MXPA98001654A - Opera current energy converter - Google Patents

Abstract

An energy converter that includes a pair of switches, each of which switches is connected between positive and negative constant DC voltage sources and a common output node that is connected to a load. A modulator opens and closes the switches in a sequence in which the electric currents controlled by the switches are in opposition to each other in the common output node and summed in the output node. The modulator includes a triangular wave generator and comparators that are connected to an error amplifier to provide a support switching voltage for the comparators. The comparators switch the control switches in response to the triangular wave generated by the triangular wave generator by increasing and decreasing the reference levels. The lack of juxtaposition and the overlap of the converter are controlled by the variation of the CD deviation level of the triangular wave.

Description

OPPOSED CURRENT ENERGY CONVERTER DESCRIPTION OF THE INVENTION The present invention relates to a pulse width modulated energy converter of opposite current ("PWM"). Power converters are widely used for many industrial and commercial purposes. Such power converters can be used to convert direct current (DC) to alternating current (AC) to be used as an AC power source or as battery chargers / dischargers, motor controls, etc. The power converters can also be used as amplifiers, both for entertainment (sound amplification) as for industrial uses. The pulse width modulated converters of the prior art (PWM) use a pair of switches to connect a load alternately to the DC power supplies of opposite polarity. A modulator opens and closes the switches in alternation (without turning both on at the same time) to produce a modulated output signal of width that is subsequently filtered by a low pass filter before being transmitted to the load. In such prior art devices, great care must be taken to ensure that both switches are not turned on at the same time, although to operate the system for high fidelity, it is desirable that the switches be turned off and on at about the same time to the extent of the possible. To handle the transistors ^ during switching, (referred to as "direct trip"), they connect inductors between the switches. In addition, circuits known as "no juxtaposition" circuits are used to create small controlled intervals between the switch conduction times. The opening and closing of the switches imposes a so-called "ripple" frequency on the output waveform having a frequency that is equal to the switching frequency of the switches. It is desirable that the magnitude of the ripple frequency be minimized, particularly to a zero output of the power converter, which is the output at which most such devices are classified, since the most common value of the input is zero for voice and sound applications. In the present invention, instead of there being no overlap driving the switch, the overlap is deliberately carried out to the maximum. In the present invention, if an output of zero is desired, SI and S2 will be switched on simultaneously and just for less than 50% of the duty cycle. If a positive output is desired, the switch that connects the positive DC voltage source to the load is turned on for more than 50% of the duty cycle, while the switch that connects the negative DC voltage source to the load is on for less than 50% of the corresponding duty cycle. If only the sum of "switch-on times" of the switches is required to be no greater than 100% of the duty cycle during normal operation of the device, but the device may be operated in a condition without juxtaposition or overlap, when the sum of the switch-on times of the switches are lower and higher, respectively, than 100% of the duty cycle. The condition "without juxtaposition" is useful, for example, to minimize ripple and energy consumption during periods of time when no input is being received by the converter. The condition of "overlap" is used to accommodate short demands on excesses of the energy scale of the converter. These and other advantages of the present invention will become apparent from the following description, with reference to the accompanying drawings, in which: BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a circuit diagram illustrating a half-bridge power converter made according to the teachings of the prior art; Figure 2 is a switch synchronization diagram illustrating the manner in which the switches of the circuit illustrated in Figure 1 are operated; Figure 3 is a circuit diagram illustrating the converter made in accordance with the teachings of the present invention; Figure 4 is a timing diagram illustrating the manner in which the circuit switches illustrated in Figure 3 are open and closed; Figure 5 is a graphic illustration of the current supplied to the output node by