GB2386266A

GB2386266A - DC-DC converter control

Info

Publication number: GB2386266A
Application number: GB0226626A
Authority: GB
Inventors: Victor Guijarro
Original assignee: ADVANCED POWER CONVERSION PLC
Current assignee: ADVANCED POWER CONVERSION PLC
Priority date: 2001-11-17
Filing date: 2002-11-14
Publication date: 2003-09-10
Also published as: GB0127593D0; GB0226626D0; WO2003044937A3; AU2002339168A1; WO2003044937A2

Abstract

The primary side of a DC-to-DC converter produces alternating positive and negative pulses separated by quiescent periods, coupled through a transformer TX to the secondary side, which comprises a bridge circuit of two transistor switches Q1, Q2 and two inductors L1, L2 with the secondary winding STX connected between the two transistor-inductor junctions, the switches being controlled from the transformer. Each of the switches Q1, Q2 is controlled from separate control winding means DX1, DX2 on the transformer via a diode D1, D2 and a control transistor Q3, Q4. Alternatively, a single control winding and further diode coupling circuitry may be used.

Description

1 2386266

DC-to-DC Converters 5 FIELD OF THE INVENTION

The present invention relates to DC-to-DC converters, particularly dynamic converters known as synchronous rectifiers that employ alternately charging and discharging capacitors or batteries to generate an 10 alternating current, which induces another alternating current using a transformer, this alternating current being converted back to a direct current by a switching circuit.

BACKGROUND OF THE INVENTION

Transformers used in electrical and electronic applications for transforming' a DC input voltage to a higher or lower voltage (and often referred to as "Buck" and "Boost" converters respectively) are well known to persons skilled in the art. A problem with known transformers is to 20 provide efficient assemblies which operate with both continuous input and output currents.

Typically DC-to-DC converters, for example using the circuits shown in Seiersen/ Scanpower US 4 899 271, include an input circuit 25 which generates a square wave AC signal from a DC signal, for example by using two capacitors whose charging and discharging are controlled by switches, the AC output being put across the primary transformer coil. A similar AC current is thereby induced in the secondary transformer coil, the transformer coil ratio and other factors being chosen such that the 30 converter circuit modifies the voltage in the desired way. The induced

current is then rectified (for example using one of the diode rectifier arrangements shown in that patent) and smoothed to produce the DC output. 5 More specifically, the secondary side of a known converter of this type is effectively a bridge circuit. Two adjacent arms of the bridge include respective MOSFET transistors, and the other two arms include inductors. The transformer secondary is connected between the transistor transistor and inductor-inductor junctions, and the output is taken across 10 the two transistor-inductor junctions. An output capacitor C1 smooths the rectified current to a substantially constant direct current.

This circuit operates in a generally four-stage cycle, with the capacitors in the primary side being charged and discharged to generate a 15 waveform in the primary side of the transformer consisting of a cycle of a positive pulse, a quiescent period, a negative pulse, and a quiescent period.

(A more detailed description of the circuit's mode of operation may be

found in "A new efficient high frequency rectifier circuit", 1991 Jun.

HFPC Proceeding, by C. Peng, M. Hannigan and 0. Seiersen.) Control circuitry controls the MOSFETs to act as switches operating in time with the signal induced across the secondary transformer coil to rectify the AC signal to a DC signal; this operation is known as synchronous rectification. In its simplest form, the control circuitry 25 consists of a connection to each MOSFET from the appropriate end of the secondary winding of the transformer. One MOSFET is turned on during the positive pulse; the other is turned on during the negative pulse. During the quiescent periods, the MOSFETs are turned off; the inductors continue to supply current to the load, the currents flowing through the parasitic 30 antibody diodes of the MOSFETs, this being known as 'freewheeling'.

The antibody diode of a MOSFET has a relatively high resistance, so this part of the switching cycle is relatively inefficient.

5 Another disadvantage of this type of circuit, particularly if the freewheeling time is very short or is dispensed with entirely, is that the MOSFETs' switching signals may overlap, causing both to be open at the same time. This short-circuits the secondary coil, which is inefficient, and can damage the converter.

One way of avoiding these disadvantages is to use a separate control circuit to drive the MOSFETs. This method is known as "control-driven" rather than "self-driven". This however increases the cost and complexity of the circuit, particularly as it must be synchronised with the primary 15 circuit's switches.

It is an object of the present invention to provide a DC-to-DC converter and method of driving it to efficiently produce a continuous output current for a continuous input current.

