures may be calculated. At208, the ARI-540 polynomial equation may be calculated as follows:
X=C1+C2*(S)+C3*D+C4*(S2)+C5*(S*D)+C6*(D2)+C7
*(S3)+C8*(D*S2)+C9*(S*D2)+C10*(D3) (8)
Where- X—can be (1) compressor capacity, (2) coolant mass flow rate, (3) compressor input power or current, (4) or energy efficiency ratio (EER);
- C—equation coefficient representing compressor performance;
- S—suction dew point temperature in ° C.; and
- D—discharge dew point temperature in ° C.
Next, at210, the compressor performance (Qcomp) may be calculated by inserting selected compressor performance coefficients (in SI units) to the ARI-540 polynomial equation (8). At212, the compressor power input may be calculated by inserting selected power coefficients (in SI units) to the ARI-540 polynomial equation (8). Based on these calculations, the thermal heat rejection component may be calculated at214 in Watts by adding Qcompand Pcomp. And finally, at216, the compressor heat loss may be calculated. Based on the results obtained above, the cooling capacity of the cooling unit may be determine. It should be noted that the compressor coefficients may be provided in both metric (SI) and English units. When taking Celsius (° C.) temperature readings for evaporating and condensing temperatures, the compressor coefficients are represented in metric units. When taking Fahrenheit (° F.) readings for evaporating and condensing temperatures, the compressor coefficients are represented in English units. It should further be noted that if the compressor is not in the evaporator air stream, the Qcomp losscomponent of equation (7) is omitted.
ExampleUsing polynomial equation (7), which employs ARI-540 polynomial equation (8), the capacity of a cooling unit may be calculated as follows. Equation (8) is employed, assuming the following coefficients for a cooling unit using a 60 Hz compressor:
| |
| | Compressor Cooling |
| Coefficient | Capacity |
| |
|
| C1 | 2.206E+04 |
| C2 | 3.403E+02 |
| C3 | −2.265E+02 |
| C4 | 4.067E+00 |
| C5 | −8.068E−01 |
| C6 | 1.352E+00 |
| C7 | 1.309E−02 |
| C8 | −1.900E−02 |
| C9 | −2.813E−03 |
| C10 | −3.881E−03 |
| |
The following estimated values for fan power, condensate production rate and compressor heat loss may be assumed:
Pf—300 Watts;
CR—1.6 pounds/hour; and
Qcomp loss—150 Watts.
And finally, the following test measurements may be employed for determining the coolant suction and discharge dew point temperatures:
Evaporating Pressure—136 psig;
Discharge Pressure—438 psig;
Suction Dew Point Temperature—47.1° F.; and
Discharge Dew Point Temperature—123.9° F.
Based on the foregoing, cooling capacity is calculated as follows:
Qcomp=6393 Watts
Pc=(6393−150−1052.6*1.6*0.2928104−300)/1000
Pc=5.45 kW
In another embodiment, the cooling capacity of the cooling unit may be determined by employing the following equation:
Pc=((SCFM*0.075*60)*Cp*DTair/3.415−Qcomp loss−Pf)/1000 (9)
where
- Pc—net sensible cooling capacity;
- 1000—converts Watts to Kilowatts;
- Cp—specific heat of air in BTU/lb-° F.;
- Qcomp loss—compressor heat loss;
- DTair—supply and return air temperature difference;
- SCFM—estimated standard volume flow rate at given fan speeds
- 0.075—density of standard air in lb/ft3; and
- Pf—fan power.
With reference toFIG. 15, at220, the return and supply air temperatures of the cooling unit are measured. Next, at222, the average return and supply temperatures are determined. At224, the temperature differential (DTair) is calculated based on the average return and supply temperatures. At226, the evaporator coil standard air flow rate at a given speed is calculated to determine SCFM, assuming a specific heat Cpat 0.243 BTU/lb-° F. The mass flow rate may be calculated at228, followed by the calculation of sensible cooling capacity at230 and232.
