WO2006137112A1

WO2006137112A1 - Power supply device and control method thereof

Info

Publication number: WO2006137112A1
Application number: PCT/JP2005/011250
Authority: WO
Inventors: Atsunori Kitsuki
Original assignee: Fujitsu Limited
Priority date: 2005-06-20
Filing date: 2005-06-20
Publication date: 2006-12-28

Abstract

When at least one of power supply devices operating in parallel has failed to cause an overcurrent state, which in turn causes an overcurrent state in other power supply devices, operation of the power supply devices in parallel operation is stopped. When the operation of the power supply devices is resumed, operation of only the device which has failed is stopped by an overcurrent protection function prior to the normal devices. The normal devices do not reach operation stop due to the overcurrent state caused by the affect of the device which has failed, until the operation stop after the operation resumption.

Description

Specification

Power supply device and control method thereof

Technical field

The present invention relates to a power supply device and a control method thereof, and more particularly to a power supply device that supplies a power to a load by operating a plurality of power supply devices in parallel and a control method thereof.

Background art

[0002] In recent years, power supply devices that use DC-DC converters (hereinafter simply referred to as “DDC”) have the necessary power of distributed power supply. In general, a configuration in which multiple small on-board DDCs are mounted on each printed circuit board It is the target. These on-board DDCs—the output current per unit is 40 to 70 A, and power is supplied to a high-end UNIX (registered trademark) server, a mainframe CPUZLSI, etc. by performing parallel redundant operation of 2 to 8 units.

[0003] When N + 1 DDC units that can normally perform redundant operation are operated in parallel, even if one of them malfunctions, the remaining N units are designed to be able to supply power of the desired capacity. At that time, so-called butt diodes are inserted into each DDC output so that the effects of the failure do not reach other DDCs. Even if a short-circuit failure occurs in one DDC due to the insertion of this butt diode, the output overcurrent state occurs in other normal DDCs due to the effect, and all DDCs in parallel operation are stopped. It is possible to prevent the occurrence of a situation that leads to

[0004] In the case of an on-board type DDC with a large output current, however, the number of matching diodes required for this increases, and as a result, adverse effects such as an increase in device scale and an increase in power loss may occur. is there. In order to avoid this, the installation of a matching diode may be intentionally omitted, especially when the failure rate of the DDC is low. In this case, if a transformer secondary side short circuit fault occurs in one DDC during parallel redundant operation of the DDC, the faulty DDC is put into an output stop state by the overcurrent protection function. At that time, other normal DDCs in parallel redundant operation may also generate a pull-in current due to the effects of the short-circuit failure, and the output may be stopped due to the overcurrent protection function.

Patent Document 1: Japanese Unexamined Patent Publication No. 2003-169471 Disclosure of the invention

Problems to be solved by the invention

[0005] In such a case, a so-called system partition down state or total system down state occurs, and the devices that monitor the operation of these parallel DDCs recognize that all these DDCs have failed at the same time. May be. In that case, all DDCs will be replaced even though only one particular DDC is out of order. At that time, if the number of units to be replaced is large, long-term replacement work is required, and as a result, the system is stopped for a long time.

[0006] The present invention has been made in view of the above problems, and when a specific one of the DDCs in parallel redundant operation falls into a secondary side short circuit fault or the like, the DDC is automatically disconnected. Then, the purpose is to provide a power supply device that can continue to supply power to the load by the remaining normal DDC and a control method thereof.

Means for solving the problem

[0007] In order to achieve the above object, in the present invention, when a plurality of power supply devices in parallel redundant operation are in a stopped operation, the operation by the overcurrent protection function is performed in the failed device before the normal device. On the other hand, control is performed so that other normal devices do not stop operation due to the failure device during that time. The invention's effect

[0008] According to the present invention, when the operation is restarted after stopping a plurality of units, the malfunctioning device is stopped and automatically disconnected from the parallel operation system. In other normal devices, an overcurrent state is temporarily generated due to the influence of the failed device until the operation is stopped as described above and the device is automatically disconnected from the parallel operation system. However, due to the control function according to the present invention, the normal device in parallel operation does not stop operating until the faulty device is stopped as described above and disconnected from the parallel operation system. Then, after the failure device is stopped and disconnected from the parallel operation system, the effect of the failure device is removed by the disconnection, so that the normal device returns to the normal operation state and the power supply to the load can be continued.

