Sometimes you are faced with a need for a simple, low-power isolated output voltage from a wide-ranging input source. Regulation may not be important, but cost and board area may be. A good solution to these requirements is a Fly-Buck power supply, which is simply a buck regulator with a coupled-winding.
Regulation is achieved by rectifying the secondary winding of the coupled inductor, when the low-side primary switch is on. This reflects a voltage to the secondary that is set by the output voltage of the buck times the turns-ratio of the coupled inductor. For an overview of the circuit operation, see Power Tip 43.
Figure 1 shows how simple a Fly-Buck can be. In this design, the sync-buck power switches are contained within the control IC and it only takes a handful of discrete parts plus a transformer to complete the design. The real trick for a successful design is the specification or selection of the coupled inductor. In particular, requirements for turns, leakage inductance, and magnetizing inductance need to be established.
In the circuit shown in Figure 1, the turns-ratio of the transformer is established by the primary and secondary output voltages. It will simply be the ratio of the primary voltage to the secondary voltage plus allowances for the diode (D1) voltage and any winding resistance drops. In this case, the relationship between the primary output voltage and the minimum input voltage needs to be understood. Clearly, the buck cannot provide an output higher than the input. If the two are too close together, the circuit may not function properly. You may be limited by the maximum duty cycle of the control since the output voltage is approximately the duty factor times the input.
The second challenge is in the circuit operation at extremely high duty factors, where the currents can be become quite high. These high currents can result from both charge conservation and the basic circuit operation. From charge conservation, the output capacitor is only charged when the switch node is low. During the remainder of the period, it sources the load current. On an average basis, to conserve charge:
This result is plotted in Figure 2 where Icharge/Iout is plotted versus D. At duty factors above 75%, the ratio is above three and climbs quite rapidly with increasing duty factor. The high current impacts regulation of the secondary output. During diode conduction, the coupled inductor places a reflected primary output voltage across the series combination of the coupled inductor leakage inductance, series parasitic resistances and the output filter capacitor.
The current waveform in the secondary shown in the bottom traces of Figure 3 are strongly influenced by the leakage inductance which will impact regulation. The leakage inductance determines how quickly the current in the secondary winding can ramp. With small amounts of leakage inductance, the currents ramp quickly to a high value which charges the output capacitor quickly. As the inductance is increased, the current rise is slowed which can result is less charge being supplied to the output capacitance and less output voltage.
Figure 4 shows the simulated impact of the leakage inductance on the secondary output regulation. This chart plots primary output voltage and secondary output voltage as a function of duty factor and leakage inductance. This was based on a 1:1 transformer with a 2.5 uH primary inductance and varying amounts of leakage. The input voltage was 5 V. The primary was loaded with 1 A of current and the secondary was loaded with 0.2 A.
The first curve is the primary output voltage, which shows a linear relationship between duty factor and output voltage. The remainder shows that there is not a linear relationship for the secondary output voltage.
There are two things that are degrading the secondary regulation. On the left, at lower duty factors, the secondary output voltage is approximately one diode drop less than the primary voltage. This could be improved with synchronous rectification. On the right, at higher duty factors, the shorter conduction time increases the peak currents and the impact of the leakage inductance becomes significant.
With large amounts of leakage, the circuit is probably not usable beyond 50% duty factor or a ratio of 2:1 between input and output. With a nominal amount of leakage, the circuit performed well up to 75% or 1.33:1. Finally, with a heroic leakage inductance, the circuit is good to 83% duty factor or a voltage ratio of 1.2:1. It should be noted as shown in Figure 2, the peak and RMS at high duty factors can be quite high. These are stongly influenced by parasitics and the easiest way to understand them is through simulation.
To summarize, the Fly-Buck is a convenient choice for a simple isolated bias voltage, but you need to be careful when considering running at high duty ratios. Peak currents can become quite high. Controlling leakage inductance allows you to push the duty factor, but anything much more than 80% is probably impractical. Please join us for the next Power Tip where we will look at electromagnetic interference (EMI) considerations in switching power supplies.
Check out TI Power Lab Notes for a designer’s prospective on his power supply designs.
For more information about this and other power solutions, visit: www.ti.com/power-ca.