Basic Components
Bipolar and MOS transistors are used to provide the switching function.
| |
Bipolar | MOS |
| Drive Circuit |
Current |
Voltage |
| Breakdown Voltage |
Higher | |
| Cost | Lower | |
| Switching Speed |
|
5 - 10 times faster |
Capacitor
Because enormous amounts of charge must be transferred during the switching cycle, electrolytic capacitors are required. However, the ESR (equivalent series resistance) of these can be quite high and may pose some problems. If the ESR must be lowered, a tantalum capacitor of about ¼ the value can be placed in parallel.
Diode
http://www.irf.com/technical-info/whitepaper/murdiodes.pdf
The diode rectifier is often the highest source of loss in a switch mode power supply. The most important parameters are: forward voltage drop and reverse recovery time. For low voltage outputs, Schottky diodes are preferred.
| Diode Type |
Forward Drop |
Reverse Recovery |
Forward Recovery |
| Fast Recovery |
1.0 |
150 nSec |
1050 nSec |
| Ultra Fast |
0.9 | 75 nSec |
50 nSec |
| Schottky | 0.5 |
<1 nSec |
0 |
Inductors
http://www.jwmiller.com/pdf2/How_to_Select_Power_Inductors_for_SMPS_EP_June%2020031.pdf
http://www.circuitprotection.com/coev/SpecifyinPowerInductorspaper.pdf
http://www.torotelprod.com/handbook1-5.pdf
If a constant current flows through an inductor, it will generate a voltage drop equal to that determined by Ohm’s law E=IR, where R is equal to the winding resistance. However, if the current changes, the voltage generated across the inductor will change according to Faraday’s law:
This equation suggests that if the current is interrupted in a vary short time span (Dt), the resulting voltage can be quite high. In practice, several thousand volts can be generated when breaking a circuit.
If a fixed voltage is applied to an ideal inductor, the current will rise linearly to infinity. A real inductor however will have the current rise linearly until the flux saturates the core, at which point the inductor essentially becomes a short circuit and the current increases to the limit of the power source. From this we note that an inductor must not be allowed to saturate. This can be accomplished by placing a current limiting resistor in series with the inductor.
Adding a resistor in series with an inductor creates a current
waveform very much like the charging voltage waveform of a capacitor. This is
because the voltage across the inductor is no longer constant and the current
can now only rise to a maximum of
The voltage and current waveforms resemble:
The time constant is given by:
At any given time, the inductor voltage and current is given by:
It takes approximately 5 time constants (5t) for the current (or voltage) to reach its final value. If the charging period is one time constant or less, the current essentially rises in a linear manner.
A diode can be added to provide a discharge path for the inductor when the switch is OFF.
Turning the switch ON and OFF results in a ‘triangle’ waveform, and if the switching time is fast enough, the current waveforms can be approximated by straight lines.
From this we observe that the average current value is somewhat lower than the peak current value. If the average current needs to be increased (or decreased) because of the load, the ON time can be increased (or decreased) accordingly.
During the ON period, energy is transferred from the input power source to the inductor. During the OFF period, energy is transferred from the inductor to the load side of the circuit.
A pulsing current is generally not directly usable by a load. Therefore a capacitor is required to accumulate the excess current and discharge it when required. This forms the basis of the simplest switching converter, namely the buck or step down converter.
Many of the design equations for a switch mode power supply can be derived by examining the inductor current waveforms.
If the inductor current falls to zero, the circuit is operating in the discontinuous mode. If there is always a current in the inductor, it is operating in the continuous mode.
Most SMPS design equations can be derived when the inductor current is just at the continuous/discontinuous threshold.
The relationship between peak inductor current and inductor value can be determined if the slope of the current waveforms can be determined. The size of the inductor can then be determined by finding the relationship between the peak inductor current and the switch OFF time.