This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

UCC 28630 5v 65W output rectifier

Other Parts Discussed in Thread: UCC28630, UCC28730, PMP11281

Finished the design of a power supply of (ucc28630) 12V 65W, this uses a diode rectifier output

We are evaluating the design of a 65W 5V.

In the schematic PMP11281 (ucc28730 5V 50W) uses a synchronous rectifier

In the case of the power supply 5V 65W (ucc28630), you can use a diode rectifier output? It is advisable?

Regards

  • Hello Pablo,

    Yes, you can use a diode for the output rectifier in your 5-V 65-W application.
    The advisability of using a diode vs. SR is a matter of meeting the performance targets for the design.
    In the case of an output rectifier, the main targets involved are usually system efficiency and size. Cost is always a consideration, of course.

    Where a system requires high efficiency and/or small size, a synchronous rectifier (SR) is usually chosen to reduce the rectification loss, especially at lower output voltages where a diode drop is a significant fraction of the output voltage. An SR reduces this drop considerably and can reduce or even eliminate the heat sink associated with the rectifier. At much higher output voltages, a diode drop may not be significant enough of an impact to justify the extra cost and complexity of an SR, although space savings may still justify an SR if the power density requirement is paramount.

    Diodes (Schottky, PN, etc) are less expensive and simpler to use, and can be justified if high efficiency and small size are not over-riding factors in the system design. The controllers you referenced will work fine with diode rectifiers, as well as SR, in the output.

    Regards,
    Ulrich
  • Ulrich
    Thank you very much, we will consider it.
    Regards
    Pablo
  • Ulrich

    To use the same components between the power supply 12V 65W and  5V 65W, we must use the output circuit as shown in the drawing.
    Is it advisable?

  • Hi Pablo,

    I can neither advise for nor against your circuit proposal, since I cannot know all of the constraints which your design must satisfy. However, I can make some observations about the output design that you can consider.

    The main trade-off consequence of using the same parts for 12V and 5V is that both designs are de-optimized in order to accommodate the limitations of each other. The main approach here appears to be to use parts with voltage ratings necessary for the 12V output and double the number parts to accommodate the approximate doubling of the current for the 5 V output. I assume that for 12V, half of these parts will be depopulated.

    Although two 5V outputs are shown, I assume that they are tied together, which essentially puts the two inductors in parallel. One should expect that the inductors will not share current equally because of component tolerances and possible layout unbalances, but with care, the current differences may not be severe. However, you should check the worst-case thermal rise to make sure not to exceed any limits. Also note that any unbalanced temperature rise may itself make the unbalance worse. A similar consideration should be made about the Schottky diode current sharing.

    Perhaps the biggest de-optimization is the transformer, which must have a turns-ratio to support the 12V, but current sizing to support the 5V. Also the auxiliary winding voltage will be more than halved at 5V, so to meet minimum voltage at 5V, the AUX voltage will be "over-sized" at 12V with extra dissipation and possibly an additional clamp or regulator needed. To deliver 65W at 5V (13A) the controller's current sense and frequency range operates at one point. To then deliver 65W at 12V with the same controller components will require a different operating point, because the demagnetization slopes are different. There is a lot to consider here. Taking the standard USB-PD approach as an example, an adapter can be made that delivers 3A at 20V, but with all the same components, it is possible to only deliver 15W (3A) at 5V. I'm not certain of this and may be wrong, but an analogy can be made that your design's 12V effective power capability (and transformer sizing) must be 156W (13A x 12V) in order to accommodate both applications. (This may depend on how the current limit is implemented by the controller.) I urge you to evaluate the stresses on all of the components of the overall design, and the operating points of the controller, for both 12V and 5V line and load range to make sure none of them exceed a rating limit or restrict an operating range.

    On the other hand, optimal design for each 12V or 5V application can be done to maximize performance for each at minimal number of parts, although the types of parts will be different. The PCB can be layed out to accommodate either Bill of Material (BOM) and it will be smaller for both designs. The cost tradeoff between separate BOMs and a single BOM must be made after assessing the optimal choices of each component in both design approaches (same parts for both, or different parts for each). I don't think there is an easy way to make this assessment other than brute-force design for all three circuits: combined 12 / 5, 12-V only, and 5-V only.

    I know I've presented mainly generalities here due to the limitations of the information available, the scope of this forum and my own limited experience. You may sense that my general tone is to shy away from the one-size-fits-all approach, because I've seen it attempted many times with few practical successes, except where the option differences were very narrowly limited.  However, that doesn't mean that you can't be successful.  I invite other viewers of this thread to contribute their own experiences, insights and advice on this topic.

    Regards,
    Ulrich

  • Hi Ulrich

    We made a mistake with the designation of the transformer. Let's use two transformer, a transformer for 5V and one for 12V.
    Thank you very much for the answer, it is very useful to us, we will consider it.

    Regards

    Pablo