Académique Documents
Professionnel Documents
Culture Documents
EE-4108
Resonant Conversion
Resonant power converters contain resonant L-C networks whose voltage and current waveforms vary sinusoidally during one or more subintervals of each switching period.
These sinusoidal variations are large in magnitude, and the small ripple approximation does not apply.
Some types of resonant converters:
dc-to-high-frequency-ac inverters Resonant dc-dc converters Resonant inverters or rectifiers producing line-frequency ac
Switching Trajectories
Turn-on or turn-off transitions of semiconductor devices can occur at zero crossings of tank voltage or current waveforms, thereby reducing or eliminating some of the switching loss mechanisms. Hence, resonant converters can operate at higher switching frequencies than comparable PWM converters Zero-voltage switching also reduces converter-generated EMI Zero-current switching can be used to commutate SCRs In specialized applications, resonant networks may be unavoidable High voltage converters: significant transformer leakage inductance and winding capacitance leads to resonant network
Current source parallel resonant converters Class E and sub class E resonant converters
and
Vd Vc 0 iL (t ) I L 0 cos 0 (t t0 ) sin 0 (t t0 ) Z0
vc (t ) Vd (Vd Vc 0 ) cos 0 (t t0 ) Z0 I L0 sin 0 (t t0 )
For iL > 0
T+ conducting D- conducting For iL < 0 T- conducting D+ conducting VAB = - 1/2 Vd VAB = +1/2 Vd VAB = -1/2 Vd + Vo VAB = +1/2 Vd + Vo VAB = +1/2 Vd VAB = -1/2 Vd VAB = +1/2 Vd Vo VAB = -1/2 Vd Vo
The voltage applied across the tank VAB depends on which device is conducting and on the direction of iL
There are three possible modes of operation based on the ratio of switching to resonance frequency.
The operating frequency is in between one-half the resonance frequency and the resonance frequency
Output Current as a function of operating frequency for various values of the output voltage
For an input voltage Vd and load resistance, V0 can be regulated by controlling the switching frequency. A constant frequency control can be implemented in a full bridge SLR. The SLR can be used where the output is not a rectified dc; e.g. induction heating where the load appears as a resistance rather than a dc voltage V0
PLR converter appears as a voltage source and therefore, more suited for multiple outputs. PLR converter can step up as well as step down the voltage, unlike SLR which can only step down the voltage. PLR converter do not posses inherent short circuit protection capability. A PLR converter can operate in a large no. of combinations consisting of the states of iL and vC. However, only three modes are commonly used. Discontinuous conduction mode S < 0 Continuous conduction mode S < 0 Continuous conduction mode S > 0
1.
In the discontinuous mode of operation a PLR exhibits good voltage source characteristics and V0 remains independent of i0. Also in this case the output varies linearly with S thus simplifying the output regulation.
2.
In the mode S > 0 the maximum change required in the operating frequency is < 50% to compensate for the output loading for a normalized output voltage of 1.0 It is possible to step-up or step-down the output voltage, i.e., V0 can be less than or greater than 1.0
3.
Output voltage as a function of operating frequency for various values of the output current
This is similar to SLR but the load is connected in parallel with a part of the series capacitor e.g. Cr2 is 1/3rd of total capacitance, C and Cr1 is 2/3rd of C. Has inherent current limiting capacity under short ckt. like an SLR and like a PLR acts as a voltage source, thus regulation of voltage to no load with a high Q tank ckt. is not a problem.
Basic circuit illustrates the operating principle at the fundamental frequency Since the square wave input current is generally supplied by a thyristor inverter, the resonant load must supply the capacitive VARs to the inverter. This implies that the load voltage V01 should lag the current I01, which is possible only at frequency S > 0
The power output of this inverter can be controlled by controlling its frequency.
As the switching frequency fS is raised above the resonance frequency f0, the power output decreases, when Id is held constant by means of a controlled dc supply.
Another possible method is to control Id keeping the switching frequency of the inverter constant. For start-up of a current fed parallel resonant inverter, the load must be in resonance with Cr prior to the inverter operation. This is achieved by means of a precharged capacitor dumping its charge on to the parallel resonant load circuit, thereafter the inverter operation is initiated.
Class-E Converters
A sharply tuned series resonant ckt. results in essentially sinusoidal output current. The input to the converter is through a sufficiently large inductor that ensures the steady state input current is a dc current source Id. The current magnitude depends on the power output. When the switch is on Id + I0 flows through the switch. When the switch is turned off, the voltage across the switch builds up slowly due to C1, thus allowing ZV turn-off. When the switch is off the ckt resonates and the capacitor voltage builds up and reaches the peak and then comes back to zero, at which instant the switch is turned on again. The converter operates at frequency slightly higher than the resonance freq.
Class-E Converters
A high Q 7 is used which results in sinusoidal current; a slight variation in fs is needed to vary the Vo. As fs increases io and therefore vR decreases A small change in fs changes the output voltage.
