Refrigeration – Automatic control – Diverse – cascade or compound refrigeration-producing system
Reexamination Certificate
2003-07-29
2004-11-09
Norman, Marc (Department: 3744)
Refrigeration
Automatic control
Diverse, cascade or compound refrigeration-producing system
C062S236000, C307S064000, C307S080000
Reexamination Certificate
active
06813897
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to power systems. More particularly, the invention relates to supplying power to at least one cooling system component.
BACKGROUND OF THE INVENTION
As chip designers are continually increasing the number of transistors that may be placed on a chip, computer systems are becoming ever more powerful. Given the increase of circuit density on chips, power consumption of these computer systems is also increasing. In some instances, when a large number of computers are provided in one location (e.g., in a data center), power consumption may be several megawatts (MW). Cooling systems for the computer systems may consume 30-40% of the total power consumption. Accordingly, cooling systems should be considered when designing a power system for the computer systems.
Computer systems often utilize a cooling system for removing heat dissipated by components of the computer systems (e.g., one or more processors, memory, power supplies, and other circuits). Excessive heat tends to adversely affect the performance and operating lives of these components. In recent years, these components have become denser and, hence, generate more heat during operation. Furthermore, when a plurality of computer systems are housed in one enclosure (e.g., a rack or cabinet) and/or stored in the same location (e.g., a data center), there is an even greater potential for the adverse effects of overheating.
In order to substantially guarantee proper operation, and to extend the life of the computer systems, it is necessary to maintain the temperatures of the components within predetermined safe operating ranges. Subjecting the components to temperatures above recommended maximum operating temperatures may result in irreversible damage to the components. In addition, it has been established that the reliabilities of computer system components decrease with increasing temperature. Therefore, the heat dissipated by the components during operation must be removed at a rate that ensures that operational and reliability requirements are met. Cooling systems are used to remove heat generated by the computer system components.
Cooling systems may include fans, air conditioning units, cooling liquid, etc., to facilitate heat dissipation. One type of conventional cooling system for computer systems is a heating, ventilating, and air-conditioning system (HVAC system). Some of the power consuming components of an HVAC system include compressors, blowers, and pumps.
These components tend to consume a significant portion of the total power consumption of the cooling system.
Power systems for computer systems and cooling systems are typically designed to meet the maximum power demand of a load and for redundancy. While these factors are important in power system design, energy efficiency is an equally important factor that is usually not given the same weight as other factors when designing power supplies.
FIG. 6
illustrates a conventional power supply
500
modeled as a black box with power entering the black box (input power) and conditioned power (output power) exiting the black box. Conditioning may include alternating current (A/C) or direct current (D/C) conversions (e.g., AC/AC, AC/DC, DC/DC, etc.), and the like. Ideally, there would be no losses between the input power and the output power. However, in reality, losses occur during the conditioning, typically as heat dissipation. Efficiency of a power supply may be measured as the ratio of output power over input power. For example, if 100 Watts (W) are input to the power supply
500
and 75 W of conditioned power exits the power supply
500
, the power supply has 75% efficiency. 25 W of heat may be dissipated by the power supply
500
. The efficiency of a power supply is usually provided by a manufacturer, but may also be measured. Energy efficiency of other power system components may similarly be determined.
FIG. 7
illustrates an exemplary efficiency curve for an AC/DC power supply input power at 200 Volts (V) and 60 Hertz (Hz). The efficiency curve of
FIG. 7
may be provided by the power supply manufacturer or determined through power measurements. Referring to the efficiency curve of
FIG. 7
, the power supply is approximately most efficient (e.g., approximately 80%) with a power output between 400 W and 450 W. Conventional power systems for computer systems use at least two power supplies for redundancy, whereby each power supply is operable to meet the power demand of the computer systems unilaterally in is case of failure of one of the power supplies. However, for the majority of their operation, both power supplies are operational and are usually designed to split the load. Therefore, if the computer systems demand 400 W, the power supplies each only operate at approximately 74% efficiency (e.g., each power supply supplying an output power of approximately 200 W at 74% efficiency per power supply). If three power supplies are used, each of the power supplies only operates at approximately 64% efficiency. Therefore, conventional power systems for computer systems typically sacrifice efficiency for other factors (e.g., redundancy), which leads to increased energy costs.
Power factor is another important characteristic related to energy efficiency of power system components, which impacts the sizing of the electrical wires and equipment that supply energy to a power supply and the cost of electricity. Power factor is the ratio of real power over apparent power (see Equation 1).
Power Factor=real power/apparent power Equation (1)
Power factor is based on the type of load on the power supply. A purely resistive load has a power factor of 1, which is ideal, because the real power is equal to the apparent power. However, for non-purely resistive loads, real power is less than apparent power, lea ding to power factors less than 1. As the difference between apparent and real power increases (i.e., with smaller power factors), more current must be generated by the power source in order to deliver a specific amount of real power to the load. For example, in a system with a power factor of 0.5, to deliver 100 W of real power (10 Amps at 10 Volts) requires the power source to provide 20 Amps at 10 Volts. In a load with a sinusoidal voltage and current, the real power is equal to the product of the RMS input voltage (V), input current (
1
), and cos (&phgr;), where cos (&phgr;) is the phase angle between the voltage and the current. Cos (&phgr;) is the power factor.
The difference between apparent and real power impacts the cost of power system components that supply power to a computer system or cooling system, because all the electrical components upstream of a power supply must be sized for a higher current. In addition, because all components dissipate some heat when current passes through them, higher currents translate into greater power wastage. To offset this cost and the cost of the greater power wastage, electrical utilities charge, in general, more for electricity provided to lower power factor loads.
Typically, power supplies for computer systems may have a power factor between 0.6 and 0.8. A poor power factor may be the result of a large amount of reactive power caused by an inductive load. The output power of a power supply can be modeled based on power factor and efficiency (see Equation 2).
Output Power=efficiency*power factor*apparent power Equation (2)
FIG. 8
illustrates an exemplary power factor curve for the power supply having the efficiency curve shown in FIG.
7
. The power factor curve shown in
FIG. 8
may be provided by a manufacturer (e.g., based on a predetermined load) or may be calculated from power measurements. Based on this power factor curve, a higher power factor is achieved generally as output power of the power supply is increased. Power factor correction circuits are generally used to improve power factor. However, power factor is typically not considered when optimizing the efficiency of a power supply or power system.
SUMMARY OF THE INVENTI
Bash Cullen E.
Farkas Keith Istvan
Janakiraman Gopalakrishnan
Patel Chandrakant D.
LandOfFree
Supplying power to at least one cooling system component does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Supplying power to at least one cooling system component, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Supplying power to at least one cooling system component will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3331436