Electric power conversion systems – Current conversion – Including d.c.-a.c.-d.c. converter
Reexamination Certificate
2001-08-09
2002-09-03
Berhane, Adolf Deneke (Department: 2838)
Electric power conversion systems
Current conversion
Including d.c.-a.c.-d.c. converter
C363S021150, C363S097000
Reexamination Certificate
active
06445598
ABSTRACT:
TECHNICAL FIELD
This invention relates to a dc-to-dc converter capable of adaptation to variations in load or input voltage.
BACKGROUND ART
As disclosed for example in the U.S. Pat. No. 5,719,755, a typical prior art comprises a direct-current power supply, a transformer having a primary, a secondary and a tertiary winding, a switching device, a rectifying and smoothing circuit, and a control circuit. Connected across the dc power supply via the transformer primary, the switching device is turned on and off under the direction of the control circuit.
The rectifying and smoothing circuit may be either of two different types. The first type comprises a rectifying diode and a capacitor. The rectifying diode is so connected to the transformer as to be reverse-biased by a voltage induced in the transformer secondary during the conducting periods of the switching device connected to the transformer primary, and forward-biased by a voltage inducted in the transformer secondary during the nonconducting periods of the switching device. The output voltage of the rectifying diode is smoothed by the capacitor.
The second type of rectifying and smoothing circuit has a rectifying diode that is so connected to the transformer secondary as to be forward-biased by the voltage induced in the transformer secondary during the conducting periods of the switching device. The output of this rectifying diode is connected to a choke coil, thence to a smoothing capacitor, and a diode is provided to form a closed circuit in combination with the choke coil and the smoothing capacitor.
The voltage regulator with the first type of rectifying and smoothing capacitor is commonly referred to as the flyback or reverse switching regulator, and that with the second type as the forward switching regulator.
There are strong demands for dc-to-dc converters of higher operating efficiency. Improving the efficiency of dc-to-dc converter depends to a large measure upon reduction of the power loss of the switching device. To this end the U.S. Patent cited above employs what is termed a quasiresonant capacitor, which is connected in parallel with the switching device. Connected in parallel with the switching device, the capacitor is gradually charged while the switching device is off, causing a gradual rise in the voltage across the capacitor and across the switching device. With use of a bipolar transistor or a field-effect transistor as the switching device, current will continue flowing therethrough after it is turned off, due to the carrier storage of the semiconductor. With the provision of the resonant capacitor as above, however, the voltage across the switching de vice does not rise too sharply after it is turned off. The result is the reduction of the switching loss, or of the loss of power equivalent to the product of the current through, and the voltage across, the switching device. Also reduced is the voltage surge or the noise when the switching device is turned off.
Preparatory to causing conduction through the switching devices the voltage across the same is gradually reduced by the resonance of the inductance of the transformer primary and the capacitance of the capacitor connected in parallel with the switching device. The switching device is turned on when that voltage is reduced to zero or thereabouts. Thus is accomplished the zero-voltage switching of the switching device, with consequent reduction of the switching loss.
In such a quasiresonant switching regulator, incorporating means for holding the output voltage constant, variations in the voltage requirement of the load manifest themselves as changes in the on-off rate (herein-after referred to as the switching frequency) of the switching device. A drop in the voltage requirement of the load, for instance, results in an increase in switching frequency. A higher switching frequency means that the switching device is actuated a greater number of times per unit length of time. Since the switching device causes a loss each time it is actuated, a loss per unit length of time also increases as the switching device is actuated oftener per unit length of time. Consequently, despite use of the quasiresonant capacitor, the efficiency of the noted prior art switching regulator did not necessarily improve.
It has been known to set a limit upon the switching frequency during operation under a light load, as taught for instance by Japanese Unexamined Patent Publication No. 8-289543. This objective has so far been attained by compulsorily imposing a lower limit on the nonconducting periods of the switching device. The actual nonconducting periods of the switching device are not permitted to fall short of the mandatory minimal nonconducting period thus imposed.
In a dc-to-dc converter with the predetermined minimum required nonconducting period, in the event of a substantive drop in the voltage requirement of the load, the switching device is not immediately turned on, but upon lapse of the predetermined minimum nonconducting period, upon completion of the production of the flyback voltage due to the release of the energy that has been stored on the transformer during the preceding conducting period of the switching device. With the completion of the production of the flyback voltage during the minimum nonconducting period, so-called ringing will occur due to the inductance of the transformer winding and the parasitic capacitance or resonant capacitance of the switching device. The switching device is turned on in the course of this ringing. The voltage across the switching device may be high due to the ringing at the end of the minimum nonconducting period, so that the switching device is turned on when the voltage across the same grows sufficiently low after the expiration of the minimum nonconducting period. This known method of controlling the switching device succeeded in material curtailment of the switcing loss.
It has later proved, however, with the fixed minimum nonconducting period, as has been the case heretofore, the switching frequency has tended to become unstable in the event of fluctuations in the input voltage or in the voltage requirement of the load. Let us consider the case in which the voltage requirement of the load changes from a first, relatively heavy state, such that the flyback voltage is generated longer than the minimum nonconducting period, to a second, relatively light state in which the flyback voltage is generated shorter than the minimum nonconducting period. The instability of the switching frequency has occurred just when the duration of the flyback voltage becomes less than the minimum non-conducting period.
The foregoing discussion of instability in switching frequency will be better understood from a consideration of
FIGS. 6 and 7
. The indicia V
1
in these figures denotes the voltage across the switching device, the voltage being due to the transformer flyback voltage and ringing voltage. The indicia T
1
, at V
13
in
FIG. 6
denotes the predetermined minimum non-conducting period.
Under a relatively heavy load, as represented by
FIG. 6
, the duration T
0
of the flyback voltage is longer than the minimum nonconducting period T
1
. The switching device will then be turned on immediately upon expiration of the flyback voltage, resulting in the continuation of the known self-excited oscillation. Then, with a gradual lessening of the load, the conducting periods T
on
of the switching device will grow less, and so will the durations T
0
of the flyback voltage, until at last the flyback voltage duration becomes shorter than the minimum nonconducting period T
1
.
As will be understood from
FIG. 7
, the switcing device will be inhibited from turning on at the end of the duration T
0
of the flyback voltage when the flyback voltage duration becomes less than the minimum nonconducting period T
1
, as above. The switching device will be turned on when the voltage across the same becomes approximately zero after the end of the minimum nonconducting period T
1
. The nonconducting period of the switching device will increase if
Berhane Adolf Deneke
Sanken Electric Co. Ltd.
Woodcock & Washburn LLP
LandOfFree
Dc-dc converter does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Dc-dc converter, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Dc-dc converter will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2884545