In yet another embodiment, cooling capacity of the cooling unit may be determined by using the following equations:
Qthr=(SCFM*0.075*60)*Cp*DTair/3.415 (10)
Qcomp=Othr−Pcomp (11)
Pc=(Qcomp−1052.6CR*0.2928104−Ccomp loss−Pf)/1000 (12)
where
- Qthr—heat rejection at condenser coil;
- Pc—net cooling capacity;
- Qcomp—compressor performance;
- Qcomp loss—compressor heat loss;
- 1052.6—amount of energy required to condense one pound of water;
- 0.2928104—converts BTU/hour to Watts;
- Pf—fan power;
- CR—condensate production;
- Cp—specific heat of air;
- DTair—condenser entering and leaving air temperature difference;
- Pcomp—compressor power consumption;
- 1000—converts Watts to Kilowatts;
- SCFM—estimated standard volumetric flow rate at given fan speeds; and
- 0.075—density of standard air.
Referring toFIG. 16, the method of calculating cooling capacity with equations (10)-(12) is as follows. At240, the temperature of air entering into and exiting out of the condenser is measured. At242, the temperature differential between the air entering into and exiting out of the condenser is calculated to arrive at DTair. Next, at244, the condenser coil standard flow rate (SCFM) is estimated, assuming a specific heat (Cp) of 0.243 BTU/lb-ft. At246, the mass flow rate (in lb/hr) is calculated. At248, the condenser heat rejection is calculated and converted to watts by dividing the result by 3.415. The compressor power input is calculated by inserting selected compressor power input coefficients (in SI units) at250 to the ARI-540 polynomial equation (8) at252. At254 and256, the compressor cooling output and the cooling capacity is calculated, respectively. With respect to the cooling capacity, the latent capacity, fan heat and compressor heat loss are subtracted from the compressor cooling output to determine the sensible cooling capacity.
And finally, in another embodiment, the cooling capacity may be determined by utilizing the following equations:
Pc=(Qtotal−Qcomp loss−1052.6*CR*0.2928104−Pf)/1000 (13)
Qtotal=M*(hsuction gas−hliquid) (14)
where
- Pc—net sensible cooling capacity;
- Qtotal—total cooling capacity;
- Qcomp loss—compressor heat loss;
- 1052.6—amount of energy required to condense one pound of water;
- CR—condensate production;
- 0.2928104—converts BTU/hr to Watts;
- Pf—fan power;
- 1000—converts Watts to Kilowatts;
- Qtotal—total cooling capacity;
- M—coolant mass flow rate;
- hsuction gas—enthalpy of coolant at evaporator coil outlet; and
- hliquid—enthalpy of coolant at thermostatic expansion valve inlet.
Turning toFIG. 17, at260, coolant gauge suction and discharge pressures are measured at the suction and discharge pipes. At262, absolute coolant suction pressure (otherwise referred to as evaporating pressure) and discharge pressure (otherwise referred to as condensing pressure) are calculated by adding 14.7 psi to the gauge pressures obtained in260. At264, the units are converted to SI units. Next, at266, dew point evaporating and condensing temperatures are calculated. At268, the ARI-540 polynomial equation (3) is employed. At270, the compressor coolant mass flow rate is calculated by inserting selected compressor performance coefficients. At272, enthalpies of the coolant at the thermostatic expansion valve inlet and evaporator coil outlet are calculated by using coolant pressures and temperatures. And finally, at274 and276, the total cooling capacity and the net cooling capacity may be determined.