the switches of the circuit illustrated in Figure 3 and their sum; Figure 6 is a circuit diagram illustrating a full bridge converter made in accordance with the teachings of the present invention; Figure 7 is a diagrammatic illustration of the circuit used by the modulator controlling the converter illustrated in Figure 3 and made in accordance with the teachings of the present invention; and Figures 8-12 are switching diagrams illustrating the manner in which the modulator of Figure 7 generates the switching pulses both in a normal state and in the overlapping and non-juxtaposed states. Referring now to Figures 1 and 2 of the drawings, which illustrate the prior art device, a prior art converter or amplifier circuit generally indicated by the number 10 is applied between a modulator 12 and a load 14 (such as as a speaker). The power converter 10 includes a switch 16 connected between the positive direct current source 18 and the node 20 and a switch 22 connected between the node 20 and a negative DC supply "24. This signal, which may vary in time, is applied to the node 20. The inductors 26, 28 are connected between the switches 16, 22 respectively and the node 20 for directing the "direct trigger" of current when the switches 16 and 22 are open and closed at substantially the same time. switches 16 and 22 are controlled by output signals generated by modulator 12 and applied to switches 16 and 22 via lines 30 and 32. The modulator receives a deviation signal at input 34 and a feedback signal from the output node 36, which is transmitted to the modulator 12 via line 38. The output node 36 is connected to the load 14. A low pass filter consisting of an inductor 40 and a cap Acitor 42, are connected between the output node 36 and the node 20 to filter the switching signals generated by the opening and closing of the switches 16 and 22. The switches 16, 22 can be MOSFETs or IGBTs (bipolar transistors of isolated gate ), both of which are well known to those skilled in the art, and are controlled by the switching signals transmitted on lines 30 and 32.

Referring now to Figure 2, the modulator generates a waveform 44 that is transmitted on line 32 to operate switch 16 and a waveform 46, which is transmitted on line 32 to operate switch 22. As it can be seen, the pulses are transmitted on the lines 30 and 32 in alternation, that is, the signal on the line 30 which closes the switch 16 is never allowed to close the switch 16 unless the switch 22 is open and vice versa . However, as discussed in the above, to handle high fidelity and voice amplification it is desirable that the switch 22 be closed immediately as the switch 16 is opened and vice versa. The inductances 26 and 28 are used to direct a "direct trip", which are the transient currents generated by the opening and closing of the switches 16 and 22 substantially at the same time. Referring now to Figures 3 and 4, an energy converting circuit according to the present invention indicated generally by the number 48 and includes a switch 50 which connects an output node 52 to a positive CD power supply 54 and a switch 56 which connects the output node 52 to a negative DC power supply 58. A diode 60 allows the current, run in free run with the output node 52 to the negative CD power supply 58 when the switch 50 is opened after being closed and the diode 62 allows the current to run in free with the output node 52 at positive CD power supply 54 when the switch 56 is reopened after being closed. The signal applied to the output node 50 by the switch 50 is filtered by the low pass filter consisting of an inductance of 64 and the capacitor 66 and a similar low pass filter consisting of an inductance 68 and the capacitor 66 filters the signal caused by the opening and closing of switch 56. Switches 50 and 56 are controlled by a modulator which will be described below in detail with reference to Figure 8. As described above, switches 50 and 56 can be implemented either as MOSFETs, IGBTSs, or any other similar device. Referring to Figure 4, the waveform 70 indicates the signal from the modulator (not shown) operating the switch 50 and the waveform 72 indicates the signal from the modulator operating the switch 56. The waveforms 70, 72 are as shown for a 50% duty cycle for each of the waveforms 70, 72. Therefore, when summed, at the output node 52, the output will be 0, since the switch 50 is connected to the output node with the positive CD supply 54 and the switch 56 is connected to the output node 52 with the negative CD supply 58. The small arrows on each of the waveforms 70 or 72 indicate the address modulation for an increased positive output at the common supply node 52 which is connected to the load 74. The waveforms 70, 72 will remain centered on each other, although the width of the waveform 72 will be decreased and the width of waveform 70 will increase torque to the positive