Brief Description of Drawings FIG. 1 is a block diagram of a power supply device according to an embodiment of the present invention.

FIG. 2 is a circuit diagram showing an internal configuration of the DDC in FIG.

FIG. 3 is a diagram for explaining a situation when a failure occurs in the configuration of FIG. 2.

FIG. 4 is an operation flowchart (part 1) when a failure occurs according to the configuration of FIG.

FIG. 5 is an operation flowchart (part 2) when a failure occurs according to the configuration of FIG.

FIG. 6 is an operation flowchart (part 3) when a failure occurs according to the configuration of FIG.

FIG. 7A is a diagram showing VZl characteristics when an output overcurrent occurs during steady DDC operation.

FIG. 7B is a diagram showing VZl characteristics when an output overcurrent occurs at the start of DDC operation.

FIG. 8A is a diagram showing the relationship between the current when an output overcurrent is generated and the overcurrent detection time during steady DDC operation.

FIG. 8B is a diagram showing the relationship between the current when the output overcurrent is generated and the overcurrent detection time at the start of DDC operation.

Explanation of symbols

[0010] 10 Inverter section

20 Synchronous rectifier circuit

21 FET

30 Control IC

22 Output overcurrent protection circuit

23 Input overcurrent protection circuit

100 Control unit

201, 202 DDC (DC— DC Connaughter)

BEST MODE FOR CARRYING OUT THE INVENTION

[0011] In the present invention, while the output voltage rises at the time of re-input (resumption of operation) after the operation of all the units is stopped due to the failure as described above in a plurality of DDCs performing parallel redundant operation, As shown, failure DDC is configured so that the input overcurrent protection circuit (see Fig. 2) operates before normal DDC. As a result, only the fault DDC is stopped by the function of the protection circuit and disconnected from the parallel operation system. As a result, failure in normal D DC The output overcurrent state due to current draw by DDC is resolved, As a result, normal DDC can continue to operate normally without being stopped.

Example 1

The configuration of the embodiment of the present invention will be described below in detail with reference to the drawings.

As shown in FIG. 1, a semiconductor circuit 1 according to an embodiment of the present invention includes a logic unit 300 that executes a functional operation provided by the circuit 1, and supplies a predetermined DC power to the logic unit 300. Monitors the output voltage of the DDC201, 202, the control unit 100 that controls the operation of the multiple DDC201, 202 that function as a power supply unit that operates in parallel redundant operation, the logic device 300 and the DDC201, 202 The voltage monitor 400 supplies the result to the control unit 100 as voltage information.

The control unit 100 supplies an ONZOFF signal to the DDCs 201 and 202, thereby starting or stopping the operation of the DDCs 201 and 202. Each DDC 201, 202 transmits a failure signal to the control unit 100 when it detects a failure state in its own device.

Further, as shown in FIG. 2, each DDC 201, 202 is supplied with a DC input voltage from a DC power source (not shown) and converts it into an AC voltage, a transformer 40, and a transformer 40. The synchronous rectifier circuit 20 that generates a desired DC voltage by supplying the AC voltage from the inverter unit 10 and performing synchronous rectification processing by PWM control, and the synchronous rectifier circuit 20 according to the ONZOFF signal from the control unit 100 It consists of a control IC 30 that starts PWM control and stops Z.

Here, the power source of the control IC 30 is supplied from the inverter unit 10, and the supply is simultaneously cut off by the disconnection of the power supply from the inverter unit 10 by the operation of the input overcurrent protection circuit 23 described later.

[0017] The synchronous rectification circuit 20 has a FET (field effect transistor) 21 as a switching element, and when this is turned on / off by the function of the control IC 30, a synchronous rectification operation by PWM control is performed.

[0018] Also, the inverter unit 10 is provided with an input overcurrent protection circuit 23, which detects an input overcurrent state in the corresponding DDC, and at the time of detection, synchronizes by turning off the power supply in the inverter unit 10. Stop the operation of the rectifier circuit 20 and stop the operation of the DDC. Make it.