In a certain embodiment, the cooling unit may be configured to provide uninterruptible cooling. Specifically, it has been discovered that by bypassing coolant from high pressure hot gas side to low temperature to the pressure suction side before the discharge pressures reach the predetermined cutoff pressure, the coolant flow rate to the condenser is reduced to reduce the discharge pressure of coolant exiting the condenser. In certain applications in which air temperature entering the condenser coil is relatively high, e.g., 100° F. or higher, the blockage of air flow into the condenser may occur. This may result in very high condensing pressures that require the high pressure cutoff switch to activate thereby cutting off power to the compressor. Obviously, the exposure to elevated temperatures may result in damage to the electronic equipment housed in the equipment storage racks.
In one embodiment, the pressure transducer at the discharge pipe may be employed to measure pressure of coolant entering the condenser. As discussed above, the cooling unit may be provided with a pressure cutoff switch, such asswitch38, which cuts off the power when the condensing pressure reaches a predetermined pressure, e.g., 550 psig. If, for example, the condensing pressure reaches a threshold predetermined pressure, e.g., 525 psig, the cooling unit may be configured to have the bypass valve open to allow a portion of the high pressure coolant to return back to the condenser. The coolant mass flow rate to the condenser coil is reduced with the heat rejection and the condenser pressure is limited so that the pressure within the condenser is below 550 psig. As a result, there may be a drop in cooling capacity, but total loss of cooling may be prevented since the compressor remains in operation.
Thus, it should be observed that by controlling the speed of the condenser fans, the condensing pressure may be reduced as needed, extremely high or low evaporating temperatures may be avoided, and high suction temperatures to the compressor may be avoided. In addition, the capacity of the cooling unit may be controlled. By manipulating the hot gas bypass valve, the condensing pressure may be reduced to prevent the cooling unit from reaching a high threshold (cutoff) pressure, the evaporating temperature may be controlled and the air temperature may be controlled as well. Also, the hot gas bypass valve may be manipulated to open during a compressor “off cycle” to expedite the coolant pressure equalization for faster and quieter compressor restarts. By closing the hot gas bypass valve periodically, the delivery of coolant to the condenser may be increased to force the delivery of oil that may be trapped back to the compressor.
As discussed, the coolingunit10 is modular and scalable so that a person designing a cooling system for the data center may select individual components. Specifically, depending on the electronic equipment deployed within the data center, and the optimum operating conditions required for the equipment, the person may employ any number of cooling units to provide primary or supplemental cooling to the particular data center. In one embodiment, the location of the cooling units within the room may be determined using a computer aided design tool. Reference is made to U.S. patent application Ser. No. 11/120,137, entitled “METHODS AND SYSTEMS FOR MANAGING FACILITY POWER AND COOLING,” filed on Apr. 7, 2005 and U.S. Provisional Patent Application No. 60/719,356, entitled “METHODS AND SYSTEMS FOR MANAGING FACILITY POWER AND COOLING,” filed on Sep. 22, 2005, which are assigned to the assignee of the present application and incorporated herein by reference. These applications generally disclose systems and methods for designing data centers and for managing equipment contained within the data center.
In one configuration, the cooling units may be packaged and shipped to the data center from a manufacturing or distribution facility. Once received, the cooling units may be assembled and otherwise installed within the data center. Specifically, the cooling units are suitably connected to a power source (not shown) and the controller to complete the installation.
As referenced above, in one embodiment, the controller may be a separately dedicated unit that controls the operation of one or more of the cooling units. In another embodiment, a main controller may be provided in one of the cooling units in place of one of the subservient controller units, with the cooling unit having the controller functioning as the main cooling unit and the other cooling units functioning as subservient cooling units. In yet another embodiment, the operation of the cooling unit may be under the control of an integrated data center control and monitoring system with each cooling unit rack having a controller that communicates with the other cooling units over the network. In one such embodiment, the controller may communicate with a data center control system to provide status of the components of the cooling system and to receive control commands for the data center control system. In one embodiment, each cooling unit includes a controller that communicates with the data center controller over a network, such as a CAN Bus network, and in one such embodiment, the data center controller may be implemented using the integrated data center control and monitoring system, such as the InfraStruXure™ data center manager sold by American Power Conversion Corporation of West Kingston, R.I., the assignee of the present invention.
