EP1303903A1

EP1303903A1 - A dc switching regulator

Info

Publication number: EP1303903A1
Application number: EP01949748A
Authority: EP
Inventors: Isaac Cohen; Andrew Skinner
Original assignee: Coutant Lambda Ltd
Current assignee: TDK Lambda UK Ltd
Priority date: 2000-07-18
Filing date: 2001-07-18
Publication date: 2003-04-23
Also published as: GB0017630D0; GB2370655B; GB2370655A; WO2002007296A1

Abstract

A DC switching regulator including a transformer (306) having a primary winding (308) and a secondary winding (310) a primary switch (302) connected in series with the primary winding (308) and having an ON condition and an OFF condition in which conditions the primary winding (308) is respectively connected and not connected across a power source (312). A storage capacitor (314) is arranged to absorb magnetization and leakage energy from the primary winding (3089 when the primary switch (302) is in said OFF condition. A unidirectional conducting device (313) is connected between the primary winding (308) and the storage capacitor (314) so as to prevent energy from the storage capacitor (314) passing back to the primary winding (308). An auxiliary switch (316) is connected in series with the storage capacitor (314) and has an ON condition and an OFF condition. An inductor (320) is arranged to absorb energy from the storage capacitor (314) while the auxiliary switch (316) is in the ON condition. This regulator has a resetting function for the main transformer (306) which prevents negative saturation of its core while providing zero volt switching of the primary switch (302) and of the auxiliary switch (316).

Description

A DC SWITCHING REGULATOR

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to DC switching regulators. These are regulators in which electrical power from a power source is delivered to an output circuit by repeatedly connecting and disconnecting the primary winding of a transformer to the power source, the output circuit being connected to a secondary winding of the transformer.

2. Background to the Invention

The present invention is particularly concerned with DC switching regulators which include circuitry to reset the transformer during the period the primary winding is disconnected from the voltage source. One known approach to providing this transformer resetting function is to use an active clamp, an example of which is shown in Figure 1 and is described in US reissued patent RE 36,098.

This regulator has a primary switch 102, controlled by a primary switch controller 104, connected in series with a transformer 106, having a primary winding 108 and a secondary winding 1 10, connected across a voltage source 1 12. A storage capacitor 114 and an auxiliary switch 1 16, controlled by an auxiliary switch controller 1 18, are connected in series across the primary winding 108.

The auxiliary switch controller 1 18 opens the auxiliary switch 1 16 when the primary switch 102 is closed and closes the auxiliary switch 1 16 when the primary switch 102 is opened. On opening the primary switch 102 magnetizing and leakage energy (associated with the transformer 106 and its leakage inductance) is transferred to the storage capacitor 1 14, which has a value large enough to provide that its voltage remains substantially constant at a value established over the preceding energy conversion cycles. The primary winding 108 is clamped to this voltage of the storage capacitor 114 during the resetting period which, in a steady state, automatically assumes a value which will reset the transformer over the open period of the primary switch 102. One of the benefits of this active clamp approach to resetting the transformer core is that zero voltage switching of the primary and auxiliary switches 102, 1 16 is possible by, typically, driving them out of phase with a small amount of deadtime. This leads to higher efficiencies and the possibility of higher frequency operation with known associated advantages.

One aspect of this topology, however, is that under large signal transient conditions a DC bias can be established in the transformer 106 which prevents zero voltage switching and allows hard commutation of the anti-parallel freewheel diodes 120, 122 associated with serni-cόriductor primary and secondary switches. If MOSFETS are used as the primary and secondary switches, the hard commutation of the body diodes can be destructive. Such large signal transients can be caused, for example, by an output short-circuit or by the rapid reduction in the duty-cycle of the primary switch 102 caused by pulse-by-pulse current limiting. A further aspect is that the DC-bias set up by a large-signal reduction in duty-cycle can also lead to negative saturation of the transformer core with resultant high currents.

