1. Flyback Power Supplies

When choosing a system topology that can produce multiple outputs from a single power supply, a flyback power supply is a smart choice. Since the voltage across each transformer winding is proportional to the number of turns in that winding, each output voltage can be easily set by the numberof turns. Ideally, if one of the output voltages is adjusted, all the other outputs will be scaled by the number of turns and remain stable.

2. How to Improve the Cross-Regulation of Flyback Power Supplies

In reality, parasitic components will collectively reduce the load regulation of the unregulated outputs. I will further explore the effects of parasitic inductance and how using synchronous rectification instead of diodes can dramatically improve the cross-adjustment of a flyback power supply.

For example, a flyback power supply can each produce two 1 A 12 V outputs from a 48 V input, as shown in the simplified simulation model in Figure 1. The ideal diode model has zero forward voltage drop and negligible resistance. The transformer windings have negligible resistance and only parasitic inductances in series with the transformer leads are modelled. 

These inductances are the leakage inductance within the transformer, and the parasitic inductance within the printed circuit board (PCB) printed wires and diodes. When these inductors are set, the two outputs track each other because the full coupling of the transformer drives the two outputs equal when the diode conducts during the 1-D portion of the switching cycle.

Figure 1 This flyback simplified model simulates the effect of leakage inductance on output voltage regulation

Now consider what happens when you introduce 100 nH of leakage inductance into the two secondary leads of a transformer and put 3 μH of leakage in series with the primary windings. These inductors can create parasitic inductance in the current path, which includes leakage inductance inside the transformer as well as inductance in the PCB and other components.

When the initial field effect transistor (FET) is turned off, there is still current flowing in the initial leakage inductance, while the secondary leakage inductance is turned on for an initial condition of 0 A for 1-D cycles. A base voltage appears on the transformer core and is shared by all windings. This base voltage causes the current in the primary leakage to ramp down to 0 A and causes the secondary leakage current to ramp up to transfer current to the load. When two heavily loaded outputs are present, current flows continuously throughout the 1-D cycle and the output voltages are well balanced, as shown in Fig. 2.

However, when one heavily loaded output and the other lightly loaded, the output capacitor on the lightly loaded output tends to undergo peak charging from that base voltage; as the current rapidly rises back to zero, its output diode will stop conducting. See the waveforms in Figure 3. The peak charging cross-modulation effects of these parasitic inductors are usually much worse than those caused by the rectifier forward voltage drop alone.

Fig. 2 Secondary winding current flows in both secondary windings when heavy loads are applied to the outputs

 Figure 3 Heavily loaded secondary 1 and lightly loaded secondary 2 with base voltage peak charging the output capacitor of secondary 2

Regardless of the load, the synchronous rectifier helps mitigate this problem by forcing current into both windings throughout the 1-D cycle. 

Figure 4 shows waveforms with the same load conditions as Figure 3, but with an ideal synchronous rectifier instead of an ideal diode. Since the synchronous rectifier stays in good condition after the base voltage is reduced, the two output voltages track each other well even with a severely unbalanced load.

Although the average current in secondary 2 is very small, the root mean square (RMS) content can still be quite high. This is because, unlike the ideal diode in Fig. 3, the synchronous rectifier can force continuous current flow during the entire 1-D cycle. Interestingly, the current must be negative for most of this cycle to ensure a low average current.

Obviously, you sacrifice better regulation to achieve higher cycle currents. However, this does not necessarily mean that the total losses will be higher. The forward voltage drop of a synchronous rectifier is usually much lower than that of a diode, so the efficiency of a synchronous rectifier at higher loads is usually much better.

3. Effect of leakage inductance on cross regulation

You can see the effect on cross regulation in Figure 5. the load on output #1 remains stable at 1A, while the load on output #2 fluctuates between 10 mA and 1A. At loads below 100 mA, when the diode is used, the cross regulation is severely reduced due to the effects of peak charging of the base voltage.

Keep in mind that the reason you only see the effect of leakage inductance is because an ideal diode and an ideal synchronous rectifier were used in these simulations. The advantage of using a synchronous rectifier is further accentuated when the effects of resistance and rectifier forward voltage drop are considered.

Therefore, to achieve excellent cross regulation results in a multi-output flyback power supply, consider using synchronous rectifiers. In addition, you may improve the efficiency of the power supply.

Fig. 4 Cross regulation between two outputs

where the 1-A load on output 1 remains stable while the load on output 2 is constantly changing, thus highlighting how the synchronous rectifier mitigates the effects of leakage inductance. 

Aipu Electronics - Modular Power Supply Production Experts reminds you that this article is from the Internet, we save the neutrality of the views in the article, we do not provide any express or implied warranty for the accuracy or reliability of the content contained in the article, and we are not responsible for the views of the article. Copyright belongs to the original author.