In Power Tip 61, we discussed controller duty factor limitations for very high conversion ratios in non-isolated boost converters. We found that the maximum duty factor limitation in the controller limited how high of a boost voltage you could create. So what do you do when you run up against this limit?
At low output powers, you could consider discontinuous operation of the boost, or you might consider adding a charge pump. But as power increases, these approaches become less viable, and you are forced to use a coupled-inductor topology such as boost or a flyback.
The figure below presents the coupled-inductor boost. While the power switch is turned on to increase the magnetomotive force (MMF) in the inductor, the output capacitor delivers current to the load. During this time, the voltage on the anode of D1 is negative, and it is set by the input voltage and the inductor turns ratio. When the switch is turned off, the MMF in the inductor drives the anode of D1 positive until it conducts to recharge the capacitor and deliver power to the load.
Note that in continuous operation mode, it is the MMF in the inductor that is continuous and the primary and secondary currents are pulsating. With the switch on, the MMF is equal to the primary inductance times the primary current squared. When the switch is turned off, the current is the same in both windings and the ratio is reduced by (1+N).
The following figure shows the continuous flyback converter. In this converter, the power switch is turned on to build MMF in the primary inductance while the output capacitor supplies load. When the power switch is turned off, the transformer voltage reverses polarity and current flows through the output diode to recharge the output capacitor and power the load.
Again, in continuous operation, it is the transformer MMF that is continuous, and the primary and secondary currents are pulsating -- in this case, all the way to zero.
With large conversion ratios, the choice between the two is subtle. The table below presents a comparison of the circuit stress between the two for a conversion of 5 to 200 volts.
Boost has slightly lower circuit stress but flyback has current limit. (See full-size table.)
This table was calculated based on equal voltage stresses on the power switches. Vr in the table is the reset voltage on the primary winding when the power switches are off. The boost has slightly lower turns ratios, because the boosting winding is stacked on the input voltage, which is similar to an auto-transformer connection. With the lower turns ratios comes a slight reduction in diode voltage stress and peak switch current. Because of this, the boost may be slightly more efficient.
The numbers in the table are very similar, and the best choice of topology is not obvious. The determining factor may be the desired response for an over current condition. If you short the output of the boost, there is nothing to limit the diode current other than the capacity of the input source. With the flyback, there is not a direction connection back to the source, allowing the controller to protect against the fault condition.
To summarize, boost controllers have a limited conversion ratio set by the minimum off-time of the controller and the operating frequency. You will need to consider alternate approaches such as charge pumps, discontinuous operation, or higher power tapped inductor topologies. While the coupled-inductor boost provides slightly better efficiency, the flyback can offer more over current protection and isolation, if desired.
Please join us for the next Power Tip where we will look at testing power supplies with high di/dt load transients.
See more about this and other power solutions.