Electric power conversion systems – Current conversion – With condition responsive means to control the output...
Reexamination Certificate
2001-10-03
2003-05-13
Patel, Rajnikant S. (Department: 2838)
Electric power conversion systems
Current conversion
With condition responsive means to control the output...
C363S127000, C327S498000, C327S439000
Reexamination Certificate
active
06563724
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to synchronous rectifiers and, more specifically, to the rapid, efficient and economical turn-off of a bipolar junction transistor (BJT) used as a synchronous rectifier.
BACKGROUND OF THE INVENTION
DC to DC power converters are typically used to stabilize or isolate a power supply signal from upstream irregularities (i.e., voltage/power surges, momentary power outages, etc.). Various transformer and non-transformer based power converters are known in the art. These power converters typically employ a rectifying device to convert either a transformed AC signal, a chopped DC or a similar signal (depending on the power converter arrangement) into a DC output signal. This output DC signal constitutes a relatively stable power supply signal.
Depending on the range of voltage (and current) for which the power converter is designed, the power converter may be used, for example, in power supplies for personal electronic devices, laptop or personal computers, engineering workstations and Internet servers. While the present invention is particularly concerned with electronic/digital logic circuits, it should be recognized that the teaching of the present invention are applicable to rectifying device operation in any voltage/current range and for any purpose.
For many years the standard power supply voltage level for electronic logic circuits was 5V. Recently, this voltage level has dropped in many instances to 3.3V and 2.5V, and there are plans within the industry to further reduce this voltage level.
As this voltage level drops, however, the forward voltage drop of the rectifying device becomes the dominant source of power loss and inefficiency. For example, a Schottky diode is typically used when a low voltage drop is desired, and a typical Schottky diode has a 500 mV forward voltage drop. This limits the theoretical efficiency of a DC to DC power converter to 80% at two volts output (before other power conversion losses are taken into account). This efficiency limit further drops to less than 67% at one volt output, and 50% at 500 mV output. These efficiency limits are deemed unacceptable.
In addition to concerns about forward voltage drop and other power inefficiencies, power converters and rectifying devices therein are expected to have high power densities. This mandates a higher switching frequency such that less energy is processed in each switching cycle, which in turn permits smaller component sizes. Switching, frequencies have risen from 5 to 20 Khz thirty years ago (where the push was to get above the audible range) up to 100 KHz to 1 MHz at present. Thus, technology that does not support rapid switching is not preferred for most rectification applications.
With respect to known rectifying devices, these include rectifying diodes (PN and Schottky junction in Si, GaAs, etc.) and rectifying transistors (bipolar and field effect). The forward voltage drop of a rectifying diode can be reduced by design, but only to around 300 mV to 200 mV before a point of diminishing returns is reached where increasing reverse leakage current losses outweigh the decreasing conduction losses. This is due to an inherent physical limit of rectifying diodes and does not depend on semiconductor material or whether the construction is that of a conventional P-N junction diode or a Schottky junction diode. For this reason, amongst others, diodes are not desirable as rectifying devices for low voltage level applications.
Rectifying transistors in which transistor driving is in “synchronism” with the direction of current flow across the transistor have increased in popularity due to their favorable forward voltage drops relative to diodes. Typically, the synchronous rectifying transistor is driven “on” to provide a low forward voltage drop when current flow across the rectifying transistor is in a designated forward direction, and is driven “off” to block conduction when current flow across the rectifying transistor would be in the opposite direction.
Both the Bipolar Junction Transistor (BJT) and the Metal-Oxide-Semiconductor Field Effect Transistor (MOSFET) have been used as a synchronous rectifier transistor, also termed a “synchronous rectifier” (SR). Although the BJT has a longer history of use as an SR, the MOSFET is used almost exclusively at present due to its fast switching speed and perceived ease of driving. BJTs are little used at present due to slow switching speeds in general, and a slow turn-off in particular.
The present invention recognizes that the BJT is a conductivity modulated device whereas the MOSFET is not. As a result of this distinction, the BJT can achieve a lower forward voltage drop for a given forward current density and reverse voltage blocking capability. A major technical cost of the lower voltage drop, however, is the presence of a conductivity modulating charge stored during the forward conduction which must be removed before the BJT can sustain a reverse voltage without high leakage currents. Removal of this charge entails a turn-off “storage time” that results in an inherently slower turn-off in BJTs than is achievable with MOSFETs which do not have such a stored charge. The lower conduction voltage of the SR BJT could be used to advantage at lower output voltages, however, if the BJT turn-off speed could be improved (in a cost-effective manner) which is a purpose of the present invention.
Various prior art circuits for turning off a BJT are discussed below after the following definitions and notes. These prior art arrangements include those that turn-off a BJT used as a conventional transistor and those that turn-off a BJT used as a synchronous rectifier.
Definitions and Notes
In the following discussion, and for the remainder of this document, the following definitions and subsequent notes generally apply to circuits using BJTs, as synchronous rectifiers unless otherwise stated or intrinsically implied:
1) A “transistor” is an active (controllable) semiconductor device with at least three electrodes, such that the signal present at one electrode controls the state of conduction between the other two electrodes.
2) The generic term Field Effect Transistor (FET) is used to include the Junction Field Effect Transistor (JFET) as well as the MOSFET, which in turn is used generically for any Insulated Gate FET (or IGFET), including the less commonly used Metal-Insulator-Semiconductor FET (MISFET) device.
3) A “positive” voltage will be that which is normally applied to the collector of a BJT relative to the emitter, or to the drain of a FET relative to the source, when the device is operating as a conventional transistor; a “negative” voltage will be one of reverse polarity.
4) The collector voltage of a BJT will be said to be “above” the emitter voltage when it is of a positive polarity (as defined above), and will be said to be “below” the emitter voltage when it is of a negative polarity, hence “above” is used generally as synonymous with “positive” and “below” is used generally as synonymous with “negative”.
5) A “positive” current flow is one which occurs in normally conducting BJTs and FETs, i.e., from collector to emitter, regardless of device polarity; a “negative” current flow is in the reverse direction.
6) A BJT conducting a given collector current with the collector voltage above the base voltage is said to be in a “linear” region of operation, and is not considered to be in an “on” state in switching applications. In the linear region, the collector voltage falls quickly with small increases in base current. When the collector voltage falls below the base voltage the BJT enters a region of “quasi-saturation” where a significant stored charge begins to accumulate in the transistor, and an increasingly large base current is required to lower the collector voltage further. At some relatively high base current the BJT enters the region of “full”, “hard” or “deep” saturation where further increases in base current do not cause a significant decrease in collector voltage, and eventually the collector voltage be
Adamson Steven J.
Patel Rajnikant S.
LandOfFree
Apparatus and method for turning off BJT used as synchronous... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for turning off BJT used as synchronous..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for turning off BJT used as synchronous... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3067830