A known approach to addressing these aspects is to incorporate a unidirectional conducting device, for example a diode, between the primary winding 108 and the demagnetisation circuit which is oriented to permit energy to pass from the transformer to the demagnetisation circuit so resetting the transformer but preventing energy flow in the reverse direction thereby preventing the possibility of negative saturation of the transformer. Examples of such arrangements are disclosed in US-A-4,736,285 and US-A- 6,01 1 ,702.

In the circuit of US-A-4,736,285, shown in Figure 2, a clamping storage capacitor 202 is connected to the primary winding 108 by a diode 204. The diode 204 is also connected to an inductor 206 and an auxiliary switch 208 in parallel with the primary winding. A diode 210 is connected in series with the inductor 206 across the power supply 1 12. The auxiliary switch 208 is placed in the ON condition when primary switch 102 is put in the OFF condition by virtue of a demagnetization winding 212 that is connected to the auxiliary switch 208 by a pulse shaping circuit (not shown).

When the primary switch 102 is placed in the OFF condition, and auxiliary switch 208 in the ON condition, the magnetization energy from the core of the transformer 106 and energy associated with leakage inductance is transferred to the storage capacitor 202 via diode 204. When auxiliary switch 208 is ON, the storage capacitor 202 causes a current to build up in the inductor 206 so extracting energy from the capacitor which is stored in inductor 206.

When the primary switch 102 is turned ON again, the core of the transformer 106 will, in steady state operation, have been reset. The auxiliary switch 208 is now switched to the OFF condition and the energy that was stored in the inductor 206 is discharged back into the voltage source 1 12 by virtue of the diode 210.

This arrangement has the benefit that it cannot hard commutate the body diode of either the primary switch 102 or auxiliary switch 208. Under transient conditions the diode 204 can be hard commutated and so should be a fast recovery type capable of withstanding this mode of operation. In particular, in the event of an output short-circuit and resultant reduction in duty-cycle this circuit will not result in negative saturation of the main transformer 106. However, the circuit is not capable of zero voltage switching and so suffers from the various known limitations this entails. SUMMARY OF THE INVENTION

According to the present invention a DC switching regulator includes a transformer having a primary winding and a secondary winding and a primary switch connected in series with the primary winding having an ON condition and an OFF condition in which conditions the primary winding is respectively connected and not connected across a power source. A storage capacitor is arranged to absorb magnetization and leakage energy from the primary winding when the primary switch is in the OFF condition. A unidirectional conducting device is connected between the primary winding and the storage capacitor and arranged to prevent energy from the storage capacitor passing back to said primary winding. An auxiliary switch is connected in series with the storage capacitor and have an ON condition and an OFF condition and an inductor is arranged to absorb energy from said storage capacitor while sai auxiliary switch is in said ON condition.

The present invention can provide a DC switching regulator in wnicn zero voltage switching can be achieved whilst obtaining optimum reset of the main transformer and which avoids the possibility of negative saturation of the main transformer. The unidirectional device prevents negative saturation of the core of the transformer. When the primary switch is turned to the OFF condition current will flow initially though the anti-parallel diode associated with the auxiliary switch. This ensures zero-volt switching ON of the auxiliary switch when effected a short time after the primary switch is turned to the OFF condition. When the auxiliary switch is later turned to the OFF condition the output capacitance of the primary switch is charged up by virtue of the inductor. This provides for zero-volt switching of the primary switch to the ON condition when effected a short time after the auxiliary switch has been turned to the OFF condition.

The storage capacitor and auxiliary switch may be connected in parallel with the inductor. The unidirectional conducting device preferably includes a diode. The primary switch and/or the auxiliary switch may be devices incorporating internal or co-packaged anti-parallel freewheel diodes such as MOSFETS.

The inductor may be selected such that the ratio between the saturated inductance of the inductor and the unsaturated inductance of the inductor prevents excess current through the auxiliary switch.