[0019] The output overcurrent protection circuit 22 is provided in the synchronous rectifier circuit 20, and an output overcurrent state in the corresponding DDC is detected, and by sending a predetermined signal to the control IC 30 at the time of detection, Control IC30 performs forced output current suppression operation by PWM control

[0020] The operation when a failure occurs in the power supply device according to the embodiment of the present invention will be described below in detail with reference to FIGS. 2 to 8B.

[0021] In step S1 of Fig. 4, the operations of the DDCs 201 and 202 are started. Specifically, when the ON signal in the ONZOFF signal is transmitted to the control IC 30 of each DDC, the control IC 30 starts the PWM control operation for the FET 21 of the synchronous rectifier circuit 20, and thus a predetermined DC voltage is generated. This is supplied to the logic unit 300 (steps S2, S3)

Here, it is assumed that a failure occurs in the control IC 30 of the DDC 201 due to some factor in step S4, so that the FET 21 of the synchronous rectifier circuit 20 is always on (state in FIG. 3).

[0023] As described above, when the FET 21 of the synchronous rectifier circuit 20 of the fault DDC 201 is always on, the circuit functions as a load circuit, and power is supplied from the other DDC 202. That is, as shown in FIG. 3, a pull-in current is generated in the synchronous rectifier circuit 20 of the failed DDC 201 (step S5), and an excessive output current is supplied from another normal DDC 202 to supply this current.

Here, the operation at the time of failure in the failure DDC 201 will be described in detail with reference to the drawings.

[0025] As a result of the failure of the synchronous rectifier circuit 20 in the DDC 201 and the FET 20 being always on as described above, a secondary circuit short circuit occurs. As a result, the transformer 40 is saturated, an overcurrent flows to the input side, and the input overcurrent protection circuit 23 is activated.

[0026] At this time, the other DDC 202 is in an output overcurrent state due to the current drawn by the fault DDC201. In such a case, if the input overcurrent protection circuit 23 is activated before the output overcurrent protection circuit 22 is activated in the DDC 202, the DDC operation is stopped by that function.

[0027] In this case, when the secondary circuit short circuit of the fault DDC201 occurs as described above, other normal When the time until the input overcurrent protection circuit 23 of the DDC202 is activated is Tl and the time until the output overcurrent protection circuit 22 of the same DDC202 is activated is T2, the following relationship holds.

[0028] Tl <T2

This state is shown in FIG. 8A.

The reason for the occurrence of such a state will be described with reference to FIGS. 7A and 7B.

FIG. 7A shows the relationship between the output current and the output voltage when an output overcurrent occurs in the steady state of the DDC. As shown in the figure, the output voltage when reaching the current value OCP where overcurrent protection works is a steady value (steady voltage).

[0031] FIG. 7B shows the relationship between the output current and the output voltage when an output overcurrent occurs in the DDC operation start state (at startup). As shown in the figure, when the current value OCP at which overcurrent protection works is reached, the output voltage has not yet reached the steady value shown in Fig. 7A, and is still in a very low voltage state. As a result, it can be seen that the power consumption at the time when overcurrent protection functions is considerably lower at the start of DDC operation than in steady operation. As a result, the power supplied from the inverter unit 10 at that time is also in a low power state, and the input current is also small.

[0032] In this way, compared to the steady operation, at the start of operation, the input current at the time when the output overcurrent protection circuit 22 operates is small. For this reason, even if a short circuit failure occurs during steady operation, the input overcurrent protection circuit 23 is activated as soon as possible for the safety of the equipment. In the event of a short circuit failure at the start, it is possible to set the output overcurrent protection circuit 22 to operate before the input overcurrent circuit 23 operates. In other words, the output current reaches the overcurrent protection function operation OCP (output) value before the overcurrent protection function operation OCP (input) value is reached as shown by the solid line in Fig. 8B. Can be set. That is, in this case, unlike above, Tl> T2

It becomes.

[0033] On the other hand, in the case of the DDC202, the FET21 is always turned on due to the failure, and the PWM control does not function. As a result, the output current cannot be reduced even when the output current reaches the OCP (output) value for the overcurrent protection function. As a result, the output current continues to increase, In order to supply the power, the input current by the inverter unit 10 also increases. Then, the input current as shown by the broken line in FIG. 8B reaches the OCP (input) value of the overcurrent protection function operation, the power is cut off in the inverter unit 10, and the operation of the fault DDC201 is stopped. Become.

