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.

Cost effective AC/DC solutions on 96W Output

Expert 2375 points
Other Parts Discussed in Thread: PMP10804, UCC28631, UCC28632, UCC28633, PMP9643, UCC28630

Below are the specs as tried to check from web that Ti does not have EVM available.

VAC: 110V ~ 265V

Vout: 24V@4A

During no load, power supply should only provide 0.1W ~ 0.5W of standby output which during 0.5W should meet the efficiency of >80%. Please advise.

  • Thanks for your question.

    While we don't have an exact reference design that meets those design requirements, we have several designs that are close to highlight the performance of our parts.

    PSR Flyback: Recommend UCC28630, and PMP10804 is the closest design.  This design doesn't run at the high input voltages, but can be with slight modification of the power stage

    SSR Flyback: Recommend UCC28600, UCC28610 and LM5023.  PMP10100 has a close output (19.5V@3.84A) with universal input.  This design can work with your requirements with slight modifications, and can be used to get a close estimate of performance. 

  • Hi Eric,

    Can you help to check since PMP10804 is the closest design, how do we achieve the input VAC: 110V ~ 265V?
  • PMP10804 should already be designed for the higher current that will flow at low line, so to convert to wide input range requires checking for sufficient voltage rating on the input bulk caps, primary FET and output rectifier.

    The input bulk caps C2 & C4 are rated for 200 V, so they would need to change to 400-V types. Similar total capacitance (182 uF) will still be required, so the caps will have to get larger in size.

    The primary FET is already 650-V rated, so it is probably already ok for high-line operation. The transformer turns ratio would need to be checked to be sure. This is not explicitly stated in the documentation, but can be inferred from the switching waveforms on page 13 of the test report. The reflected output voltage looks to be about 120-130 V, superimposed on the peak-rectified AC input. So Nps ~ 5.3 (130 V / (24 V + 0.5 V)). So at 265 V ac, the peak input voltage + reflected output voltage ~ 505 V, allowing up to 150 V for leakage spike clamping.

    The present clamp (D2/D4) is too high - this will clamp ~ 280 V above Vac peak, or 280 + 375 = 655 V, and will be higher due to initial spike before the clamp can take effect. For Vreflected ~130 V, a clamp of ~180 V would be recommended, so D2, D4 should change to 2 x 91-V , or 100 V + 82 V. Even then, depending on the transformer leakage, the Vds spike will need to be checked at 265 V ac, it might still be close to the 650-V rating. If so, a small high-voltage capacitor (10 - 100 pF) in the C5 position would help reduce the Vds spike.

    The output diode D1 is rated for 100 V, but for operation up to 265 V ac, this would need to increase to at least 150 V.

    Vdiode = (265 * sqrt(2) / 5.3) + 24 = 95 V + margin for spikes.

    Alternatively, PMP9643 could also be considered. This is rated for 65 W (24 V @ 2.7 A) continuously, but can deliver up to 140 W transiently. But if there is sufficient additional heatsinking or airflow, that design can deliver up to ~95 W continuously - above this, the UCC28630 overload timer will trip out, depending on the overload power level, duration and repetition rate. If the UCC28630 is changed to a version with no overload timer (UCC28631, UCC28632 or UCC28633, depending on the required feature mix), then the design can be run up to the CC-mode limit of ~5.5-6 A (130-140 W), if there is enough additional heatsinking and airflow. For continuous operation at the higher power, the transformer may need to be made slightly larger, again this depends upon the application ambient temperature and airflow.

    I hope this helps answer your question, if so plase click the "verify answer" button.

  • Thanks Bernard
  • Hi Bernard,

    If would like to make changes on the output at 24V@10A? Is it possible to do so?
  • That's big increase from 96 W to 240 W!

    Is this a continuous rating, or a transient peak? 240 W is a lot of power, probably too much for a single-stage flyback.

    The peak and rms currents would be very high, eff would not be good.

    The input AC current would be very peaky with very poor PF (power factor). In fact, depending on the applicatio, PFC would probably be required at this power level.