The present invention also encompasses a switch mode power supply including a DC switching regulator according to the present invention. Such power supplies include, for example, DC-DC convertors, particularly high density DC-DC convertors, and multiple output AC-DC power supplies for industrial applications. Because the power supplies of the present invention are efficient at recovering transformer leakage energy efficiently, they are particularly suitable for powering medical apparatus which generally require stringent approvals in terms of high isolation.

The present invention also encompasses a method of resetting a main transformer including a primary winding and a secondary winding and the primary winding being connected in series with a primary switch, having an ON and an OFF condition, across a power source. The method includes the steps of passing magnetisation energy and leakage from said primary coil to at least a capacitor in series with an auxiliary switch having an ON and an OFF condition whilst said primary switch is in the OFF condition via a unidirectional conducting device, and passing the energy stored in the storage capacitor to an inductor also during said OFF condition of the primary switch.

Objects, features and advantages of embodiments of the present invention will be apparent from the following detailed description of exemplary embodiments of the present invention taken in conjunction with the accompanying drawings in which common components have been given common reference numerals. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic description of a prior art DC switching regulator;

Figure 2 is a schematic diagram of a further prior art DC switching regulator;

Figure 3 is a schematic diagram of a first embodiment of a DC switching regulator according to the present invention;

Figure 4 is a schematic diagram of a further embodiment of a DC switching regulator according to the present invention;

Figure 5 is a schematic diagram of a further embodiment of a DC switching regulator according to the present invention, in which the storage capacitor is provided with a bias resistor; and

■ Figure 6 is a pair of graphs showing the transformer magnetization current and reset inductor current of a further described embodiment.

DETAILED DESCRIPTION OF THE INVENTION

Referring to Figure 3 of the drawings, a power supply 300 includes a DC switching regulator 300A and an output circuit 300B. The DC switching regulator includes a main transformer 306 having a primary switch 302 controlled by a primary switch controller 304 in series with a primary winding 308 of the transformer 306 also having a secondary winding 310. The primary winding 308 and primary switch are connected across a power source 312.

A diode 313 connects the primary winding 308 to a demagnetising/reset circuit comprising a storage capacitor 314 in series with an auxiliary switch 316 controlled by an auxiliary switch controller 318 connected in parallel with an inductor 320.

Each switch 302, 316 (and the equivalent switches of the below described embodiments) must have associated with it an anti-parallel freewheel diode D which may be inherent in the switch structure (eg if a MOSFET) or added as a discrete component.

The steady state operation of the embodiment of Figure 3 is as follows. Assuming the primary switch 302 has reached the end of its ON period it will be turned to its OFF condition by primary controller 304, the auxiliary switch being in the OFF condition. The storage capacitor 314 will have been charged up during previous switching cycles of the converter to a clamping voltage which will fully reset the main transformer 306 during the forthcoming OFF period of the primary switch. The value of the storage capacitor 314 is chosen such that the voltage across it remains substantially constant, during operation of the regulator.

During this OFF period of the primary switch 302 (simply OFF, period for brevity), magnetising energy stored in the transformer 306 and -energy stored in the leakage inductance of transformer 306 is delivered to the storage capacitor 314. Current flowing through the anti-parallel diode of auxiliary switch 316 results in a near zero voltage across the auxiliary switch 316 even though in the OFF condition. The auxiliary switch 316 can therefore be switched to the ON condition with zero voltage switching a short time after the primary switch 302 is turned to the OFF condition.

Energy will then, during this OFF period, be transferred from the storage capacitor 314 to the inductor 320.

By the end of the OFF period, if the inductor 320 is suitably selected in known manner, the current in the transformer 306 will have been reduced to zero prior to the auxiliary switch 316 being turned to the OFF condition. The inductor 320 then causes a current flow through diode 313, which discharges the output capacitance of the primary switch 302. The primary switch 302 can then be turned to the ON condition (starting the ON period of the regulator) under zero voltage switching conditions. The primary switch 302 is switched to the ON condition with a short delay after the auxiliary switch 316 is switched to the OFF condition to avoid short-through.