SUMMARY OF THE INVENTION

According to the present invention, there is provided a DC-to-DC converter comprising a primary side supplying alternate positive and 25 negative current pulses to a transformer which is in turn coupled to a secondary side, the secondary side comprising a series pair of inductors and a series pair of transistor switching elements connected in parallel across the transformer, the output being connected between the junction between the inductors and

the junction between the switching elements, and control circuitry coupling the transformer to the switching elements, characterized in that the control circuitry comprises control winding means connected in series with diode means between the 5 gates of the switching and control transistors, and, for each switching element, a control transistor connected between the gate and base of the switching element.

BRIEF DESCRIPTION OF THE DRAWINGS

Various DC-to-DC converters embodying the invention will now be described, by way of example, with reference to the accompanying drawings, in which: Fig, l shows a prior art primary coil and secondary circuit;

15 Fig. 2 shows a primary coil and an embodiment of the secondary circuit; Fig. 3 shows a primary coil and a variant of the secondary circuit; Fig. 4 shows various waveforms associated with the Fig. 2 circuit; Fig. 5 shows a primary coil and a further variant of the secondary 20 circuit; and Fig. 6 shows a primary circuit and a further variant of the secondary circuit. DESCRIPTION OF THE PREFERRED EMBODIMENTS

The primary coil and secondary circuits of a known converter are shown in Fig. 1 (the other elements of the primary circuit, which are conventional, are not shown).

The secondary coil STX is connected in series between two inductors L1, L2. Two Field Effect Transistors (MOSFETs) Q1, Q2

(shown here with their parasitic capacitances Cql, Cq2 and parasitic antibody diodes Dql, Dq2) are connected in parallel with the secondary 5 coil STX. These MOSFETs act as switches operating in time with the signal induced across the secondary transformer coil to rectify the AC signal to a DC signal; this operation is known as synchronous rectification.

The output voltage is drawn off from between the two MOSFETs Ql, Q2 and the coupled terminals of the inductors L1, L2. The capacitor C1 10 smoothes the rectified current to a substantially constant direct current.

The gates of the MOSFETs Q1, Q2 are coupled to the secondary-: coil STX as shown. This converter operates in a generally four stage cycle The secondary winding waveform is as shown at Vdx in Fig. 4.

In the first stage of the cycle, a first primary capacitor is discharged through the primary coil. As the capacitor discharges through the primary coil PTX, it produces a potential difference across the primary coil such that the lower connection of the coil is, say, positive with respect to the 20 upper connection; this induces a potential difference in the same direction across the secondary coil STX. Therefore MOSFET Q1 is switched off while MOSFET Q2 is switched on. The current flows from coil STX through inductor L1 to the load R3 and back to the coil STX via MOSFET Q2. In the second stage the first capacitor is set to be recharged while the primary coil is clamped to zero voltage, and so the secondary coil is similarly clamped, switching both the MOSFETs off. The inductor Ll continues to supply current to the load, the current flowing through the 30 parasitic antibody diode Dq2, this being known as 'freewheeling'.

In the third stage the second primary capacitor is discharged through the primary coil such that the lower terminal of the primary coil PTX is negative with respect to the upper terminal, inducing a similar voltage 5 across the secondary coil STX. In a similar manner to the first stage of the switching cycle, MOSFET Ql is switched on while MOSFET Q2 remains off, and current flows from earth through MOSFET Q1, through the secondary coil and inductor L2 and to the load R3.

10 In the fourth stage, the primary coil PTX is again has zero volts clamped across it, similarly clamping the secondary coil, and turning off both MOSFETs Q1, Q2. As for stage 2 of the switching cycle, the inductor L2 continues to cause current to flow around the secondary circuit, via the antibody diode Dql of MOSFET Q3.

During stages 2 and 4 of the switching cycle, where the circuit is freewheeling, MOSFETs Q1 and Q2 are conducting in turn through their respective parasitic antibody diodes.

20 Referring to Fig. 2, a preferred form of the converter comprises an input connected to a primary circuit, and an output connected to a secondary circuit, the primary and secondary circuits being coupled by a transformer having a primary coil connected to the primary circuit and a secondary coil connected to the secondary circuit. The primary circuit 25 converts direct current into an alternating current, and provides this current across the primary coil so that a similar current is induced in the secondary coil (only the primary coil of the primary circuit is shown in the drawings).

