Electrical transmission or interconnection systems – Plural load circuit systems – Plural sources of supply
Reexamination Certificate
2000-01-18
2003-01-07
Sircus, Brian (Department: 2836)
Electrical transmission or interconnection systems
Plural load circuit systems
Plural sources of supply
C307S126000, C700S286000
Reexamination Certificate
active
06504266
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to the power up of an electronic system in response to the application of AC power and, more particularly, to power up of a system which has multiple independent power supplies.
BACKGROUND OF THE INVENTION
In many electrical systems, such as computers, it is often desirable to have several independent power supplies operational in the system simultaneously. The term “independent” is used here to mean, among other things, that each supply has its own connection to the main AC power source and produces all, or a portion of, the total DC power which is used to run the system. In most cases, however, the DC power outputs of the power supplies are connected in parallel. There are several advantages inherent in multiple supplies.
For example, the ability to add power supply units to the system after the initial configuration allows the number of power supply units to be tailored in accordance with system power requirements. In a computer system with pluggable units such as CPU's, memory, I/O cards, and disk drives, the power consumption of the system may vary dramatically depending on the current set of installed units. The number of power supplies that are installed can be selected to match the actual power load.
Further, extra power supplies can be added for redundancy. In contemporary terminology, systems can be made “N+1” redundant with respect to the power subsystem by adding one more power supply than is actually needed to run the system. In the event of the failure of one supply, such systems can continue operating normally while the failed supply is being replaced.
In general, it is often required that, following a loss, and subsequent reapplication, of the main AC power, the system remain in the same state that it was in before power was removed. In other words, the system must have some facility for remembering the state of its power supply subsystem (ON or OFF) at the time of the AC removal, so that, following the reapplication of AC power, the system can resume its previous state. The removal of AC power could occur either intentionally when the system is powered up or unintentionally in the case of an AC power failure. If the system was “ON” before the power removal, it is often necessary for the system to correctly power up and resume normal operation.
However, when such a system has multiple power supply units and each unit has an independent connection to the main AC power, a difficulty can arise in correctly recovering from a temporary lapse of the main AC power. In particular, since each power supply has its own connection to AC power, it will recover when AC power is applied and it is possible that the separate AC power connections might not recover simultaneously.
For example, consider a system with “M” independent power supplies which, while in the “ON” state, loses AC power. Further, assume that the system requires at least “N” operational power supplies to power it, where “N” is less than or equal to “M”. Depending on the source and routing of AC power in the facility where the system is located, it is possible that the “M” power supplies may experience the reapplication of AC power at different times, or “steps” over several seconds or longer. This staggered power application could be due either to fault conditions in the facility, or to a controlled reapplication of AC power via circuit breakers.
When power is reapplied, if “L” power supplies receive AC power in the first such “step”, where “L” is less than “N”, then the power consumption requirements of the system will exceed the capabilities of the “L” supplies and the operational power supplies will be overloaded. In this situation, the operational power supplies experience over-current faults. Typically, once a power supply enters such an over-current fault state, it must be manually serviced before it will again attempt to power the system. This service can be accomplished by manually turning the supply OFF and then ON again.
At the next “step” in the facility's recovery of AC power, and at each subsequent “step”, if less than “N” supplies receive AC power at the same time, then they will also enter an over-current state. Thus, it is possible for all the supplies in the system to enter the over-current fault state, in two or more steps. Most system designs would require manual intervention once in this state before the system could be successfully powered up.
SUMMARY OF THE INVENTION
In accordance with the principles of the invention, a method and apparatus dynamically determines the number of power supplies required to power up a system without overloading any supply. The power subsystem is required to wait until the determined number of power supplies has received AC power before attempting a system power on.
In accordance with one embodiment, the amount of required power is determined before power up based on the actual power load present. The actual load is determined by sensing load indicators in each load device and computing the total power load.
In accordance with another embodiment, the number of power supplies which have received AC power is determined by detecting when AC power has been applied to each power supply and computing the total amount of power available for power up.
In accordance with a further embodiment, power up is delayed when the number of power supplies that have received AC power is insufficient to power the system without an overload situation occurring.
REFERENCES:
patent: 3300648 (1967-01-01), Rockefeller, Jr. et al.
patent: 3379893 (1968-04-01), Cavanaugh
patent: 3617812 (1971-11-01), Deter
patent: 4370562 (1983-01-01), Palazzetti et al.
patent: 4694192 (1987-09-01), Payne et al.
patent: 4720758 (1988-01-01), Winslow
patent: 4951171 (1990-08-01), Tran et al.
patent: 5017799 (1991-05-01), Fishman
patent: 5142165 (1992-08-01), Allard et al.
patent: 5191520 (1993-03-01), Eckersley
patent: 5268850 (1993-12-01), Skoglund
patent: 5317366 (1994-05-01), Koshi et al.
patent: 5483656 (1996-01-01), Oprescu et al.
patent: 5583419 (1996-12-01), Haller
patent: 5604421 (1997-02-01), Barnsley
patent: 6046513 (2000-04-01), Jouper et al.
patent: 6233693 (2001-05-01), Berglund et al.
patent: 6301674 (2001-10-01), Saito et al.
patent: 2343044 (1973-08-01), None
patent: 11-18295 (1999-01-01), None
Kudirka & Jobse LLP
Rios Roberto J.
Sircus Brian
Sun Microsystems Inc.
LandOfFree
Method and apparatus for powering up an electronic system... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for powering up an electronic system..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for powering up an electronic system... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3070974