The energy stored in the inductor 320 during the OFF period is then recovered to the power source or transferred to the load via the transformer 306 during the ON period.

Assuming that the primary and auxiliary switches operate at duty- cycles of D and (1-D) respectively, the steady state voltage on the storage capacitor for volt-seconds balance for the two inductive components will be: (1 ) Vc = Vin.D/{1 - D)

A charge balance also exists where the charge delivered to the capacitor during the OFF period from the transformer 306 equals the charge extracted into the inductor 320 during the same period. When suitably selected the inductor 320 will have a variable dc bias which maintains this charge balance with changes in converter loading and. hence changes in the ramdunt of leakage energy delivered from the transformer to capacitor 314 during the OFF period.

Under transient conditions, charge balance in the capacitor 314 and volts-second balance in the transformer 306 and inductor 320 will not exist. It will be clear to someone skilled in the art that under large signal transient conditions a DC bias can be generated in transformer 306 for increasing duty-cycle D or in the inductor 320 for reducing duty-cycle which could prevent zero-voltage switching conditions from existing whereby one of the switches could be switched into the ON condition with current flowing in the anti-parallel diode of the other switch. If diode 313 is sufficiently fast it can commutate whilst the anti-parallel diode is still conducting so preventing hard commutation of this diode and its resultant potential failure mode. This is especially advantageous where the anti-parallel diode is the body diode of a MOSFET.

Referring now to Figure 4, a power supply 400 is as the embodiment 300 of Figure 3 except that a storage capacitor 41 , an auxiliary switch 416 and an auxiliary switch controller 418 are positioned in series with the inductor 320 and parallel to the primary auxiliary switch 302 in the DC regulator 400B. The circuit of Figure 4 operates in an identical fashion to that of Figure 3.

Referring now to Figure 5, a power supply 500 is similar to that of Figure 3 except in a DC regulator 500B the positions of the auxiliary switch 316 and storage capacitor 314 are reversed, and the polarity of the voltage source 312 has been reversed as has the orientation of the diode 313. Further, there is an additional component, namely a resistor 510 in series with inductor 320 across the voltage source 312 to charge the storage capacitor 314 to the supply voltage prior to start-up when both the primary switch 302 and the auxiliary switch 316 will be in the OFF condition

This is a benefit in topologies using voltage feed forward control with little or no soft-start. In applications employing pre-regulation of the supply voltage to the regulator such as off-line converters employing power factor correction, the regulator may be constrained to start when the duty-cycle D is less or equal to 50% whereby the capacitor 320 is pre-charged to a level which guarantees full reset of transformer 106 during the OFF period.

The description above assumes that inductor 320 is suitably selected to enable zero-voltage switching of the two switches which will be dependent on the energy stored in the inductor 320 during the OFF period. Energy stored in the inductor 320 during the OFF period is required to discharge the output capacitance of the switches and supply energy to the leakage inductance of the transformer 306 during the switching interval between the OFF and ON periods if the transition from the OFF condition to the ON condition is to occur with zero-voltage switching of the main switch 302.

In forward converter circuits with high leakage inductance for example designs where the transformer 306 has relatively loose coupling between its primary winding 308 and secondary winding 310 then zero-voltage switching can be achieved over a wide range of load and line utilising a linear inductor with an inductance comparable to that of the magnetising inductance of transformer 306. In flyback designs, the inductor is only required to recover the leakage energy from the transformer 306 , a higher value of inductance would be selected.

Under large signal transient conditions in which the duty-cycle is rapidly reduced, diode 313 prevents negative saturation of the transformer 306 but the loss of energy balance between that of the reset circuit and main transformer causes an increase, from the steady state value, of the current in the inductor 320 which can lead to its saturation. By winding the inductor 320 on a low permeability core with a large number of turns, the ratio between its saturated and unsaturated inductances can be controlled so as to prevent excess current passing through the auxiliary switch 316 without recourse to additional current limit circuit protection for the auxiliary switch 316.

