GB2087171A - Static inverter - Google Patents

Abstract

A static inverter using a high frequency transformer (13) with an alternating pulse width modulated square input (Figure 2D) voltage and capable of regenerative operation even with a reactive load by the avoidance of rectifiers in the secondary winding and having instead one or more bilateral switches (23) arranged to be operated in synchronism with the secondary voltage. The transistors of modulating circuit 14 are controlled by a source (16) of modulating high frequency of 25 KHz. A microwave output is obtained via a filter 22. In a modification (Figure 3) two transistors (41) are switched alternately by the source (16) and the output is obtained by switching bilateral switches (23) connected to toppings on the secondary of transformer (13) in sequence (1-6) so that from a pulse output (Figure 4C) a microwave output (Figure 4D) may be obtained from the filter 22. The bilateral switches may include a transistor (46, Figure 5) connected across a diagonal of a diode bridge (44). An alternative demodulator to the switches (1-6) has a single secondary winding with bilateral switches a, d, and b,c operated alternately to provide the a.c. output. <IMAGE>

Description

SPECIFICATION Static inverter This invention relates to a static inverter primarily for providing a low frequency output voltage from a d.c.

supply and one object is to provide such an inverter which enjoys the advantages obtainable form using a high frequency transformer while yet it can be used regeneratively with power being returned from the low frequency side to the d.c. supply.

Thus, the inverter can be used as a stand-by power supply for supplying a.c. mains from batteries or for recharging the batteries from the mains and can be used for supplying reactive loads from the d.c.

supply preferably with provision for variable speed drive for example for a machine tool or even for a vehicle.

According to the present invention, a static inverter comprises a high frequency transformer, an input circuit arranged to produce from a d.c. supply a high frequency primary voltage, at least one bilateral switch connected between a tap on the transformer secondary and low frequency output terminals, and means for operating the, or each, switch in synchronism with the secondary voltage.

A bilateral switch is a switch that, unlike a transistor or thyristor, can, when closed, pass current in either direction. One example is a pair of transistors connected back to back each shunted by a diode, the diodes being connected back to back, and means for turning both transistors on together to close the switch.

Another example is a for-rectifier bridge network with supply terminals at opposite corners, and a transistor across the other corners.

Preferably the primary voltage is of square-form and preferably is an alternating voltage. Thus, it may consist of square pulses alternately positive and negative with periods of zero voltage between them.

Such a square-pulsed primary voltage can be pulse width modulated at the output low frequency and then the preferred arrangement is for there to be a bilateral switch connected between each end of the secondary winding and one output terminal while the other output terminal is connected to a centre tap on the secondary winding of the transformer. The bilateral switches can be operated so that for the positive half cycles of the low frequency output the respective ends of the winding are positive with respect to the centre tap and during the negative half cycles the switches are operated when their respective secondary winding ends are negative with respect to the centre tap.

Such an arrangement has the advantage of a high frequency transformer because the primary voltage can be at any desired high frequency and since there are no rectifiers on the secondary side of the transformer the arrangement is suitable for reactive loads permitting regeneration from the oututto the d.c. supply.

In another arrangement, the alternating squarepulsed primary voltage is not modulated, but the pulse width is set in accordance with the desired output amplitude. Modulation is achieved on the secondary side of the transformer as amplitude modulation by virtue of a number of taps along the secondary winding, each of which is at a different number of turns from a common tap connected to one of the output terminals. Each tap has its own bilateral switch and they are operated in synchronism with the secondary voltage but in a sequence so that each pulse provides one short component of the low frequency output voltage to be achieved after smoothing.

In a further arrangement, the alternating squarepulse primary voltage is neither modulated nor is the pulse width controlled. Modulation is achieved by switching the second winding taps mentioned above at the same frequency, but with a controlled phase delay in relation to the secondary pulsed voltage.