In the secondary circuit, two inductors L1 and L2 are connected in 30 series across the secondary coil of the transformer, the junction of the

inductors forming terminal of the output. Two opposing MOSFETs are connected in series across the secondary coil in parallel with the two inductors, the junction between the MOSFETs forming the other terminal of the output. A capacitor is connected across the output terminals to 5 smooth the output.

Also included in the secondary circuit are two drive coils DX1, DX2 in series closely coupled to the transformer. (These two coils can be regarded as a single coil with an earthed centre tap.) Each outer terminal 10 of each coil DX1, DX2 (the outer terminal being the terminal not connected to the other coil) controls the gate of a respective MOSFET Q3, Q4. The drain of each of MOSFETs Q3, Q4 is connected to the outer terminal of DX2, DX1 respectively via a diode and resistor as shown, the diode biased to allow current to flow to the drain. The source terminal of 15 each MOSFET is earthed, as is the junction between the two drive coils.

The drains of MOSFETs Q3, Q4 are connected to the gates of MOSFETs Q1, Q2 respectively. As for the known circuit, the sources of MOSFETs Q1, Q2 are earthed. The drain of MOSFET Q1 is connected to 20 the upper terminal of the secondary coil SIX, and also to inductor L1; similarly, the drain of MOSFET Q2 is connected to the lower terminal of the secondary coil SIX, and also to inductor L2. These inductors are then coupled to form one terminal of the output, the load being connected between the output and earth. As in the known circuit, a capacitor is 25 connected in parallel to the load.

The MOSFETs Q1 and Q2 are shown here with their parasitic capacitances Cql and Cq2 and parasitic antibody diodes D1 and D2 respectively.

As the capacitor in the primary circuit discharges, resulting in a voltage across the primary coil (say such that the lower terminal is positive), potential differences in this direction will be induced across the secondary coil and the upper and lower drive coils. The lower terminal of 5 the lower drive coil is at a positive voltage, turning MOSFET Q4 on, and earthing the capacitance Cq2, which discharges. When the capacitor is discharged, the input to MOSFET Q2 is zero and it switches off. At the same time, the current flows through the diode Dl to maintain the input of MOSFET Ql in an 'on' state and charge the capacitance Cql. Also, the 10 upper terminal of the upper drive coil is at a negative potential, turning MOSFET Q3 off. No current flows through D2. The voltage waveform of the drive winding and the resulting gate voltages for MOSFETs Ql, Q2 are shown in Fig. 4. (The lower drive coil's lower terminal will be similar but inverted.) Like the drive coils, the secondary coil STX has a potential difference induced across it. Current flows from the earth through MOSFET Ql from source to drain. From here the current flows through inductor L1 to the load R3, as well as from the secondary coil through L2 20 and again to the load R3.

During stage 2 of the converter's cycle, the voltages across the secondary coil and the driving coils are clamped at zero. Accordingly, MOSFETs Q3 and Q4 are switched off. The gate of MOSFET Q1 is held 25 at a positive value by the capacitance Cql, and therefore MOSFET Ql remains on. Inductor Ll maintains current flow through MOSFET Ql and load R3.

When the second capacitor is discharged in the primary circuit, the 30 upper terminals of the secondary coil and the upper and lower drive coils

will be made positive relative to their lower ends. Since the upper terminal of the upper drive coil is at a positive voltage, MOSFET Q3 will be turned on, discharging the capacitance Cql. No current flows through diode D1, and the gate of MOSFET Q1 is zero, turning MOSFET Q1 off.

5 MOSFET Q4 is switched off by the negative potential of the lower drive coil's lower terminal. Current flows through the diode D2 charging the capacitance Cq2 and switching MOSFET Q2 on.

The potential difference across the secondary coil STX causes 10 current to flow through MOSFET Q2, inductor L2 and the load R3.

Current also flows from the upper terminal of the secondary transformer coil through inductor L 1 to the load R3.

During the fourth stage of the converter's cycle, the coils are again 15 clamped at zero volts, and MOSFETs Q3 and Q4 both off. The gate of MOSFET Q2 is held positive by the capacitance Cq2, and therefore MOSFET Q2 remains on. Inductor L2 causes current to flow the earth through MOSFET Q2 from source to drain and out to the load R3.

20 It will be seen that the use of two MOSFETs Q3 and Q4 and diodes D1 and D2 to control the gates of the MOSFETs Q1 and Q2 by utilising the parasitic capacitance of MOSFETs Q1 and Q2 ensures that MOSFETs Q1 and Q2 remain open even when the potential across the secondary coil is clamped at zero volts, so that the inductors L1 and L2 are permitted to 25 induce current to flow during the freewheeling periods. In utilising the synchronous rectifiers rather than their anti-parallel body diodes during the freewheeling period, the conversion is more efficient.