By selecting inductor 320 such that its characteristics during steady- state operation of the regulator remain substantially linear then the increased average current required in inductor 320 during the OFF period for increased regulator loading is achieved with increased dc bias. This increased current will tend to have less resistive heating effects in the winding of inductor 320 than a design, which generates high AC current such as where a non-linear inductor is used.

To achieve zero-voltage switching at turn-on of the main switch 302 in circuits with low levels of leakage inductance for example forward convertors with tight coupling between the primary winding 308 and secondary winding 310 the current in inductor 320 at the time that the auxiliary switch 316 switches into the OFF condition must of similar magnitude to the reflected load current and greater than the reflected load current where the leakage inductance tends towards zero.

Such current can be achieved by reducing the inductance of both the transformer and inductor so the peak value of the magnetizing current approaches the value of the magnetizing current approaches the value of the reflected load current.

Under such conditions, the magnetizing current will supply the load current as it gradually increases in the leakage inductance of the transformer as well as charging the snubbing capacitance of the primary switch. In such an arrangement: a large amount of inductive energy will circulate in the converter, resulting in increased conduction losses and requiring a physically large reset inductor; and the dV/dt on the transistors at the turn OFF of the auxiliary switch will vary significantly with load, making zero volt switching operation difficult when the load and line vary widely; and the increased magnetizing current will essentially double the turn OFF current and dV/dt of the primary switch thereby diminishing the zero volt switching effectiveness.

These characteristics can be eliminated, if desired, by winding the inductor on a saturable core that has a flat hysteresis loop. Such an inductor will have an inductance much larger than the magnetizing inductance of the transformer and at the start of the operation its current will not be sufficient to extract from the storage capacitor all the energy transferred from the magnetizing inductance of the transformer. Consequently, the voltage on the storage capacitor will tend to raise above the value given by equation (1 ) above and a net DC voltage will develop across the inductor.

This voltage will have a DC bias current in the inductor and eventually its core will start to saturate towards the end of the conduction cycle of the auxiliary switch. As a result, a narrow triangular pulse of current will develop which will keep increasing in amplitude until the average current flowing out of the storage capacitor through the inductor will equal to the average of the current flowing into the storage capacitor from the transformer and the voltage on the storage capacitor will reach the value predicted by equation (1 ). (See Figure 6).

If the saturated inductances of the inductor is Lsat, the following equation will apply at equilibrium.

(2) 1/2.Lm.((Vinmax/Lm).Tsw.Dmin)² = 1/2.Lsat.(lpk(L))²

At no load condition, we can obtain a rapid discharge of the snubbing capacitance to zero following the turn OFF of the auxiliary switch if we design for the following constraint:

Cs.Vinmax² = lpk(L)².Lsat (Cs being the snubbing capacitance).

By choosing Ipk(L) to be equal to the reflected load current, we can have equal dV/dt at turn ON and turn OFF and we can solve Eq. 2 and 3 for Lm and Lsat:

(3) Ipk(L) = (Ns/Np).lo

(4) Lsat = Cs.(Vinmax²/lo .Ns²)).Np²

(5) Lm - Tsw².(Dmin²/Cs)

When output current increases to its maximum, the storage capacitor will absorb the energy stored in the leakage inductance of the transformer in addition to the magnetizing energy. As a result, the reset inductor will have to extract the additional energy from storage capacitor and the amplitude of the current pulse will increase.

This increase in current if very desirable as the current in the inductor has to deliver the load current in addition to discharging the snubbing capacitance.

The optimal value for the transformer's leakage inductance LI is value equal to the saturated inductance Lsat of the reset inductor, but the circuit can operate at either higher or lower values if Lm and Lsat are adjusted appropriately.