In certain embodiments, the cooling unit may take the form of a kit for cooling a data center. Depending on the volume of space of the data center, the components of the kit are scalable to meet the cooling requirements of the data center. In one embodiment, the kit comprises a predetermined number of cooling units adapted to be interspersed within rows of equipment racks in the data center. The cooling units may embody the cooling unit described above.
Thus, it should be observed that cooling units of embodiments of the present invention are particularly configured for scalable and modular implementation within a data center. The cooling system may be provided in kit form that may be installed by personnel having no particular training in cooling system installation and no specialized tools. One benefit of the cooling unit is that it may be movable within a data center, or to another data center, when environmental conditions or needs within the data center change. Another advantage is that each cooling unit is self-contained, in that only power and communication needs to be delivered to each unit. No external cooling systems are required.
In addition, since the cooling unit may be provided as an in-row product, the cooling unit may be positioned to intake the hottest air in the data center and to cool it slightly below ambient temperature. This design feature eliminates the inefficiency of mixing hot air with the room temperature air to get a warm mixture. The design also significantly decreases latent cooling provided by the air conditioner, thereby potentially eliminating the need for humidification. The improvements to efficiency may best be seen by the fact that the foot print of a cooling unit (e.g., cooling unit10) may be decreased (e.g., by up to thirty percent) to obtain the same cooling performance. Specifically, the provision of movable cooling units having casters and leveling feet improves the efficiency and the scalability of the cooling system. To assist the operator in optimizing the locations of the cooling units, the cooling capacity of each unit may be monitored by the operator, along with the flow rate, coolant and air inlet and outlet temperatures, and pressure differentials. These readings enable the operator to place the cooling units where each cooling unit may neutralize the maximum amount of heat, while providing higher flexibility to the operator in the room design and layout and removing the constraint of having air conditioners positioned around the periphery of the data center. From a power perspective, each cooling unit operates under direct current, thus providing some level of flexibility to the input power provided. Thus, a cooling unit no longer needs to be built for a specific voltage.
As described above, the cooling unit of embodiments of the invention may be further provided as part of an integrated data center control and monitoring system. When used with such an integrated control and monitoring system, the cooling unit is easily removable for service and relocation to another position within the data center. The cooling unit may also be integrated into an existing cooling system of the building housing the data center, for example and used in conjunction with one or more CRAC units to provide additional cooled air where needed in the data center.
The cooling unit may be provided with a predictive failure determination module by utilizing a number of factors. Specifically, through the controller, each cooling unit may be designed to notify the data center operator when certain parts, such as motors, fans, or any other part subject to wear, are near the ends of their useful life. The provision of such a module will enable a reasonably timed preventative maintenance action to be performed and to save possible downtime. The notification may be delivered to the display of the cooling unit, or provided to the operator of the data center through the integrated control and monitoring system. In addition, a controller of the cooling unit configured as a main controller may compensate for a failure of a particular cooling unit by increasing the output of other cooling units positioned near the failed cooling unit.
With the cooling unit of embodiments of the present invention, it is observed that the need for a raised floor is eliminated. By eliminating the raised floor, costs associated with designing and providing the raised floor are eliminated. In addition, the equipment housed by the equipment racks may be better anchored to the floor of the data center for enhanced earthquake resistance. The number of suitable sites for server rooms or data centers is increased because rooms with relatively low headroom may now be utilized. Additionally, the need for raised floor ramps is eliminated.
The cooling unit of embodiments of the present invention is faster to install than prior systems. Since the cooling unit includes a closed loop cooling system, only power and communication need be connected to the cooling unit. Thus, the data center looks more professional.
Having thus described several aspects of at least one embodiment of this invention, it is to be appreciated various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be part of this disclosure, and are intended to be within the spirit and scope of the invention. Accordingly, the foregoing description and drawings are by way of example only.