If the conditions are such that MOSFETs Q3 and Q4 are on 30 simultaneously, this could form a short circuit across the secondary coil.

However, it will also be seen that here, this causes the gates of MOSFETs Q1 and Q2 to both be earthed. Hence MOSFETs Q1 and Q2 are switched off and no short circuit across the secondary coil is formed.

5 Depending upon factors such as the switching frequency of the circuit and the MOSFETs used, the parasitic capacitance across the gate and source of the MOSFETs Q1, Q2 could be supplemented with discrete capacitors. 10 The transistors shown here are e-channel enhancement mode MOSFETs, but of course other MOSFETs, or other types of switch such as npn bipolar transistors could be used in a similar manner. Further, MOSFET pairs could be replaced by two switching elements which shared the characteristics of being switched on to conduct in response to a first 15 signal, whilst requiring a second input to actively switch them off.

Referring to Fig. 3, rather than using two driving coils, a single driving coil DX may be coupled with the secondary coil, the upper and lower terminals being connected to MOSFETs Q3 and Q4 respectively, 20 while the upper and lower terminals are also connected to the gates of MOSFETs Q4 and Q3 respectively (again via diode and resistor pairs D1, R1 and D2, R2). The upper and lower terminals of the driving coil are earthed when a negative potential is induced using diodes D3 and D4. It will be seen that this circuit operates in an equivalent manner to the two 25 driving coil circuit of Fig. 2. (Doides D3 and D4 effectively replace the centre tap of the coils DX1, DX2 of Fig. 2 when those coils are regarded as a single centre-tapped coil.) By using suitable switching elements, the potential from a single 30 drive coil, or even the secondary coil itself, could be used to generate the

potentials which are increased (or decreased) for the switches to operate during some or all of the freewheeling period. Using potentials tapped from drive windings gives cleaner signals than signals derived from the secondary coil, as this will include spikes generated by the switches, and 5 the drive windings will not have the secondary coil's voltage drop caused by the transformer impedance and which cause delays in turning off the switches. Also, the number of turns of the drive windings may be selected independently of the secondary coil to obtain correct gate voltages.

10 Rather than the current doubler secondary circuit shown here, the inventive principles described herein could of course be applied to other topologies with synchronous rectifier circuits, such as forward, pushpull, half bridge, full bridge, and other voltage or current multipliers.

15 Figs. 5 and 6 show further variants for the control windings DX1 and DX2 generating the control signals. In Fig. 5, the windings DX1 and DX2 are coupled to the inductors L1 and L2 respectively instead of to the main secondary winding STX. In Fig. 6, the windings DX1 and DX2 are coupled to the primary circuit inductors Lip and L2p respectively. These 20 circuits operate in essentially the same way as the circuits of Figs. 2 and 3.

Claims

Claims

5 1 A DC-to-DC converter comprising a primary side supplying alternate positive and negative current pulses to a transformer which is in turn coupled to a secondary side, the secondary side comprising a series pair of inductors and a series pair of transistor switching elements connected in parallel across the transformer, 10 the output being connected between the junction between the inductors and the junction between the switching elements, and control circuitry coupling the transformer to the switching elements, characterized in that the control circuitry comprises control winding means connected in series with diode means between the 15 gates of the switching and control transistors, and, for each switching element, a control transistor connected between the gate and base of the switching element.
2 A converter according to claim 1 wherein the control circuitry 20 includes a first control MOSFET the gate of which is coupled with a first potential induced from the transformer, and the drain of which is coupled to a second potential induced from transformer via the diode means.
3 A converter according to claim 2 wherein the control circuitry 25 includes a second control MOSFET the gate of which is coupled to the first potential, and the drain of which is coupled to the second potential via the diode means.
4 A converter according to either of claims 2 or 3 wherein the first 30 and second potentials are provided by a common control coil.

5 A converter according to any of claims 2 to 3 wherein the first and second potentials are provided by respective control coils.
5
6 A converter according to any previous claim wherein the alternating current supplied across the primary coil is generated by a direct current input to the primary side.
7 A secondary circuit according to any previous claim.
8 A circuit substantially as herein described and illustrated with reference to Fig. 2 or 3.
9 Any novel and inventive feature or combination of features 15 specifically disclosed herein within the meaning of Article 4H of the International Convention (Paris Convention).