The invention may be carried into practice in various ways, and two embodiments will now be described by way of example, with reference to the accompanying drawings, of which: Figure 7 is a circuit diagram of one embodiment of static inverter; Figure 2 is a set of voltage characteristics appearing in the circuit of Figure 1; Figure 3 is a circuit diagram of a second inverter emoding the invention; Figure 4 is a set of voltage characteristics appearing in the inverter of Figure 3 Figure 5 is a diagram of an alternative bilateral switch to that shown in Figure 1; and Figure 6 is a diagram of an alternative arrangement of bilateral switches on the secondary side of the transformer of Figure 1; and Figure 7shows a modification of the circuit of Figure 3.

The inverter is for producing a low frequency a.c.

supply at 11 from a d.c. source 12, for example a battery pack for use as a stand-by mains power supply. The inverter uses a transformer 13 which is a high frequency transformer, and so can be smaller, lighter and cheaper than a transformer designed for use at mains frequency. For example, the frequency at which the transformer 13 is designed to operate might be about 25 KHz.

The primary winding of the transformer 13 is supplied with an alternating pulse width modulated square-wave, as shown at Figure 2D, and that is derived from a modulating circuit 14 including two pairs of transistors A, and A' and B, and B', each pair of transistors being connected in series across the d.c. supply 12. Each of the four transistors is shunted buy a diode 15. The ends of the primary winding of the transformer 13 are connected respectively to the junction points of the transistors A and A1 and the transistors B and B1.

The bases of the four transistors are controlled from a source of modulated high frequency 16 which operates at the 25 KHz frequency, to cause the transistors A and A1 to be switched on alternately as indicated at Figure 2A and to cause the transistors B and B1 to be switched on alternately as shown at Figure 2B. The phase relationship of the voltages 2A and 2B modulated at the desired low output frequency.

If transistors A and B are on together, the opposite ends of the primary winding will be at the same potential and the primary will be short-circuited, but if transistor A is on at the same time at transistor B1, current can flow through the transformer primary in one sense while in the other half cycle at the corresponding instant when the transistor B and the transistor A' are both on, current can flow through the transformer primary in the other sense.

Figure 2C shows that when there is no phase difference between the switching of the transistors A and B, there is zero input voltage across the transformer primary, as shown at 17. 18 shows the voltage waveform across the transformer primary, when there is a 60 lag between the switching of transformer A and the switching of transistor B. 19 shows the voltage when the phase lag is 1200, and 21 shows how when the switching of the two transistors is completely out of phase, there is a continuous voltage across the primary winding alternately of opposite polarity to reproduce the switching signal 2A.

Thus, since the phase difference between the switching of the transistors A and B is continuously modulated at the low output frequency, the voltage appearing across the primary winding of the transformer 13 will be as shown at Figure 2D, which is an alternating square-wave at high frequency, pulse width modulated at low frequency. Because the transformer 13 is a high frequency transformer, it can pass the square edge of that voltage characteristic, which is essentially reproduced at a larger amplitude in dependence on the turns ratio of the transistor.

The secondary winding of the transformer 13 has a centre tapping connected to one of the output terminals 11, and the two ends of the winding are connected through respective bi-lateral switches 23 to the other output terminal. A low-pass filter 22 is included between the winding and the output terminals.

Each of the bi-lateral switches 23 consists of a pair of back-to-back transistors 24 each shunted by one of a pair of back-to-back diodes 25 with terminals 26 connected between the common emitters and the common bases of the two transistors. Thus when a switching signal is supplied at 26, both transistors 24 are rendered conducting, so that current can flow in either sense between the end of the secondary winding and the output terminal. The two switches 23 are switched in synchronism with the voltage across the transformer secondary winding so that the switch 23 is on whenever the upper end of the winding is positive, and is off when it is negative while the switch 23 connected to the lower end of the secondary winding is on when the lower end is positive and off when it is negative.That produces, after filtering at 22, the positive half of a low frequency sine wave, as indicated at 28 in Figure 2F.

Forthe negative half cycle 29, the arrangement for switching the bi-lateral switches 23 is reversed, and the switch connected to the upper end of the secondary winding is on whenever the upper end is negative.

In that way the desired low frequency approximation to a sine wave is achieved using a high frequency transformer, and it is not necessary to provide rectification because of the bi-lateral switches 23. That means that regeneration of power from the output 11 to the input 12 is possibie, without having to have a special regeneration inverter, and this makes the system particularly useful for reactive loads in which a sinusoidal current can thus be produced.

