JPH0334711B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a method for reducing switching output from the DC-DC converter without impairing the high ground impact characteristics and high efficiency power supply characteristics, which are the advantages of the floating DC-DC converter type DC supply circuit. The present invention relates to the configuration of a DC supply circuit that can remove noise, particularly common-mode component noise.

The DC supply circuit is a circuit that supplies DC communication current from the exchange to terminals such as telephones via subscriber lines.

The DC supply circuits currently commonly used in exchanges are implemented by relay switch windings. The performance required for a DC supply circuit is as follows:
(1) Power feeding function that supplies DC, (2) Loop monitoring function that detects the flow of DC, (3) Interface function with voice calls, (4) High balance against AC induced noise, which is common mode component noise. These include absorption characteristics and absorption capacity. In the case of a relay switch winding, the winding has a large inductance, and the two divided windings are bifurcated so that the characteristics of both windings match. In addition, the winding direction of the two-divided winding is specified, and the impedance of the winding becomes low for common-mode component noise and absorbs this noise into the station power supply, while for balanced component signals such as voice calls, The winding has a high impedance and its existence can be ignored. The performance of the winding of the relay switch listed above satisfies the above-mentioned performance requirements for the DC supply circuit. Although this DC supply relay switch has excellent performance, the relay switch is an electromagnetic component.
There are limits to miniaturization in order to achieve inductance that has high impedance up to low frequencies.
Furthermore, since it is necessary to make the characteristics of the two divided windings the same, there is a limit to the reduction in manufacturing costs.

Furthermore, the communication current supplied to the terminal using the DC -48V local power supply ranges from 20mA to more than 100mA. Conventional relay switches and DC supply circuits that simulate them using electronic circuits have an internal DC resistance of 440Ω, so the current output is 100mA.
If this amount of current flows, the internal power consumption will be over 4.4W, which generates too much heat to be incorporated into the integrated circuit. In addition, since the DC supply circuit is a circuit for subscribers, and a large number of circuits are installed in the exchange, the large amount of power consumed in the circuit has to be avoided due to the thermal design of the mounting surface. There are several items that require consideration from this perspective as well.

In recent years, miniaturization due to electronicization of DC supply circuits,
Considerations are being made to reduce costs through integrated circuits and mass production. One direction to consider is DC-DC
There is a high-efficiency DC supply circuit using a converter method. Conventional relay switches and DC supply circuits that simulate them using electronic circuits have the disadvantages mentioned above, but DC-DC converter type DC supply circuits require a voltage drop of 440Ω to achieve the same supply current characteristics. Since the desired characteristics are obtained by using the inductance of a coil such as a transformer without using a coil, in principle it is a high-efficiency supply circuit that does not generate heat in the supply circuit.

There are several types of DC-DC converters, but what all types have in common is that a pulsed current is passed from the primary power source through a switch that opens and closes in pulses to the inductance element. This means that the current appearing in the rectifier circuit on the opposite side is rectified to become a DC power supply that is supplied to the subscriber line. When a pulse transformer winding is used as an inductance element, the winding to which the switching element and the local power supply are connected and the winding to which the rectifier circuit and subscriber line are connected can be realized using separate windings of the transformer. winding can be electrically floating from ground.
The former winding will be referred to as the primary side winding, and the latter winding will be referred to as the load side winding. If the winding on the load side and the load circuit are made floating, the impedance to the ground (also called longitudinal impedance or common-mode component impedance) becomes extremely high, so it is easy to realize a large unbalanced attenuation characteristic to the ground. This shows that among the performances required of the DC supply circuit described above, (4) high balance characteristics and absorption capacity for AC induced noise, which is in-phase component noise, can be easily achieved. Therefore, this floating DC-DC
Research has been carried out on the practical application of converter-type DC supply circuits, but floating type circuits have had the drawback of not being able to suppress switching noise, especially common-mode component noise. This situation will be explained using drawings.

FIG. 1 shows the configuration of a conventional floating DC-DC converter type DC supply circuit. Also, the second
The figure shows typical waveforms of each part in FIG.
In FIG. 1, 100 is a clock generation circuit;
01 is a pulse width modulation (PWM) circuit, 102 is a rectifier circuit, 103 is a filter circuit, SW100 is a switch element, T100 is a pulse transformer, D100
is a rectifying diode, C100 is a rectifying capacitor, and A and B are terminals. The operation of the circuit will be explained with reference to FIGS. 1 and 2.

