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.

UCC29950: UCC29950

Part Number: UCC29950
Other Parts Discussed in Thread: TMS320F28379D, UCD3138

Do not know if this is the correct  controller for LLC.

SLUA733 is the most exciting expose I have read on LLC.

We are developing a  380vDC to  5v/80amp converter due to small size available using LLC.

So our optimization will be very different yet in line with Dr. Leisten treatment of LLC. It would be necessary to plot out our use case

Is it possible to get the Mathcad ( or whatever he used to calculate & plot)  file ?

Pl see attached.

Otherwise, it is almost impossible to obtain the correct range for Lm, k, Cr,  etc. and miss out using the exact treatment of LLC ever done & applied.

robin

  • Hi Robin Gangopadhya18

    Thank you for your interest in UCC29950.

    SLUA733 presents graphs that should cover a very wide range of requirements.

    Normalising the parameters means that many different designs are mapped onto a single point on each graph.

    If you are able to share the details of your requirement then I assist you with this process.

    The information required is the range of input voltage, and the range of output voltage and load current the LLC must be able to supply.

    There is no real problem with sharing the Mathcad 15 file, but it will be difficult to follow since it was never intended for release.

    Thanks

    Joe Leisten

  • Joe

    Thanks for the response.

    Sure, I can describe the inputs, outputs etc.

    #1 Input voltage is fixed with some Tol sich as 5%. And nominal value is 380V, well regulated from 3 phase PFC( controlled by 1/2 Delfino MCU)

    #2 Output is fixed at 5. 5v max load current is 80 amp. User specifies that they may not drive any load. 

    So we plan to keep min static load of, say 3W( limited by available area for heat sinking)

    Our most important task is to calculate Lr, Lpri, turns ratio. From there calculate Cr. Our suitable resonance frequency will be sub 200 kHz.

    All Switching devices are GaN.

    The output is synchronously rectified using UCC7138.

    I have not found any clear cut results for these from the app notes I have reviewed.  But yours appear to get us the most optimal values for these parameters.

    Here is the ringer: we have planned to use the Interlevaed LLC stage that TI has design guidelines for using TMS320F28379D( Delfino). Firmware TI has developed needs to be modified to suit our app but its capabilities are exactly what our customer would want.

    When I posted these in the digital power forum, I could not get these values with rigor. 

    Then luckily, I bumped into your design guide & analysis we rated above any other we have seen yet the end results would apply for the digital control as well.

    We would convert the mathcad  to run  in matlab! ...in a way will be forced to follow. Yet, given how late we are on this, we are surely eager to get pcbs fabricated asap. For now,  getting these values without going through the translation, of course, will highly desirable.

    Any help will be highly appreciated.

    -robin

  • Hi Robin Gangopadhya18,

    Does 5.5V at 80A represent the full load on each output or the combined full load?

    For the LLC design we need to know the absolute maximum load that the LLC stage will be expected to carry. This is normally the Over Current Protection load. It is normally higher than the expected full load.

    As an example let us imagine that we are designing a single LLC stage to deliver an absolute maximum current of 80A at 5.5V.

    Design calculations based upon the application note are attached in PDF form.

    Best Regards

    Joe LeistenDesign_Calcs.pdfDesign_Calcs.pdf

  • Joe

    Wow: this is awesome!

    Nominal output current is 80 amp. OC is programmed in the MCU.

    I would have needed a lot more study of the app note to get such crisp answers....because of the conflicting design guidelines one finds in, say, UCD3138 HB-LLC implementation. 

    We can get a ratio of 34.545 due to planar construction. At least very close to this value.

    With Lm such a low value, it has to be an air-core "transformer".

    So the secondary side will have approx 11.2 nH

    I am wondering if leakage inductance will not make it lose efficiency unless we have coupling =1 or close to it with air-core.

    Do you think this is implementable?

    -robin

  • Joe:

    A clarification: since the MCU code implements Interleaved LLC with tight load sharing, each LLC will carry max 40 amp load.

    fyi

    robin

  • Hi Robin,

    I have attached a revised calculation for your 40A current limit.

    You will notice a significant increase in the required Lr and hence Lm.

    For this calculation I selected an Lm/Lr ratio of 5. If you wish to have a higher Lm value the selecting one of the other curves with a higher Lm/Lr ratio is also possible. It will not affect much the final operating efficiency.

    Cheers

    Joe Leisten

    Design_Calcs_2.pdf

  • Thanks a lot Joe.

    Let me run it & give you feedback on this comes out.

    This post resolves the  issue.

    If more questions come up, we will post new question.

    robin