JPS61112534A - Control system of power supply - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a power supply control method for determining an abnormal unit among a plurality of power supply units operating in parallel redundantly.

With the recent trend of data processing devices going online, it has become necessary to increase the reliability of the data processing devices.

For this reason, a parallel redundant operation method has been introduced from the conventional parallel operation method to power supplies. For example, if one redundant unit is installed, even if one power supply unit fails at a time, It has become possible to prevent a data processing device to which power is supplied from the power supply device from going down.

However, in such a parallel redundant operation method, because the output voltage does not change due to the redundant configuration, the faulty unit cannot be detected and operation continues without redundancy. When a failure occurs, there is a risk that the system will go down.

However, regular maintenance and one-time maintenance are usually required for data processing equipment.
．． Since initial diagnosis is performed when power is turned on, a method has been awaited that can effectively detect faulty units during periodic maintenance.

[Conventional technology]

FIG. 3 is a block diagram showing a conventional parallel redundant operation system, in which 1 is a load, 20 to 22 are power supply units (PSU 1 to PSU 3), and A is a low voltage detection circuit.

OC is the DC output voltage, AC is the AC input voltage, and C is the interface connector.

In the following description, for convenience, only one redundant unit will be described.

Currently, due to some circumstances, the power supply unit (PSU 1)
When a failure occurs inside the power supply unit) (P
SU1) 20 is no longer able to supply the output voltage DC to the load 1.

However, it has a redundant parallel configuration and the load 1 is light.
Power supply unit (PSIJ 2) 21. Power supply unit(
PSU 3) 22 alone is sufficient to supply power to the load l, and the output voltage DC does not drop, so the low voltage detection circuit A does not detect the faulty unit and there is no problem in operation. becomes.

[Problem that the invention seeks to solve]

Therefore, in the conventional system, the power supply unit (PSU
1) Since 20 continues to operate without redundancy due to failure, the abnormality cannot be discovered until the second failure occurs, and there is a risk of inducing a system down for load 1. There was a problem that there was.

In view of the above-mentioned conventional drawbacks, the present invention provides a method in which, for example, one redundant unit is provided and when a failure occurs in one of the power supply units operating in parallel redundantly, the faulty power supply unit can be detected. The purpose is to

[Means for solving problems]

The purpose of this is to provide, for example, one redundant unit in a device that includes a plurality of power supply units 7) operating redundantly in parallel, and a power supply control section that performs on/off control of the power supply units. When one of the power supply units becomes abnormal and its output is turned off, the power supply control section turns off each power supply unit by 1 minute,
When the power supply unit that has been turned off is a normal unit, an overcurrent will drop and the output voltage will drop. When the power supply unit that has been turned off is abnormal, it will be detected that the output voltage does not drop and the abnormal power supply will be removed. This is achieved by the power supply control method of the present invention, which performs control to discriminate between

[Effect]

That is, according to the present invention, for example, when one redundant unit is provided and a plurality of power supply units are operated in a parallel redundant format, if an abnormality occurs in one of them, the DC output voltage DC is Even when it could no longer be supplied to the load, the low voltage detection circuit A might have detected the DC output voltage DC.
Due to the above redundant configuration, it is not possible to detect the abnormal unit, so when performing regular maintenance or turning on the power, each power supply unit must be turned off by the power supply control unit (UPC). Therefore, if the power supply unit turned off is normal, an overcurrent will drop and the above DC output voltage OC will drop, and an alarm signal from the low voltage detection circuit A will be detected, and if the power supply unit is abnormal, Since it is possible to recognize an abnormal unit by detecting that the DC output voltage DC does not drop, it is possible to prevent continuous operation without redundancy and eliminate the risk of inducing a system failure. There is an effect.

〔Example〕

Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram illustrating an embodiment of the present invention, and FIG. 2 is a flow diagram illustrating an example of the operation of detecting a faulty unit by implementing the present invention. Also in this embodiment, for convenience, the number of redundant units will be limited to one.

In FIG. 1, the same symbols as in FIG. 2 indicate the same objects, and 3 indicates the power supply control unit (UPC
) and is controlled, for example, by a service processor (SVP) 4 which is usually connected to the data processing device.

Now, assuming that the power supply unit I-(PS(11) 20) is in trouble, the power supply unit
SU 2) 21 and power supply unit (PStl 3)
22 alone can supply power to load 1 and continue operation.

In this state, during periodic maintenance or when turning on the power, the power supply control unit (UPC) 3 is controlled by the diagnostic program executed by the service processor (SVP) 4 and the power supply unit (PSU 1) 20 ~(PSU 3
) 22 to one unit and observe the change in the DC output voltage OC.