A clock generation circuit 100 generates a clock that determines the primary side switching frequency of the DC-DC converter, and this clock is input to a pulse width modulation (PWM) circuit 101 to provide PWM timing. This PWM signal is as shown in the waveform example of voltage b in FIG. 2, and in FIG. 2, the rising edge of the clock voltage a is used as the PWM timing. Switch element using PWM signal as drive signal
Apply to SW100. In the waveform of voltage b in Figure 2, when the voltage b is at a high level, the switch element SW100 becomes conductive, and when the voltage b is at a low level, the switch element SW100 becomes conductive.
When SW100 is made non-conductive, the pulse transformer T10 is connected to the pulse transformer T10 through the switch element SW100.
A current as shown by current I in FIG. 2 flows from the ground potential to the local power supply potential -V _B in the primary winding (between terminals 1 and 2) of the power supply. Current I in the primary winding
When current flows, the current also tries to flow in the other windings of the pulse transformer T100, but in the case of the winding on the load side (between terminals 3 and 4), the direction of the induced current is determined from the winding direction of the winding. This is the direction in which it flows out from the terminal 4 to the load. The direction of this current is the rectifier diode D100
As a result, while the current I is flowing through the primary winding, the load impedance of the load side winding becomes extremely large, and no current flows. On the other hand, when the switch element SW100 becomes non-conductive and the current in the primary winding becomes zero, a current in the opposite direction tends to flow in the winding on the load side. This direction is the direction in which the fluid flows out from the terminal 3 toward the load side. At this time, the diode D100 becomes forward biased and conducts, and the current indicated by the arrow i in FIG. 1 flows through the rectifying capacitor C100. An example waveform of current i is shown in FIG. Current i charges capacitor C100 with the polarity shown in FIG. Capacitor C100 is charged while current i is flowing, and discharges its charge toward the load while current i is not flowing, which is repeated every clock cycle. This discharge causes current to flow to the load and becomes a supply current to the terminal.

Another winding (between terminals 5 and 6) of the pulse transformer T100 and the rectifier circuit 102 create a PWM modulation signal e, which is sent to the pulse width modulation circuit 101.
sent to. The relationship between the winding direction of the winding between terminals 5 and 6 and the current switching polarity in the rectifier circuit 102 is the same as the winding direction of the winding on the load side and the current switching polarity of the rectifier diode D100. When the pulse occupancy rate of the signal represented by voltage b, which is the control signal for switching element SW100, is constant, the total amount of power transferred via pulse transformer T100 is constant regardless of the size of the load. Therefore, the voltage across the rectifying capacitor C100 changes in proportion to the square root of the DC resistance of the load.
This voltage appears between terminals 5 and 6 in proportion to the ratio of the number of turns between the winding on the load side and the winding between terminals 5 and 6, so the voltage information between terminals 5 and 6 is passed through the rectifier circuit 102. The obtained voltage e has information on the DC resistance value of the load of the winding on the load side. Furthermore, by comparing the voltage e with an appropriate threshold voltage, it is also possible to perform a DC loop detection function.

The filter circuit 103 has a characteristic of allowing direct current to pass through but cutting off alternating current, so when viewed from terminals A and B to which subscriber lines are connected, the rectifying capacitor C100, which is a battery, is visible for direct current, resulting in low impedance. With alternating current, the filter circuit 103 has a high impedance, so the direct current supply circuit becomes invisible. Although omitted in FIG. 1, the alternating current audio signal is terminated and processed by the audio signal processing circuit.

Common mode component noise is generated as follows. In the operating principle described so far, while the current I flows through the primary winding, the rectifier diode D100 becomes reverse biased, and no current flows through the load winding, but in reality Since stray capacitance exists in the lead wires of the terminals 3, 4 and other parts, a pulsed current flows through the stray capacitance. In principle, all the voltage induced in the winding on the load side is applied to both ends of the rectifier diode D100, and if the system on the load side is a floating type, this voltage will not go out to the subscriber line, but there is stray capacitance. is divided by the ratio of the reciprocal of the stray capacitance of each of terminals 3 and 4, and a voltage waveform is generated between the ground and each terminal. FIG. 3 is a diagram illustrating this situation.

FIG. 3 is a diagram in which the winding on the load side of FIG. 1 up to the rectifying capacitor C100 is extracted and stray capacitance is added. The arrow shown by the dashed-dotted line in the figure shows the path through which leakage current flows through the stray capacitances C300 to C302 when the switching element on the primary side is conductive and the rectifier diode D100 on the load side is non-conductive. ing. Crossing points 3 and 4 of the circuit in Figure 3 are the pulse transformers T100 in Figure 1.
Terminals 3 and 4 match. diode 100
Even if the third
If there is a path shown by the dashed line in the figure, the stray capacitance C300~
terminals 3 and 4 depending on the impedance ratio of C302 and rectifying capacitor C100 to the pulse.
And a ground voltage waveform is generated at terminal f. Here, since the capacitance of the rectifying capacitor C100 is generally extremely large compared to the stray capacitances C300 to C302, the voltage waveform of the terminal 3 and the voltage waveform of the terminal f match sufficiently. As shown in Figure 1, this terminal 3
is directly connected to one of the pair of subscriber lines, and the terminal f is also connected to the other pair of subscriber lines via the filter circuit 103.
Since the ground impedance of the subscriber line and the terminal equipment is larger than the AC impedance of terminal 3, the almost identical voltage waveforms at terminals 3 and f are output to the subscriber line as in-phase component noise.

