• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 3 answers
  • Subscribers 64 subscribers
  • Views 518 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Original question:

UCC28780: ucc28780- still no RUN

UCC28780: ucc28780-CONFLICTING RESULTS FROM sluc666REV c

robin gangopadhya18
Mastermind 6296 points
Part Number: UCC28780

Jaden

A while ago Ulrich had sent in the post his calculations for 12v 4 amp using our range of DC-DC converter input, all GaN approach.

But in redoing the calculations in my version of SLUC664 rev c, I cannot confirm the values he got. In particular, Lpri is absolutely out of whack. I believe this comes about because of the selection of units.

Is efficiency expressed as 93 or .93?

I looked in the expression for Lpri in the UCC28780 datasheet- that gives a value unrelated to anything we have seen before & surely not practically possible.

I took screenshots- But I can send you the spreadsheet as well by email.

Appreciate any help in clarifying the issues.

SLUC66CCALCULATIONS_DC12V4A.pptx.

previous_vals.pptx

robin

  • 0
    TI__Genius 9790 points

    Hi, Robin

    I've assigned this thread to Jaden, please wait for his feedback.

    BR,

    Yunsheng

  • 0 in reply to Yunsheng Qu
    TI__Expert 7080 points

    Hi Robin,

    I found two obvious mistakes in your calculation screenshot .

    1.  Page1 Line22 , Minimum Bulk Voltage Target  is 375V while the minimum line input voltage on Line 19 is 265V , for DC input , I think the minimum bulk voltage should be less than or equal to minimum line input voltage.

    2. Page2 Line38 and 39 . the PRIMARY to Secondary turn ratios  is 24 , and primary turns is 48.  primary turns is too large .

    I have filled your input and output in the calculation tool , but didn't change the Mosfet (it will minor impact calculated Lm) and get Lm is 149uH , that is maybe a  reasonable value. attached it for your reference.

    Thanks.

    UCC28780 Excel Design Calculator (Rev. C) SLUC664C--for robin.xlsx

  • 0 in reply to Jaden Ning
    Mastermind 6296 points

    Thanks much Jaden. Especially for redoing the spreadsheet.

    It is always a bit confusing what to do with input voltage ranges when our whole product line planned will have 370V typ DC from PFC stage. with perhaps 5% variation.

    That gives you upper limit of 388v max to 350 min

    We can impose 200V at startup: will that not for fixing  Rvs1 value only?

    Load variation can be assumed to be min to max: for 12V case you treated, Lpri is very practical value .

    But when we go to 5V  with max load 4 amp,  same input ranges, Lpri shoots to 891 uH with fsw=160 kHz.

    Of all these, what is realistic- I mean, I can make Lpri be anything by changing Fsw but there is no set point for Fsw. It is what the controller decides from various observations at its pins.

    My set up now is for 5V transformer with turns ratio 28:1  for Pri:sec, Aux:sec is 3:1

    I can make Lpri anything by changing the gap.

    Previous calculations with some input values gave us 440 uH & that is where the setup is right now.

    And there is SOME set of values in the input which will give this value of Lpri!

    Right now, we are testing with an external PWML,  so that we can make sure all related voltages & currents are as expected from calculations. So far they are.

    If 440uH is useful for the range  of inputs & power, I would appreciate your comments on these & we can have the controller installed & hope it operates safely 

    robin

  • 0 in reply to robin gangopadhya18
    TI__Genius 9790 points

    Hi, Robin

    Jaden is out of office for several days, Please expect a response by end of day Thursday this week.

    BR,

    Yunsheng

  • 0 in reply to Yunsheng Qu
    TI__Expert 7080 points

    Hi, Robin,

    I am so confused with  what is your exactly input voltage range and output voltage ?  I see 12 V as well as 5V in your description . you have two power modules or you just want one power module to compatible with 12V and 5V outputs  ?  why I ask this question is because you were talking about two transformer design with different turn ratios and inductance.

    Based our experience , for universal input range 90Vac~264Vac , and 5V ~20V output , maximum 65W design . the transformer Lm range is  75uH ~200uH . that is depends on what switching frequency you want the power stage running .

    I don't think Lm=440uH is a good start point . even I don't know what the exactly input voltage for your design . one potential issue is you may can not start up the controller .. since the controller has CS pin fault detection on the first pulse.  the voltage on CS pin after 2us turn on time should be higher than 0.25V . large Lm means current ramp up rate is slow . it may can't touch 0.25V threshold , for more information about CS pin fault detection , please check datasheet. 

    Transformer inductance should not be a block for your design , if you find the switching frequency is high , increase Lm, Vice Versa.

    Lm vs Fsw this is a natural behavior for CRM mode operation.

    Thanks.

  • 0 in reply to Jaden Ning
    Mastermind 6296 points

    Jaden

    Thanks for your review. Apologize for the delay in replying to the issues you raise in this post.

    A full disclosure: given that our input is 380VDC, I really do not understand the implications of various inputs one must provide in the SLUC664C calculations. So excuse if my inputs look arbitrary. They ARE.

    [But lately after reading your replies, I have come to a better grasp of these input ranges. Most important being startup voltage( 365 uA from vs) and Vcs at startup.]

    Our product line up using UCC28780 has several types: DC and AC.

    But the DC case is important.

    Input: 380V dc from a PFC regulator with  3ph input ac 400 Hz.

    We have several different products.

    out of module 1: 5v @ 4 amp

    Module 2 output: 12v @ 4amp

    Module 3: output 15V 8 amp

    Each module can have the same pri with different Secondary turns

    All primaries have - 28turns. (FYI: Core Ae is 108 mm2)

    Secondary turns are

    5v: has 1 turn

    12v and 15v will have 3t

    AUX is adjusted accordingly

    for 5v output AUX turn: 4

    For the other two: 4 as well.

    Your point about high Lm is well taken.

    In fact this large value may be the reason we do not get the first PWML.

    But then we calculated these from SLUC664 C spreadsheet.

    Ldi/dt with dt= 2usec, L=800 uH approx for a certain combination of inputs & fsw

    With this Lm, assuming 350V min startup, you get 875 mA in 2 usec. With  .4 Ohm Rcs, that is>.28 require at start up

    Does this appear consistent with the controller behavior?

    robin

  • 0 in reply to robin gangopadhya18
    TI__Expert 7080 points

    Hi Robin,

    In design calculator , The minimum BULK voltage is used for transformer turn ratios and Lm calculation , and the maximum input voltage is used for evaluate the voltage stress on primary switches and secondary SR . Brown_in voltage is used for RVS1 calculation .

    If your minimum bulk voltage is 350Vds , The turn ratios and Lm you mentioned make sense .

    Thanks.

  • 0 in reply to Jaden Ning
    Mastermind 6296 points

    Jaden:

    A relief!

    Appreciate your time for reviewing my post and a great relief that we concur on the values we get from SLUC664C.

    That said, I did not bring up the ac case we are implementing on yet another proto just as you indicated earlier to reproduce eval kit results using the same values.

    Let us see what we get.

    If all goes well & the proto works as expected, we will get back to the DC case with high DC  input voltage.

    thnx again.

    closing this post.

    robin