output of increase. In contrast, for the increased negative output, the waveform 72 will be increased and the waveform 70 will be reduced. Referring to the waveforms illustrated in Figure 5, the waveform 76 represents the current in the inductance 64, the waveform 78 represents the current in the inductance 68 and the waveform 80 represents the sum of the shapes of upper and lower wave illustrated in Figure 5, which is the current at the output node 52. It will be noted that waveforms 76 and 78 include a waveform superimposed on the waveform, indicated between the displacements between the peaks A and B on the waveform. This ripple frequency or printed waveform is a result of the opening and closing of the switch 50 in the case of the waveform 76 and the switch 56 in the case of the waveform 78. It will be noted that a waveform similar is printed on the waveform 80 although the frequency of this waveform is twice the frequency of the waveform printed on the curves 76 and 78. This is due to the fact that the switches 50 and 56 are driven in different moments, thus printing a ripple on the waveform summed twice the frequency of any individual commutator. Higher frequency undulations are desirable, since they are easier to filter. It will also be noted that at the output 0, as indicated in X in Figure 5, that the ripple amplitude is substantially zero. This is due, as indicated in Figure 4, that switches 50 and 56 are driven at about the same time in the zero output current, albeit in opposite directions, which adds up to zero. Therefore, the ripple is set to zero at zero output. As described above, CD to AC converters are typically tested to a specification that specifies a ripple at zero output. By using the invention, this ripple amplitude at the zero output will be minimized. The ripple specification at zero output is selected because, in the case of sound amplification, the noise imposed by the undulation is more noticeable and annoying in the pauses when the output is not present. In the case of voice amplification, this is the most common condition, due to pauses between words, etc.

It will also be noted that small inductances such as inductances 26, 28 in Figure 1, which were required to handle direct firing are not necessary in the invention since the switches are isolated by inductances 64, 68 which are a part of the low pass filter. Furthermore, as will be explained below, the lack of juxtaposition in which the zero output can be imposed for a longer period of time is easily achieved with the modulator necessary to operate the converter 48. In addition, the modulator is substantially simplified in comparison to the modulators required by the prior art and no circuits are required to achieve the lack of juxtaposition, which is achieved by the simple deviation of a CD deviation over a triangular waveform, as will be explained below. Referring now to Figure 6, the load 74 is connected to a full bridge converter consisting of two half-bridge converters 48A, 48B, each of which are identical to the half-bridge converter 48 illustrated in Figure 3. Accordingly , each of the elements of each half-bridge converter 48A, 48B retains the same reference number, although with the addition of the letter "A" or "B". The phase deviation of the operation of the switches of one half-bridge of a complete bridge converter in relation to the other half-bridge is also known. Accordingly, the ripple frequency of the current supplied to the load 70 is duplicated again over that of each half-bridge.It is also known to add additional half-conductors, each of which, in the case of the invention, will increase the ripple frequency As noted above, the higher frequency ripples are easier to filter.With reference now to Figure 7 of the drawings, a modulator generally indicated by the number 82 includes an error amplifier 84 which adds the difference between an input signal representing a desired level in the output node which is fed to the internal amplifier 84 over the input 86 and a feedback signal which is received by the internal amplifier 84 over the input 88. The signal of feedback from the input 88 is taken from the output node 52. The output of the error amplifier 84 is transmitted to the input of inversion 87 of a comparator 89, the output of which is transmitted on the line 90 to operate the switch 56. The output of the error amplifier 84 is also fed through an inverter 92 and is then transmitted to the inversion input 94. of a - - ^ • "- ^ j? & i. 12 comparator 96, the output 98 of which is connected to operate the switch 50. The modulator 82 also includes a triangular wave generator generally indicated by the number 100, which receives a square wave input (generated in any conventional manner) in 102 and generates a triangular wave output on line 104 and transmitted to positive input 106 of comparator 89 and positive input 108 of comparator 