The level of common mode component noise is pulse transformer T10
0 Voltage induced in the winding on the load side and stray capacitance C30
It depends on the capacitance ratio of 0 to C302, and may reach several tens of V. When this common-mode component noise is applied to a pair of wires with a finite degree of ground balance, the common-mode component is converted into a balanced component depending on the degree of unbalance, and becomes noise that is difficult to remove. The drawback was that it could enter the route as induction noise, degrading the quality.

The present invention aims to solve these drawbacks, and connects both ends of the winding on the load side of a pulse transformer and both ends of a rectifying capacitor via a semiconductor element that allows current to flow in only one direction. This prevents a large amplitude signal induced in the winding on the load side of the pulse transformer from being transmitted to either terminal of the rectifying capacitor, even if there is stray capacitance. Examples will be described below with reference to the drawings.

FIG. 4 shows an embodiment of the present invention. This corresponds to Fig. 1 showing the conventional circuit configuration, but the difference is that the winding on the load side of the pulse transformer T100 (terminal 3
A second rectifying diode D400 is connected between the terminal 3 of the rectifying capacitor C100
Connect terminal 3 - second rectifier diode
This is the point where current flows in the path D400 - rectifying capacitor C100 - first rectifying diode D100 - terminal 4. The operation of the second rectifying diode D400 in the DC-DC converter circuit is similar to that of the first rectifying diode D400.
It performs a rectifying operation similar to that of the rectifying diode D100.

Moreover, since the impedance of the second rectifying diode D400 shows the same change as that of the first rectifying diode D100, this second diode D4
00 has a very large influence on the formation of a side flow path for pulsed signals due to stray capacitance.

FIG. 5 shows an extracted circuit diagram of the circuit from the winding on the load side to the rectifying capacitor in the circuit configuration according to the present invention, and a diagram showing the influence of stray capacitance.

The configuration shown in FIG. 5 differs from that shown in FIG. 3 in that the impedance of each element from the winding on the load side to the rectifying capacitor C100 is balanced. The winding on the load side and the rectifier circuit directly connected to the winding have a balanced circuit network configuration from direct current to high frequency, and the circuit network is sparse from the ground over the entire frequency range.

Since there are rectifying diodes D100 and D400, these rectifying diodes D100 and D40
When 0 is on or off, the connection between the rectifying capacitor C100 and the winding on the load side is extremely loose even when stray capacitance is taken into account, and the current flows through terminals f and g to terminal 3. is disappearing. Therefore, all voltages induced in the windings on the load side are equally applied to the rectifying diodes D100 and D400, and no ground signal is generated at the terminals f and g. In addition, it is possible to prevent in-phase component noise from being output to the subscriber line.

In the example, the second rectifier diode D400
However, the present invention is not limited to this, and when the switching element SW100 on the primary winding side is conducting, the impedance is extremely large, and when the switching element SW100 is in the disconnected state, the impedance is extremely small. Any element is sufficient. For example, this goal can be achieved by switching transistors.

Since the DC supply circuit is a circuit installed to support subscribers, it is extremely densely packaged, and requires a noise level of -70 dBm or less. According to the present invention,
The ability to suppress common-mode component noise, which is a cause of noise increase, is a floating DC type that has many advantages.
-Practical application of DC converter type DC supply circuit,
This has a great effect on downsizing the DC supply circuit and reducing power consumption.

[Brief explanation of the drawing]

Figure 1 is a diagram of a conventional floating DC-DC converter type DC supply circuit, Figure 2 is an example of waveforms at each point in Figure 1, and Figure 3 is common-mode component noise generation in the conventional configuration of Figure 1. FIG. 4 is a circuit diagram showing an embodiment of the present invention, and FIG. 5 is a diagram showing the state of leakage current in the configuration of FIG. 4. 100... Clock generation circuit, 101... Pulse width modulation (PWM) circuit, 102... Rectifier circuit,
103...Filter circuit, SW100...Switching element, T100...Pulse transformer, D1
00, D400... Rectifier diode.

Claims (1)

[Claims] 1 Floating type DC-DC that supplies direct current communication current through subscriber lines connecting exchanges and terminal equipment
In a converter-type DC supply circuit, both ends of the winding on the load side of a pulse transformer and both ends of a rectifying capacitor are connected via a semiconductor element that allows current to flow in only one direction. circuit.