Thus it is possible to vary the output frequency by varying the frequency of the pulse width modulation on the characteristic of Figure 2D to achieve a variable speed drive, for example, for a machine tool, or for a vehicle. It is also possible by controlling the switching of transistors A and A1 and B and B1 to produce a desired output wave shape.

The inverter of Figure 1 used pulse width modulation on the primary side of the transformer to introduce the low frequency reference, and in the embodiment of Figure 3 that reference is supplied as amplitude modulation on the secondary side of the transformer.

In this case a stepped square wave as shown in Figure 4A is supplied to the transformer primary winding by means of a pair of transistors 41 connected between the positive side of the d.c.

source 12 and the respective ends of the primary winding, each shunted by a reversed diode 42. The centre tap of the winding is connected to the negative side of the source 12. The transistors are controlled by the high frequency source 16 so that in a complete cycle, first one of the transistors 41 is fired, and then both are extinguished, and the other transistor is fired, and then both are extinguished again. The voltage characteristic of Figure 4A is reproduced at the secondary winding of the transformer 13 at an increased voltage.

The second winding has a tap 33 slightly to one side of centre, and a number of other taps (shown here as being six in number) at numbers of turns distant from the tap 33 which increase in accordance with a quarter cycle of a sine wave. Each tap, other than the tap 33, is connected to one side of the low pass filter through a bilateral switch 23 controlled at 26. In Figure 3 the switches are numbered 1-6 in the order in which they are fired.

Operation of the switches 23 is in synchronism with the high frequency source 16, so that they are fired in turn, each during the period when one of the thyristors 31 and 32 is fired. The firding sequence is indicated at Figure 4B.

When the switch 1 is operated, the voltage supplied to the filter 22 will be corresponding to the first tap and will be small, as shown in Figure 4C at 34.

When switch number 2 is operated, the voltage will be rather greater, as indicated at 35, and so on, and it will be appreciated that since the taps are alternately on opposite sides of the tap 33, and the peaks in the square wave 4A are alternately of opposite polarity due to the alternate firings of the thyristors 31 and 32, all of the voltages such as 34 and 35 will be the same polarity throughout the first half cycle. That cycle is completed by firing the switches 23 in the reverse order to produce the second quarter cycle.

Then switch number 1 is fired again as indicated at 36 and because the peak in the characteristic 4A will now be of the other polarity, the voltage on the secondary side will now be of the opposite sense to start the second half cycle. It will be clear that the second half cycle will built up in the same way as indicated at 37. Then after smoothing in the filter 22 the low frequency output 4D will be produced.

There can of course be as many taps and switches 23 as is desired in accordance with the number of steps in the output characteristic 4C, and they may be arranged at numbers of turns from the tap 33 corresponding to a sine wave as described above, or may be at other distances if any output wave of different form is required. The switching of the switches 23 has to be in synchronism with the source 16 for operating the thyristors 31 and 32 and the order of firing is determined by a logic circuit in accordance with the output characteristic required.

The order would be as described above for producing a sine wave. The output frequency can be varied by varying the frequency of the source 16, but can only be a fixed fraction of the high frequency of that source 16.

The amplitude of the signal at 11 can be varied by varying the width of the pulses in Figure 4A by controlling the time in each cycle when thetransis- tors 41 are on.

In an alternative arrangement, the width of the primary pulses is not varied, but the switching of the bilateral switches 23 occurs with a controllable phased delay in relation to the secondary voltage so that only a selected portion of each second voltage pulse is used to build up the output sine wave.

It would be possible to use the primary circuit 14 of Figure 1 in the two Figure 3 arrangements, but because of the switched tap system of Figure 3 it is not necessary to have a defined input zero, as is the case in Figure 1 where it is important that the primary winding is short-circuited effectively at moments of zero input voltage.

Figure 5 shows an alternative to the bilateral switch 23, which alternative uses a single transistor.