Hereinafter, a case where the input overcurrent protection circuit 23 is activated before the output overcurrent protection circuit 22 is activated by the operation described above with reference to FIG. 8A (step S6) will be described with reference to FIG. Furthermore, unlike this, the case where the output overcurrent protection circuit 22 is activated first (step S7) will be described with reference to FIG.

[0035] In the case of FIG. 4, in steps S41 and S61, an output overcurrent state during steady operation occurs, and the input overcurrent protection circuit 23 operates in both DDC201 and 202. As a result, when the power supply from the inverter unit 10 is cut off in steps S42 and S62, the operations of the DDCs 201 and 202 are stopped. As a result, a failure signal is transmitted from each control IC 30 to the control unit 100 (steps S43 and S63).

[0036] Note that the power supply of the control IC30 is also cut off when the power supply from the inverter unit 10 is cut off as described above. Due to the configuration of the IC30, for example, by applying an open collector type output structure, etc. The failure signal is maintained even if the signal is cut off. This failure signal is canceled by this detection when the output of the synchronous rectifier circuit 20 returns to the steady value in each DDC 201, 202.

[0037] The control unit 100 that has received the failure signal of each of the DDCs 201 and 202 performs corresponding alarm detection (step S31). In this case, since the alarm detection is based on the failure signal from the two DDCs 201 and 202, the preset “re-injection process when detecting alarms in multiple DDCs” is performed in step S32.

That is, the operation of the DDCs 201 and 202 is restarted by starting the supply of power from each inverter unit 10 (step S35).

[0039] In this case failure DDC201, becomes FET21 is always ON state again due to a failure of the control IC 30, similarly to the above secondary side short-circuit condition current draw occurs for generating (scan Tetsupu S45, S46) ₀ result input The overcurrent protection circuit 23 is activated (step S47), and the power supply from the inverter unit 10 is again turned off (step S48) as described above. As a result The operation of DDC201 stops again (step S49).

[0040] As described above, the power supply from the inverter unit 10 is cut off (step S48), so that the power supply to the control IC 30 is cut off (step S51). As a result, the failure DDC 201 is released from the on state of the FET 21 that is always on (step S52).

[0041] In this way, the fault DDC201 in which the input voltage is disconnected at the inverter unit 10 by the input overcurrent protection circuit 23 does not saturate the transformer 40, and as a result, no current is drawn.

On the other hand, in the normal DDC 202, the output overcurrent protection circuit 22 operates to reduce the output current by PWM control (steps S65, S66). That is, in this normal DDC 202, an overcurrent is generated until the input overcurrent protection circuit 23 of the fault DDC201 is activated (step S47). As a result, the output overcurrent protection circuit 22 is activated to maintain the output drooping state. In this case, since the output overcurrent protection circuit 22 is activated during the rise of the output voltage, the input current is reduced and the input overcurrent protection circuit 23 is not activated (step S67). Therefore, the output is not stopped (step S68). .

[0043] As described above, when the input overcurrent protection circuit 23 of the failure DDC201 is activated, the operation of the DDC201 is stopped and the short-circuit state is eliminated, so that the normal DDC202 is released from the overcurrent state. Accordingly, the DDC 202 shifts to normal operation and resumes power supply to the logic unit 300 (step S69).

[0044] Since the output voltage of the DDC 202 that has shifted to the steady state in this way becomes a steady value, the failure signal is canceled as described above (step S70). On the other hand, in the failure DDC201, the failure signal is maintained because the failure state is maintained. When this is detected by the control unit 100, the control unit 100 recognizes that the DDC 201 is a fault DDC (step S37).

In this way, in the embodiment of the present invention, the control unit 100 can identify the DDC 201 whose output has been stopped (step S49) after being turned on again as a fault DDC. In this way, in the case of N + 1 parallel redundant operation, if all the units are stopped (steps S42 and S62), the system is restarted again (step S35). As a result, it is possible to automatically identify the fault DDC201 and to shift to normal operation with N units (N = 1 in this example). did It is possible to effectively reduce downtime by using force S.

[0046] That is, in the past, multiple DDCs were determined to have failed due to a short circuit failure on the secondary side, and all DDCs were replaced together, and the system was not restarted until the replacement was completed. For this reason, the down time of the system becomes long, which sometimes becomes a big problem. On the other hand, in the embodiment of the present invention, it is possible to identify the failure DDC and shift to the steady operation as it is by executing the re-input processing as described above, and thus effectively reduce the down time. It becomes possible.