96. Triangle wave generator 100 is and includes a deviation control 110 for adjusting the CD level of the waveform generated by the triangular wave generator 100 upwards or downwards. The deviation control 110 may be responsible for the dynamic conditions, such as a pause in the case of the voice transmission and a temporary demand for power and access that is normally provided by the converter. Referring now to Figures 8-12, the operation of the modulator 82 and the converter 48 will be described. Referring to Figure 8, the wave triangular T generated by the triangular wave generator 100 is centered at zero. The output of the error amplifier 84 is also assumed to be zero. In this case, both comparators 89 and 96 light up at a point A and turn off at a point B. Consequently, waveforms 70 and 72 are generated, as illustrated in Figure 4. Both of the waveforms 70 and 72 are turned on and off at the same time and have a period exactly half of the duty cycle (the duty cycles between the points of light on the triangular wave T, such as between points A and A or B and B). As discussed above, the switch 50, which is connected to the positive DC supply voltage, is turned on and off by the waveform 70 and the switch 56, which is connected to the negative DC supply voltage is turned on and switched off by waveform 72. Since both switches 50 and 56 are turned on in exactly the same time period and are turned on exactly halfway through the duty cycle, the voltages transmitted through those switches are summed at the node exit 52 which will be zero. Referring now to Figure 9, the triangular wave T remains centered at zero although the output of the amplifier 84 is a unit +1. This output is fed through the inverter 92 and transmitted to the comparator 96. Accordingly, the comparator 96 is turned on when the triangular wave exceeds a unit -1 as indicated at point C and will turn off when the triangular wave falls below the unit, as indicated by F. Similarly, comparator 89 will turn on when triangular wave T exceeds a +1 unit as indicated in D and will turn off when triangular wave T falls below a unit as indicated in E. Accordingly, waveform 70 is turned on at C and turned off at F, and waveform 72 is turned on at D and off at E. As can be seen in Figure 9, waveforms 70 and 72 are centered with each other. Accordingly, since the waveform 70 and 72 are summed at the output node 52 a positive output is supplied to the load 74. Referring to FIG. 10, where it is assumed that the output of the amplifier 84 is a negative unit , the "on" time of switches 50 and 56 is inverted from that shown in Figure 9, as indicated by waveforms 70 and 72. Therefore switch 56 will turn on at point G and turn off at point J, and switch 50 will turn on at point H and turn off at point I. The sum of the turn-on time of waveforms 70, 72 will still remain the same for one cycle of triangular wave T. Figure 11 illustrates the modulator 82 and the converter 48 that are operated in an overlap condition, such as would occur in response to a high demand for transient energy. It is assumed that the output of the amplifier 84 is zero. In the overlap condition, the deviation control 110 is operated to deflect the T waveform upwards one unit, so that one unit is centered above zero instead of being centered on zero as in the case of FIGS. -10. The dotted lines in Figures 11 and 12 include the position of the triangular waveform having the CD deviation of the waveform without being shifted. Accordingly, both of the comparators are turned on at point K and turn off at point L, generating identical waveforms 70 and 72 that operate switches 50 and 56. However, it will be noted that the sum of the times of " "on" switches 50 and 56, as indicated by waveforms 70 and 72, exceed a duty cycle of the triangular wave T. Referring to FIG. 12, a condition without juxtaposition is illustrated, in which the switches 50, 56 are turned on for a total cycle less than the triangle wave cycle T. The condition without juxtaposition is used when, for example, no converter output is desired for a period of time. The condition without juxtaposition is achieved by deviating the triangular waveform T down one unit, operating the deviation control 110. As indicated in Figure 12, the triangular waveform T has been moved down one unit, in this way instead of being centered at zero, it is centered one unit below zero. It is also assumed that the output of the amplifier 84 is zero. In this case, as illustrated in Figure 12, both of waveforms 70 and 72 are generated when the triangle increases above zero, as indicated at point M, and are finalized when the triangle wave falls below Accordingly, waveform 70 'and 72 turn on and off at the same time, although their total duty cycle is less than a triangular wave cycle T as a result of the downward deviation in the deviation of DC from the triangular wave T.