Four diodes 44 are connected in a bridge configuration as shown in Figure 5, with the supply terminals 45 connected to opposite corners of the bridge and the transistor 46 connected across the other corners.

The arrangement of the diodes is such that when the transistor is on, current can flow between the terminals 45 in either direction, but when the transistor is off the current cannot flow in either direction.

Then Figure 6 shows an alternative type of demodulator to that shown on the secondary side of the transformer in Figure 1. In this case there are four bilateral switches 23, one connected between each end of the secondary winding, and each side of the filter 22. The switches 'A' and 'D' are turned on for one half cycle of the output, and the switches 'B' and 'C' are turned on for the other with the output plurality being determined by the phase relationship used. Although this arrangement has twice as many bilateral switches as in Figure 1 the voltage stress across each switch is reduced.

In the Figure 3 arrangement, the zero reference point can be defined as the secondary of the transformer as described above, or could be defined as the primary side by modifying the primary circuit as shown in Figure 7. A diode 48 is connected between the common point of the diode 42 and the positive side of the supply 12, and a transistor 49 is connected between that common point and the negative side of the supply, and can also be controlled from the H.F. source 16 to connect the negative side to the common point.

Claims (17)

1. A static inverter comprising a high frequency transformer, an input circuit arranged to produce from a D.C. supply a high frequency primary voltage, at least one bilateral switch connected between a tap on the transformer secondary and low frequency output terminals, and means for operating the, or each, switch in synchronism with the secondary voltage.

2. An inverter as claimed in Claim 1 in which the primary voltage is an alternating voltage.

3. An inverter as claimed in Claim 2 in which the primary voltage consists of square-pulses alternately of opposite polarity.

4. An inverter as claimed in Claim 3 including a pulse width modulator for moduiating the primary voltage at the low frequency.

5. An inverter as claimed in Claim 4 including a bilateral switch connected between each end of the transformer secondary and one output terminal together with a connection from a centre tap on the transformer secondary to the other output terminal.

6. An inverter as claimed in Claim 5 in which during alternate half cycles of the low frequency output the switches are arranged to be operated whenever their respective end of the secondary winding is positive with respect to the centre tap, and during the other half cycles of the low frequency output the switches are arranged to be operated when their respective ends of the secondary winding are negative with respect to the centre tap.

7. An inverter as claimed in Claim 4 including a bilateral switch connected between each end of the transformer secondary and each output terminal.

8. An inverter as claimed in Claim 7, in which the ends of the secondary are arranged to be connected to alternate output terminals during alternate output cycles.

9. An inverter as claimed in any of Claims 1-3 including a number of taps on the transformer secondary winding each connected through its own bilateral switch with one output terminal, and a common tap on the secondary winding connected to the other output terminal, each of the number of taps being at different number of turns from the common tap.

10. An inverter as claimed in Claim 9 including logic means for switching the bilateral switches in synchronism with the secondary voltage but in a sequence in dependence on the desired output wave form.

11. An inverter as claimed in Claim 9 including logic means for switching the bilateral switches at the same frequency as the secondary voltage but with a phase delay.

12. An inverter as claimed in any of the preceding claims including a low pass filter between the secondary winding and output terminals.

13. An inverter as claimed in any of Claims 1-3 including means for controlling the width of the pulses of the primary voltage in dependence on the desired output amplitude.

14. An inverter as claimed in any of the preceding claims in which the input circuit comprises two pairs of transistors, the transistors of each pair being connected in series with each other and each being shunted buy a rectifier, the primary winding of the transformer being connected between the junction points between the transistors of the respective pairs, and the four rectifiers constituting a full wave rectifier between the transformer primary and the d.c. supply.

15. An inverter as claimed in any preceding claims in which each bilateral switch comprises a pair of transistors connected in backto back each in reverse parallel with a diode, and means for switching thetransistortogether.

16. An inverter as claimed in any of Claims 1-14, in which each bilateral switch comprises a diode bridge with supply terminals connected to two opposite corners, and a transistor connected across the other corners.

17. A static inverter arranged substantially as herein specifically described with reference to Figure 1 or Figure 3 of the accompanying drawings.