Note that the DDCs 201 and 202 activate the internal auxiliary power supply from the input voltage, thereby operating the control IC 30 to control the synchronous rectifier circuit 20, that is, the secondary side circuit. Therefore, when the input voltage is cut by the operation of the input overcurrent protection circuit 23 (step S48), the synchronous rectification circuit 20 does not function, and as a result, the transformer 40 is not saturated, and no drawing current is generated on the output side. It will be.

Next, a case where the output overcurrent protection circuit 22 operates before the operation of the input overcurrent protection circuit 23 in the event of a failure (steps S81 and S91) will be described with reference to FIG.

[0049] In this case, although the failure DDC201 instructs the control IC 30 to reduce the output current by the function of the output overcurrent protection circuit 22, the output current actually increases due to the control IC failure as described above. The operation of narrowing down is not performed (step S82). For this reason, the output overcurrent state is maintained, and as a result, the input overcurrent protection circuit 23 also operates (step S83).

[0050] As a result, in the failure DDC201, the empty power supply to the inverter unit 10 is cut off as described above, and the power supply to the control IC 30 is also cut off (steps S84, S85, S87). As a result, the on state of FET 21 that was always on in the failure D DC201 is released as described above (step S88).

[0051] The failure in which the power supply from the inverter unit 10 is cut off by the input overcurrent protection circuit 23 in this way does not cause the transformer 40 to saturate in the DDC201, and as a result, no drawing current occurs.

[0052] On the other hand, in the normal DDC 202, the output overcurrent protection circuit 22 is activated so that the output current is reduced by PWM control (steps S91, S92). That is, in this normal DDC202, an overcurrent state occurs until the input overcurrent protection circuit 23 of the fault DDC201 is activated (step S83). The output droop state is maintained by the output overcurrent protection circuit 22 working. In this case, the output overcurrent protection circuit 22 is activated while the output voltage rises, so the input current is reduced and the input overcurrent protection circuit 23 is not activated (step S93). Therefore, the output is not stopped (step S94). .

[0053] As described above, when the input overcurrent protection circuit 23 of the failure DDC201 is activated and the failure DDC201 is disconnected from the parallel operation system, the normal DDC202 is released from the overcurrent state force, and the operation proceeds to the normal operation. The power supply to is resumed (step S95).

In this case, the control unit 100 receives a failure signal from the failure DDC 201 (step S86) and detects an alarm (step S71). In this case, unlike the case of step S32 in FIG. 5, since the detection of the stop of only one unit (DDC201) is detected, the above “re-injection processing by alarm detection in multiple DDCs” is not performed (step S72).

In this embodiment, for convenience of explanation, two DDCs DDC201 and 202 for parallel redundant operation are used.

Although the description has been given assuming that (N = 1), the present invention can be similarly applied even when there are three or more units. In that case, there will be multiple normal DDCs 202. Each of the plurality of normal DDCs 202 has the same configuration as the one normal DDC 202 described above, and performs the same operation as the operation of the one normal 202 described above.

Claims

The scope of the claims

[1] A method for controlling a power supply device that supplies power by operating a plurality of power supply devices in parallel,

When an overcurrent state occurs due to the failure of at least one of the power supply units in parallel operation, and the overcurrent state also occurs in other power supply units due to the failure, a plurality of power supply units in parallel operation are A stage of stopping operation,

When restarting the operation of the multiple power supply units, the faulty device is stopped by the overcurrent protection function before the normal device, while the other faulty devices are stopped until the faulty device stops operating. A method for controlling a power supply apparatus, comprising a step of preventing the operation from being stopped depending on the influence of a corresponding failure.

[2] When parallel operation of multiple power supply units including a faulty device is resumed, the faulty device does not function to control the output current due to the fault, resulting in an output overcurrent state, resulting in an input overcurrent state. Causes the malfunctioning device to stop operating,

On the other hand, in the normal device, the force that increases the output current due to the current drawing by the faulty device is controlled by the output current control function, so that the increase in the output current is suppressed, so that the input overcurrent does not exceed the input current. 2. The control of the power supply device according to claim 1, wherein the operation is stopped due to detection of the power supply, and then the operation of the failed device is stopped so that the cause of the excess of the output current is resolved and then the normal operation state is reached. Method.