Avg VT = Vd, avg (VR=VLr=0) D=0.5 Pmax, Ipk 3Id, Vpk 3.5Vd
Optimum mode
Class-E Converters
R< Ropt Vpk decreases with smaller duty ratio, but Ipk increases. Advantages: No switching losses & EMI Single switch topology Sinusoidal output current Disadvantages: High pk voltage and currents in switch and LC elements
Non-Optimum mode
Classifications
Waveforms
Voltage is regulated by varying the switching frequency
A practical circuit
Serious limitations
Waveforms
Comparison of ZCS & ZVS resonant switch converters: Both techniques use variable frequency control to regulate output voltage. ZCS: High peak current > Load current (Vd/Z0) For natural turn-off I0 < Vd/Z0
The inductor current must reverse direction during each switching cycle
Commonly used
EMI Problems
Effect on Machine Insulation Machine Bearing Current Problem Machine Terminal Overvoltage with Long Cable
Machine
CF
Machine Stator Winding Stray Capacitor Coupling to Stator and Rotor by dv/dt Effect
Machine L
With initial current IL0 in inductor, initial voltage VC0 on capacitor and with switch in off state, the capacitor voltage is given by eqn:
0 Lr
sin 0 (t t0 ) ( I L 0 I 0 ) cos 0 (t t0 )]
HighFrequencyLink Inverter
High-Frequency-Link Inverter
High-Frequency-Link Inverter
Why are we interested in resonant mode power conversion? With the earliest switched-mode power converters, it became obvious that higher frequencies allow smaller L's and C's and this, in turn, should lead to smaller, lighter, and (hopefully) less costly systems. The down side to moving to higher frequencies, however, are the problems of greater susceptibility to parasitic capacitance and leakage inductance, greater stress in the switching devices, and increased EMI and RFI. A resonant mode system offers the potential of achieving the benefits while sidestepping many of the disadvantages of higher frequencies.
With a resonant circuit in the power path, the switches can be configured to operate at either zero current or zero voltage points in the waveform, greatly reducing their stress levels; the resonant sine wave minimizes higher frequency harmonics reducing noise levels; and since the circuit now requires inductance and capacitance, parasitic elements may enhance rather than detract from circuit performance. With these benefits, power systems operating in the range of 500 kHz to 2.0 MHz are now practical and in fact are already being produced by a few leading edge manufacturers.
Advantages: 1. 2. 3. 4. 5. Zero current switching Low component stress Low EMI Useful parasitic elements Improved diode recovery
Disadvantages:
1. Greater complexity 2. Higher peak currents 3. New technology learning curve
To bring some order and ease in understanding the broad range of circuit choices which are possible as we move to resonant mode operation, it helps to establish a system to classify resonant topologies by defining the following operating characteristics:
1. Is the load in series or in parallel with the resonant circuit elements? 2. Is the control system a fixed or variable frequency type? 3. Does current ( or voltage) in the resonant circuit flow continuously or is it equal to zero for some portion of the switching cycle? For discontinuous operation, it is also important to know: 4. Is the switching designed for zero current or zero voltage activation, and 5. Does the energy in the resonant circuit flow in only one direction or is there a full cycle before it returns to stop at zero?
Discontinuous Resonance The discontinuous operating mode works by supplying constant packets of energy to the load with the rate, i.e. frequency, determined by load power demand. The most popular and important class of resonant converters with variable frequency and discontinuous current is often called Quasi-Resonance. The general properties of any resonant converter are completely dependent on these options, so they are a good basis to use as a starting point in understanding the principles involved. Series or Parallel Loading Since resonant converters operate by putting energy into a resonant circuit and then transferring some or all of it into the load, There are two ways to accomplish this as shown in Figure 5. If the load is in series with the resonant circuit elements, as in Fig. 5A, we call it a series loaded converter and the operating characteristics tend toward a current source with a high impedance output.
Fixed or Variable Frequency Resonant converters may be configured for either constant or variable frequency operation, but these choices infer significant differences in their operation. Fixed frequency control systems use conventional pulse width modulation to change the output in response to a control input, as shown in Figure 6. This forces a fixedfrequency system to have at least one non-zero switching transition and possibly two, thereby voiding one of the more significant reasons for choosing to use a resonant mode topology. This would usually preclude its use unless system considerations required a synchronized frequency operation.
Variable frequency operation, however, needs to be subdivided by the third classification: whether the resonant circuit current is continuous or discontinuous. A circuit operating in the continuous resonant mode uses the slope of the resonant circuit impedance curve to control the output. As shown in Figure 7, the circuit can operate either above or below resonance but the principle is the same: that the control circuit changes the frequency to move either toward or away from resonance, and thereby controls the amount of energy which is transferred into the resonant circuit and therefore to the load.
While many practical systems have used continuous conduction, variable frequency operation, there are several disadvantages: 1. The non-zero switching adds stress to the transistors. 2. As the frequency approaches resonance, peak currents or voltages can get very high, adding stress to the resonant components. 3. The control transfer function is very nonlinear following the resonant impedance curve.
The major advantage of the continuous mode of operation is that the frequency varies over a much smaller range than with the discontinuous mode.
As indicated in Figure 8, quasi-resonant circuit waveforms are not sinusoidal, but have two essentially linear portions interspersed with two sinusoidal portions. A quasi-resonant converter control loop is usually configured as shown in Figure 9 with a pulse generator driving the resonant circuit at a repetition rate determined by the control circuit. The pulse generator may be set for constant pulse width defined by the resonant circuit or set to sense zero crossing of either current or voltage. With maximum loading and low line voltage, a quasi-resonant converter can approach continuous resonance as a limit when the individual pulses run together .