(Editor's note: to see a linked list of all entries in the Power Tip series, click here.)
(Additional editor's note: We've updated this Power Tips entry to include a short, informative, no-nonsense video by the author. The video player is at the bottom—so check it out.)
Have you ever come across the need to generate an isolated power supply for gate drive, isolated sensing or communication circuits? In this Power Tip, we will take a look at a circuit that can do this with minimal parts count, complexity, and cost. This circuit finds use when you have a low input voltage available and the powered circuits allow some (five percent) supply-voltage variation.
Figure 1 shows one example of this technique. It shows an IC especially developed for this requirement; however, any synchronous buck circuit that allows sink operation can be used. This circuit known, as the asymmetrical, half-bridge flyback (or flybuck), operates much like a synchronous buck regulator.
Figure 1: A synchronous buck provides an isolated power supply.
(Click on image to enlarge)
A FET totem pole connected to the input voltage feeds an inductor-capacitor filter. The output of the filter is then regulated through a voltage divider and through the negative input of an error amplifier. The error amplifier controls the duty factor of the FET totem pole to maintain a DC voltage at the sense point.
The voltage on C6 is close to the duty factor times the input voltage. Like a buck power stage, the voltage-seconds on the inductor must equal zero. This circuit, however, adds a coupled winding to the inductor and uses a diode to rectify the reflected inductor voltage when the low side FET is on.
Since the voltage across the inductor during this time is equal to the output voltage, ideally, the output of the circuit will be regulated. Therefore, differences in the voltage drops in the primary and secondary will degrade regulation. In this circuit, voltage regulation with load will be significantly impacted by the forward voltage drop of the diode D1. However, the diode can be replaced with a FET to improve load regulation.
Just as with a coupled-inductor SEPIC, parasitic components in this topology can impact circuit performance. During the on-time, the circuit is pretty benign and most of the current flows in the magnetizing inductance of the coupled inductor, T1, charging C6. The output capacitor, C3, provides load current.
However, during the off-time, the two capacitors are placed in parallel through the coupled windings of the inductor. These capacitors have different voltages and the only thing that limits the current flow between the two is the parasitic components in the loop. These parasitics include the ESR of the two capacitors, the winding resistance of the coupled inductor, the resistances of the low-side MOSFET and diode, and the leakage inductance of the coupled inductors.
Figure 2 shows simulated currents at different leakage inductance values. The top set is the current in primary of T1 and the bottom set is the current in the output diode, D1. The leakage inductance is varied from a very tightly coupled inductor of 10 nH to a very loosely coupled one of 1 µH. In the tightly coupled case, peak current is much higher and is essentially limited by the resistances throughout the loop.
Figure 2: Low leakage increases circulating currents.
(Click on image to enlarge)
In the loosely coupled case, peak currents are much lower. The higher leakage helps to improve efficiency of the power supply by reducing the RMS currents. Figure 2 shows the comparison. A loosely coupled inductor has as much as a 50% reduction in current flow, which reduces losses by 75% in some components. The down side to loosely coupling is that the regulation of the output voltage is degraded.
Figure 3 shows load regulation results for a converter much like Figure 1. If the load current is constrained, this converter provides “good enough” regulation in many cases. The impact of the diode-junction voltage variation, as well as ringing, is seen at light loads.
Figure 3: Flybuck load regulation Is good enough in many cases.
(Click on image to enlarge)
A minimum load or Zener clamp may be required to reduce these light-load effects. With heavy loading, the circuit parasitics degrade the regulation. Consequently, reducing them improves results. For example, replacing the diode with a synchronous switch significantly improves the load regulation.
To summarize, a flybuck converter is an attractive topology which solves the need for an inexpensive, simple, isolated power supply that can tolerate some variation (5 to 10%) on the output voltage. At output of 5 V, efficiency can be good (80%) with diode rectifiers, and can be further improved with synchronous rectifiers.
Please join us next month when we will discuss minimizing the effect of distributed capacitance in a transformer. For more information about this and other power solutions, visit: www.ti.com/power-ca.
Chen and Chen; “Small-Signal Modeling of Assymetrical Half-Bridge Flyback Converter,” IPEMC 2006.
About the author
Robert Kollman is a Senior Applications Manager and Distinguished Member of Technical Staff at Texas Instruments. He has more than 30 years of experience in the power electronics business and has designed magnetics for power electronics ranging from sub-watt to sub-megawatt with operating frequencies into the megahertz range. Robert earned a BSEE from Texas A&M University, and a MSEE from Southern Methodist University.
"In the past, I have seen some self oscillating dc/dc converters. Will they be simpler as compare to this?"
yes hm, just look Bob Boschert's first switching mode power-supply some thirty years ego, that was the basis of his company "Boschert Computer products, he used 2 [two] transistors for a 25W off line power-supply....
I could not find the details about the controller TPS55010 indicated in the picture. Is it a new device, not launched yet?
Also I would be interested to learn about asymmetrical half-bridge flyback (I know flyback, synchronous buck but not "flybuck"). Unfortunately, I don't have access to IEEE and can't refer the paper mentioned in the "Reference".
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.