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.

PMP30446: modification for 6A output current

Part Number: PMP30446

TI PMP30446 reference design is 780V to 390V 1.5A DC/DC 585W buck converter. We have an application with the same input range (3 phase 480VAC, which is 680VDC after rectifying), but we need 90V 6A min. at the output. In fact, output voltage can be within 70V and 150V range, but the current must be 6A min.Please suggest how to modify your reference design to obtain required output.

TIDT177.PDFTIDM559.PDF

  • Hello Robert,

    Since the PMP30446 is a Buck converter, in theory you don't have to modify the working principle to get 90V on the output, just changing R41 and recalculate for the voltage you need. By the way, here the output current is pretty much higher therefore you need to consider several aspects.

    Please also note that the output terminals do not have any reference to primary ground, so you have to consider this stage as "floating".

    Here are the components that should be recalculated:

    1) Mosfets Q1 and Q4: the peak maximum voltage on each FET is Vin(max)/2, so you can use still 600V rated components. Of course the current is 4x, then please select a FET capable of supporting this current. Very likely you will need a heatsink, because the switching losses will be also higher.

    2) Diodes D1 and D2: the current stress here needs to be also analyzed for the same reason of point 1).

    3) Please consider that this converter works in critical conduction mode CrM (boundary between CCM and DCM); you may want to recalculate the coupled inductor L2 according to the switching frequency you want, and the maximum peak current (which is actually 2 x Iout, due to CrM mode).

    4) The turn ratio primary (Np, pins 1-8 to 11-12, or 1-2 to 5-6) to auxiliary winding (Na, 3-4) of L2 should be recalculated as well. The actual inductor has a ratio of Np/Na=16:1. When Vin = 780V and Vout = 390V, during Ton we have on pins 3 and 4 a voltage = 1/2(Vin-Vout) / Npa = 1/2(780-390) / 16 =12.19V; during Toff, we have - 1/2(Vout) / 16 = - 12.19. You can then calculate the ratio to get the same voltage during Toff, where the voltage during Ton will be a consequence. Again, for example, in order to get 12V during Toff, and Vout = 90V, then the turn ratio should be 90V / 12V = 7.5. During Ton you will have on this winding: 1/2(680 - 90) / 7.5 = 38.66.

    5) As you can see from point 4), high ratio of Vin(max) / Vout involves high switching losses and stress over pin 5 of U3 (ZCD). This is not a problem, but this stage delivers its best when Vout is pretty high and the current low.

    7) A new value of the resistors R22//R23 is needed because of the current being higher: please consider also higher loss on these current sense resistors

    8) Finally the output capacitors C6, C7, C12 and C13 should be modified to support the transient response specification, as well as ripple current.

    9) Small fine-tuning in the loop compensation values (C40, C20 and R39) in order to get safe phase and gain margin.

    Best regards,

    Roberto

  • Hello Roberto,

    Could you give any more specific tips regarding 3)? How to choose correct inductor L2 parameters at given switching frequency? Why for PMP30446 application a 160uH inductor was chosen?    

    Also regarding to your calculation in 4), to get auxiliary voltage Vaux = 12V during Toff with Vout = 90V, the ratio Np/Na shouldn’t be Npa = ½*Vout/Vaux = ½*90/12 = 3.75? Therefore voltage on auxiliary winding during Ton would be Vaux = ½*(Vin-Vout)/Npa = 78.66.  

    Best regards,
    Pawel


  • 750317489_trafo.pdf

    Also, in the attached L2 documentation from 10/21/2021 the inductor has 320uH. Which one (160uH or 320uH) was used in the final design? 

  • And how to choose an inductance of auxiliary winding? 

  • Hello Pawel,

    The part number of L2 inductor is correct (750317489) but the revision is not updated. The rev. of this inductor should be R01, rather than R00, see:

    750317489r01 Prelim Spec Sheet.pdf

    Regarding the other two questions, it is possible to calculate the value of the coupled inductor with the following method:

    Select first the minimum switching frequency you want to allow the converter to run. In my case it was 22 KHz. This will be the switching frequency at minimum Vin and maximum load. Then calculate L (inductance of the single winding) with the following formula:

    L = (1-Vout/Vin_min)*Vout*Eff/(8*Iout_max*Fsw_min).

    In my case I have: Vin_min=450V, Vout=390V, estimated efficiency (Eff)=0.95, Iout_max=1.5A, Fsw_min=22KHz. This makes L = 187.12uH. Please subtract here the external 15uH inductance, then you get 172.12 úH, which is close to the value I selected of 160uH.

    You may want to recalculate your inductance according to the formula above. Then, when you calculate L, you can also get the maximum switching frequency by replacing Fsw_min with Fsw_max, Vin_min with Vin_max. The efficiency can stay the same, since this has small impact on the calculation.

    Voltage on the output of the auxiliary winding:

    The formulas 1/2(Vin-Vout) / Npa (during Ton) and  - 1/2(Vout) / Npa (during Toff) are still valid to calculate the positive and negative output voltages on auxiliary winding. 

    Auxiliary winding inductance: you don´t need to calculate it, just only the turns ratio.

    Best regards,

    Roberto

  • Dear Roberto, 

    thank you for your help. I have another question. Could you provide a documentation for T2 transformer? 

    Best regards,

    Pawel 

  • Dear Pawel,

    I don´t have the internal structure of the transformer (type of core material, number of turns, thickness of the wires, isolation layers etc.) but I do have the datasheet, please have a look to it.

    750314878r01 spec.pdf

    Best regards,

    Roberto