Claims (19)

1. CLAIMS 1. The power converter is characterized in that it comprises a pair of switches, each of the switches being connected between two sources of constant voltage and a common output node for each of said switches, the output node that is connected to a load, and modulating means for opening and closing the switches in a sequence in which the electric currents controlled by such switches are in opposition to each other, the electric currents being summed at the output node. The power converter according to claim 1, characterized in that the modulation means includes means for opening and closing each of said switches in a controllable work cycle. 3. The power converter according to claim 2, characterized in that the modulating means includes means for controlling the output of the output node by increasing and decreasing the duty cycle of one of the switches while decreasing and correspondingly increasing the work cycle of the other switch. The power converter according to claim 3, characterized in that the sources of constant voltage include a positive voltage source and a negative voltage source, one of the switches being connected between the voltage source -positive and the voltage node. common output and the other switch being connected between the negative voltage source and the common output node, first one-way conducting means, connected between a switch and the negative voltage source to allow current to flow from the output node common when such a switch is opened and second one-way conduction means connected between the other switch and the positive voltage source to allow current to flow from the common output node when the other switch is opened. The power converter according to claim 1, characterized in that the modulating means includes means for closing one of the switches, closing the other switch and opening the other switch before opening the first switch. The power converter according to claim 1, characterized in that one of the switches is connected to the output node through a first inductance and the other switch is connected to the output means through a second inductance. The power converter according to claim 1, characterized in that the modulator means includes means for closing one of said switches and then closing the other switch before the first switch is opened. The power converter according to claim 1, characterized in that the modulating means include means for generating a first electrical signal for controlling a switch and a second electrical signal that controls the other switch, triangular wave generating means for generating a triangular wave signal, and means for generating both of the first and second electrical signals as a function of the same triangular wave signal generated by the triangular wave generating means 9. The power converter according to claim 8, characterized because the modulating means includes a pair of comparators for generating the first and second electrical signals, one of the comparators comparing the triangular wave signal with a reference signal, the other comparator comparing the triangular wave signal with the inverse one reference signal 10. The energy converter according to claim ication 8, characterized in that the modulating means include means for moving the CD level of the triangular wave signal to create the lack of juxtaposition between the output of such switches. 11. A full bridge energy converter characterized in that it comprises a half-bridge on the upper side and a half-bridge on the lower side with a load connected between the half-bridges, each of the half-conductors that includes a pair of switches, each switch of each semiconductor that is connected between constant voltage sources and a common output node for each switch of each semiconductor, the output nodes of the half-conductors that are connected to the load and modulation means for opening and closing such switches of the half-bridges in a sequence in which the electric currents controlled by the switches of each half-bridge are in opposition to each other, the electric currents being summed at the corresponding output node of each semipuente. The full bridge energy converter according to claim 11, characterized in that the modulating means includes means for phase shifting for the operation of the switches of a half bridge with respect to the operation of the switches of the other half bridge. The full bridge energy converter according to claim 11, characterized in that the modulating means includes means for controlling the output node output of each half bridge by increasing and decreasing the duty cycle of one of the switches of each half bridge in so much that it decreases and increases correspondingly the work cycle of the other switch of each half bridge. The power converter according to claim 13, characterized in that the constant voltage sources include a positive voltage source and a negative voltage source, one such switch of each half bridge being connected between the positive voltage source and the common output node and the other switch of each semiconductor being connected between the negative voltage source and the common output node, first one-way conducting means connected between a switch of each half-bridge and the negative voltage source to allow a current flow from the corresponding half-bridge common output node when one switch of each half-bridge is open and second one-way conducting means connected between the switch of each half bridge and the positive voltage source to allow current to flow from the common output node of the corresponding semipuente when the other switch of each semipuente is a open 15. The "method for the operation of an energy converter comprising the steps of connecting voltage sources to a common output node through switches that connect each of the voltage sources to the output node, and open and close the switches in a sequence in which the electric currents controlled by the switches are in opposition to each other and summing the currents in the output node 16. The method for the operation of an energy converter in accordance with the claim 15, characterized in that the method includes the step of opening and closing the switches in a controlled work cycle 17. The method for operating an energy converter according to claim 16, characterized in that the method includes the step of controlling the exit of the output node by increasing and decreasing the duty cycle of one of the switches as long as it is correspondingly increased and decreases the duty cycle of the other switch. 18. The method for operating an energy converter according to claim 15, characterized in that the method includes the steps of closing one of the switches, closing the other switch and opening the other switch before opening the first switch. The method for operating an energy converter according to claim 15, characterized in that the method includes the steps of closing one of the switches and then closing the other switch before opening the first switch.