[3] The fault that causes the overcurrent is a fault in which the switching element that constitutes the synchronous rectifier circuit that controls the output current by PWM control is always turned on. The method for controlling a power supply device according to claim 1 or 2, wherein the constant-on state of the switching element is canceled by stopping the switching element.

[4] The plurality of power supply devices that are operated in parallel each has an input overcurrent protection unit having a function of stopping the operation of the synchronous rectifier circuit when the input current value exceeds a predetermined value;

The output overcurrent protection unit according to any one of claims 1 to 3, further comprising an output overcurrent protection unit having a function of reducing the output current by PWM control when the output current value exceeds a predetermined value. Control method of power supply.

[5] When the operation resumes after the failure until the output voltage of each power supply rises, the input current is lower than that during steady operation due to the low output voltage. The output current protection unit functions before the operation of the protection unit, and the function of the output current protection unit suppresses the increase in output current due to current drawing by the faulty device, and as a result, the increase in input current is also suppressed and input overcurrent protection is suppressed. On the other hand, in the case of a faulty device, the function of the output current protection unit does not work effectively due to the failure, and as a result, the output current cannot be reduced and the input current cannot be suppressed. The method for controlling a power supply device according to any one of claims 1 to 4, wherein the operation is stopped by the input overcurrent protection unit.

6. The method of controlling a power supply device according to any one of claims 1 to 5, wherein the plurality of power supply devices that are operated in parallel comprise DC-DC converters.

[7] A power supply device that supplies power by operating a plurality of power supply devices in parallel, and at least one of the plurality of power supply devices in parallel operation has caused an overcurrent state, and the influence thereof causes another In this power supply unit, when an overcurrent condition occurs, the power supply unit is configured to stop operation of a plurality of power supply units in parallel operation.

When restarting the operation of the multiple power supply units, the faulty device is stopped by the overcurrent protection function before the normal device, while the other faulty devices are stopped until the faulty device stops operating. A power supply unit configured to perform control so that operation is not stopped depending on the influence of the corresponding failure.

[8] When resuming parallel operation of multiple power supply units including a faulty device, the faulty device does not function to control the output current due to the fault, resulting in an output overcurrent state, resulting in an input overcurrent state. Causes the malfunctioning device to stop operating,

On the other hand, in the normal device, the force that increases the output current due to the current drawing by the faulty device is controlled by the output current control function, so that the increase in the output current is suppressed, so that the input overcurrent does not exceed the input current. 8. The power supply device according to claim 7, wherein the power supply device is configured to reach a normal operation state after the cause of the excess of the output current is resolved by the operation stop of the faulty device after the operation stop due to detection of the failure.

[9] The fault that causes the overcurrent is a fault in which the switching element that constitutes the synchronous rectifier circuit that controls the output current by PWM control is always turned on, and the control power supply of the synchronous rectifier circuit is supplied by stopping the operation. The power supply device according to claim 7 or 8, wherein the power supply device is configured such that the normally on state of the switching element is canceled by stopping the switching.

[10] The plurality of power supply devices that are operated in parallel each have an input overcurrent protection unit having a function of stopping the operation of the synchronous rectifier circuit when the input current value exceeds a predetermined value;

The power supply device according to any one of claims 7 to 9, further comprising an output overcurrent protection unit having a function of reducing the output current by PWM control when the output current value exceeds a predetermined value.

[11] When the operation restarts after the failure until the output voltage of each power supply rises, the input voltage is lower than that during steady operation due to the low output voltage. The output current protection unit functions before the operation of the protection unit, and the function of the output current protection unit suppresses the increase in output current due to current drawing by the faulty device, and as a result, the increase in input current is also suppressed and input overcurrent protection is suppressed. On the other hand, in the case of a faulty device, the function of the output current protection unit does not work effectively due to the failure, and as a result, the output current cannot be reduced and the input current cannot be suppressed. The power supply device according to any one of claims 7 to 10, wherein the power supply device is configured to stop operation by an input overcurrent protection unit.

12. The power supply device according to any one of claims 7 to 11, wherein the plurality of power supply devices that are operated in parallel comprise DC-DC converters.