At this time, there is no redundancy, so if the normal power supply unit (PSU 2) 21 is turned off, power supply unit I- (PSU 3) 22 alone will not be able to supply power to load 1, and the DC output The voltage OC drops due to an overcurrent, and the low voltage detection circuit A detects the drop in voltage and issues an alarm signal, but an abnormal power supply unit (PS
When U1) 20 is turned off, the DC output voltage DC does not change, so the alarm signal is not generated.

Furthermore, when there is redundancy, even if all power supply units are turned off one by one, the DC output voltage DC does not change, so it can be determined that everything is normal.

The above faulty unit/unit identification operation will be explained in more detail with reference to the flowchart of FIG.

Step 50: One power supply unit is in parallel redundant operation.

Step 51: A failure occurred in one power supply unit, but it continues to operate.

Steps 52, 53: Service processor (SVP)
) In step 4, execute the program for periodic maintenance and enter the power supply unit number N to turn off the power supply unit.
= 1.

Step 54: Turn off the power supply unit N.

Step 55: An alarm signal is detected, and if the signal can be detected, the process jumps to step 61, but if the alarm signal cannot be detected, the process moves to the next step.

Step 56: Set N=N+l.

Step 57: Check whether N-n+1, N=n+1
When , all power supply units have been turned off, so move on to the next step. However, when N≠n+1, there are still power supply units that need to be turned off, so proceed to step 5.
Return to 4.

Step 58: When all power units are turned off,
It is checked whether an alarm signal has been detected, and if an alarm signal has been detected, the process moves to step 59, but if no alarm signal has been detected, the process jumps to step 60.

Step 59: Since the unit for which an alarm signal was detected in step 61, which will be described later, is determined to be normal, the unit for which the alarm signal could not be detected is determined to be abnormal.

Step 60: All power supply units are turned off and no alarm signal is detected, which means that there is no change in redundancy, so it is determined that everything is normal.

Step 61: Detecting an alarm signal means that
Since turning off the power supply unit means that redundancy is lost, it is determined that the power supply unit I-N was operating normally.

As mentioned above, the above operation flow is limited to when the redundancy level is 1 unit, but when the redundancy level is 2 units, by executing turning off for 2 units, for example, for all combinations, It goes without saying that a faulty unit can be detected.

Therefore, it can be seen that, in general, even when the redundancy level is m, by applying the present invention and performing m-redundant off for all combinations, it is possible to detect a faulty unit. .

〔Effect of the invention〕

As described above in detail, the ta control method of the present invention is effective when, for example, one redundant unit is provided and a plurality of power supply units are operated in a parallel redundant format.
Even if an abnormality occurs in the stand and the DC output voltage DC cannot be supplied to the load, if the DC output voltage DC is detected by the low voltage detection circuit A, the redundant configuration described above will prevent the abnormal unit from supplying the DC output voltage DC to the load. Therefore, when performing regular maintenance or turning on the power, the power supply control unit (UPC) can turn off each power supply unit by one unit, and the power supply unit that has been turned off will be automatically turned off. If it is normal, the DC output DC will drop due to overcurrent droop, and an alarm signal from the low voltage detection circuit A can be detected, and if the power supply unit is abnormal, the DC output voltage DC will not drop. Since it is possible to detect and recognize an abnormal unit, there is no need to continue idle operation without redundancy, and there is an effect that there is no danger of inducing a system failure.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention. FIG. 2 is a flowchart illustrating an operational example of detecting a faulty unit by implementing the present invention. FIG. 3 is a diagram showing an example of a configuration when parallel redundant operation is performed using a conventional method. It is. In the drawing, 1 is the load. 20 to 22 are power supply units (PSU 1 to PSU 3)
. 3 is a power supply control unit (UPC). 4 is a service processor (SVP). 50 to 61 are each step of the faulty unit detection operation. A is the low voltage detection circuit, and DC is the DC output voltage. AC is alternating current input voltage. are shown respectively. Leather I picture grass 3 times

Claims (1)

[Claims] m units of redundant units are installed and are operated in parallel redundantly.
In a device including +n power supply units and a power supply control unit that performs on/off control of the m power supply units,
When one power supply unit among the above m+n power supply units becomes abnormal and its output is turned off, the above power supply control unit turns off each power supply unit to m units and all combinations of said m units. However, when the m power supply units that were turned off were normal units, overcurrent drooped,
When the output voltage decreases and an abnormal unit is included in the m power supply units turned off, control is performed to detect that the output voltage does not decrease and determine the abnormal power supply unit. A power control system featuring: