Miscellaneous active electrical nonlinear devices – circuits – and – Gating
Reexamination Certificate
2000-09-14
2003-09-02
Callahan, Timothy P. (Department: 2816)
Miscellaneous active electrical nonlinear devices, circuits, and
Gating
C327S376000, C327S377000, C327S387000, C363S020000, C363S021030
Reexamination Certificate
active
06614288
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to switched power supplies, and more specifically, to a circuit which controls the operation of a switch in such a power supply. The circuit uses an adaptive control loop to change the state of the switch at the optimum time to minimize switching stress and power loss. The inventive circuit may also be used to control the change of state of a switch in other types of power controlling circuits in which the optimal switching time is dependent upon the voltage through the switch.
2. Description of the Prior Art
Switching or “switch mode” power supplies use a semiconductor device as a power switch to control the application of a voltage to a load. In a flyback power converter, for example, when the power switch is in the “on” position (i.e., the device is conducting), the voltage is applied across a transformer, which stores energy in its magnetic core as the current through it increases. When the power switch is in the “off” position (i.e., the device is not conducting), the inductor voltage increases as the circuit attempts to compensate for the reduction in current by generating a back emf. The inductor voltage “flies back” above the input voltage and is typically clamped by a diode (or the body diode of a second switching device) at the output voltage level. With the diode being conductive, the stored energy in the transformer is transferred to an output capacitor and the other elements of the load. The current through the transformer then decreases until the energy in the magnetic core is depleted. The power switch is then turned back on to start another cycle. Just before being turned on, the voltage across the power switch is greater than the input voltage. As the power switch is turned on, the voltage falls and current in the switch increases, resulting in a loss of power.
The switching operation of the power switch is often controlled by a clock signal, with the duty cycle of the switch (the relative “on” versus “off” time during the switching period) determining the output voltage of the circuit. This method of controlling the switch operation is termed “pulse width modulation” (PWM).
In some configurations of switched mode supplies, the load of the circuit may include a resonant network (typically an inductor and a capacitor) which acts to smooth the output signal and provide a back emf in the form of a sinusoidally varying waveform. This provides a zero-voltage or zero-current condition through the power switch which can be used to define the desirable switching point(s).
A “half-bridge” (or Class-D inverter) is another type of switch mode power converter topology which is used to provide a dc source. In such a configuration, a center-tapped dc source is provided by using two smoothing capacitors in conjunction with two switching devices. The switching devices are operated so that they are, alternately switched “on.” This can be accomplished by driving one of the switching devices with a clock signal which is an inverted version of the clock signal for the other device.
The switching devices in switch mode power converters are subjected to high stresses and potentially high switching power loss as a result of the switch being changed from one state to another while having a significant voltage across it. These effects increase linearly with the switching frequency of the PWM. Another drawback of switched power circuits is the electromagnetic interference arising from the large di/dt and dv/dt caused by the switch mode operation.
The noted disadvantages of switch mode power circuits are reduced if each power switch in the circuit is caused to change its state (from “on” to “off” or vice versa) when the voltage and/or current through it is zero or at a minimum. Such a control scheme is termed “zero-voltage” and/or “zero-current” switching. In the case of switching at a minimum voltage, the control scheme is termed “low-voltage” switching. It is therefore desirable to switch the switching device at instances of zero or minimum voltage in order to reduce stress on the switch and power loss of the power supply or converter.
In Zero Voltage Switching (ZVS) power converters, during each switching cycle the switch voltage is driven to zero by the action of the inductive load, and ideally the switch is then turned on. ZVS Resonant converters have a large LC tank to ensure that there is always sufficient inductive energy to drive the switch voltage to zero. In contrast, active Clamp ZVS Forward and Flyback converters have a relatively small inductance between switch and transformer, with smaller inductive energy and hence lower losses than resonant converters. However, the small inductance between switch and transformer causes two problems. Firstly, there is only a small amount of energy stored in the inductance to ring the switch voltage down towards ground. If there is insufficient energy for the switch voltage to reach zero, then the switch should ideally be turned “on” when the voltage is at a minimum, before it rings back up again. This corresponds to Low Voltage Switching (LVS) operation. Secondly, the LC tank circuit formed by the small inductance and switch capacitance has a high resonant frequency. This means that there is only a small window of time for turning the switch “on” at zero (or low) voltage. If the switch state is changed too soon or too late, there will be a potentially significant power loss associated with the switching operation.
One method of controlling the switch action in order to reduce power losses is for the switch control circuit to use a fixed time delay before turning the power switch on, where the delay is chosen to be approximately long enough for the switch voltage to have reached zero volts or its minimum value. Examples of this approach are found in the Unitrode UCI 875 range of Phase Shift Resonant Controllers, and the UC1580 range of Active Clamp/Reset PWM controllers, manufactured by the Unitrode Corporation. However, a disadvantage to this approach is that it allows for no variation in the amount of the fixed delay to take into account differences in the load, operating conditions, or parts tolerances arising in a high volume production environment. This situation can cause the power supply or converter to operate in a sub-optimal manner, because if the switching event occurs even a small amount of time after the point of minimum switch voltage, the switch voltage may be much higher, resulting in a large power loss. Another disadvantage of this approach is that the fixed delay must be designed into the switch controller. This creates a burden on the designer to optimize the component values for the circuit.
An improved approach to controlling the action of the power switch is to detect a zero voltage condition across the power switch. Examples of this approach are found in the Unitrode UCI 861 range of Resonant-mode Power Supply Controllers and the UCI 872 range of Resonant Lamp Ballast Controllers. In such devices, the power switch field effect transistor (FET) drain terminal is connected via a high value resistor to a “Zero Detect” pin, which generates a synchronization pulse (for the control circuit oscillator) when the resonant waveform falls to zero.
While this approach is an improvement over circuits which change the state of the switching device based only on a fixed delay, it does have a disadvantage. This is because the zero detect method relies on the existence of a zero voltage condition to trigger the switch control signal. The resistive element used as part of the switch voltage sensor senses the actual voltage across the switching element and hence does not generate a switch control signal until that voltage is exactly zero. Thus, the “Zero Detect” method found in the Unitrode devices requires the converter to be designed with sufficient inductive energy to ensure that the switch voltage always falls to zero at the end of each cycle. If the switch voltage does not fall to zero, then the Unitrode controller may default to a mode of operat
Dagan Marc
Sawtell Carl Keith
Smith David Anthony
Astec International Limited
Callahan Timothy P.
Coudert Brothers LLP
Luu An T.
LandOfFree
Adaptive drive circuit for zero-voltage and low-voltage... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Adaptive drive circuit for zero-voltage and low-voltage..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adaptive drive circuit for zero-voltage and low-voltage... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3098762