It should be noted that in the circuit described here, unlike in many other zero-volt switching topologies, the time required to discharge to zero the snubbing capacitance prior to the turn ON of primary switch is essentially independent of line and load so it is possible to obtain zero-volt switching under all conditions with a constant delay time.

The time required to charge the snubbing capacitance following the turn OFF of the primary switch will vary widely with line and load, but since the auxiliary switch conducts through its body diode for half of the Tsw.(1-d) period, the delay time in the drive of the auxiliary switch can be made long enough to accommodate the variations.

The embodiments of the present invention described herein are intended to be taken in an illustrative and not a limiting sense. Various modifications and changes may be made in these embodiments by persons skilled in the art without departing from the scope of the present invention as defined in the appended claims.

Claims

1 claim:

1. A DC switching regulator including: a transformer having a primary winding and a secondary winding; a primary switch connected in series with said primary winding and having an ON condition and an OFF condition in which conditions said primary winding is respectively connected and not connected across a power source; a storage capacitor arranged to absorb magnetization and leakage energy from said primary winding when said primary switch is in said OFF condition; a unidirectional conducting device connected between said primary winding and said storage capacitor arranged to prevent energy from said storage capacitor passing back to said primary winding; an auxiliary switch connected in series with said storage capacitor and having an ON condition and an OFF condition; and an inductor arranged to absorb energy from said storage capacitor while said auxiliary switch is in said ON condition.

2 A DC switching regulator as claimed in claim 1 , in which said storage capacitor and auxiliary switch are connected in parallel with said inductor.

3. A DC switching regulator as claimed in claim 1 or 2, in which said energy storage capacitor is provided with a means to pre-charge said capacitor prior to operating the regulator.

4. A DC switching regulator as claimed in any preceding claim, in which said unidirectional conducting device includes a diode.

5. A DC switching circuit as claimed in any preceding claim, in which said primary switch and/or said auxiliary switch is a MOSFET.

6. A DC switching regulator as claimed in any preceding claim in which said inductor selected such that the ratio between the saturated inductance of said inductor and the unsaturated inductance of said inductor so as to prevent excess current through said auxiliary switch.

7. A DC switching regulator as claimed in any preceding claim in which said inductor has a core which saturates, in use, and has a substantially flat hysteresis curve.

8. A DC switching regulator as claimed in any preceding claim in which a resistor is used to pre-charge said energy storage capacitor prior to operating the said regulator.

9. A switch mode power supply including a DC switching regulator as claimed in any preceding claim and an output circuit.

10. A method of resetting the main transformer of DC switching regulator, said main transformer including a primary winding and a secondary winding and said primary winding being connected in series with a primary switch, having an ON and an OFF condition, across a power source, the method including the steps of: passing magnetization energy and leakage from said primary coil to at least a capacitor in series with an auxiliary switch having an ON and an OFF condition whilst said primary switch is in the OFF condition via a unidirectional conducting device; and passing the energy stored in said storage capacitor to an inductor also during said OFF condition of said primary switch.

1 1 A method as claimed in claim 10, in which at least some of the energy stored in said inductor is recovered to said power source and/or transferred to a load connected to said secondary windings during said ON condition of said primary switch.

12. A method as claimed in claim 10 or 11 in which said auxiliary switch is switched to said ON condition after said primary switch is switched to said OFF condition by way of zero voltage switching and said primary switch is switched to said ON condition after said auxiliary switch is switched to said OFF condition by way of zero volt switching.

13. A method as claimed in any one of claims 10 to 12, in which the inductor has a core having a flat hysteresis and which saturates during operation of the DC switch regulator.

14. A DC switching regulator substantially as hereinbefore described with reference to any of Figures 3 to 6 of the accompanying drawings.

15. A method of resetting the main transformer of a DC switching regulator substantially as hereinbefore describe with reference to any of Figures 3 to 6 of the accompanying drawings.