• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Subscribers 63 subscribers
  • Views 684 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

RE: LM25574 Inverting Buck-Boost--Webench problems?

Martin Risso
Intellectual 470 points
Other Parts Discussed in Thread: LM46001

Based on your application notes, I would like to use an inductor value large enough to keep in CCM mode at light load (20 mA minimum, 100 mA maximum). Is it possible to enter custom value inductors for the requirement 

Lmin (Vin x D) / Fsw x 2 x Iout min

  • 0
    Intellectual 470 points

    correction Lmin = (Vin x D) / (Fsw x 2 x Iout min)

  • 0 in reply to Martin Risso
    TI__Genius 15810 points

    Hello,

    Please send us your design requirements. As such for any current mode control designs, it is not a good idea to make the inductor too large at light loads. The issue is that the slope of the sensed current is too shallow and can cause problems in regulation if there is too much noise in the circuit. 

    Regards,
    Akshay

  • 0 in reply to Akshay Mehta
    Intellectual 470 points

    Vin min 14V

    Vin max 16V

    Vout -15V

    Iout min 15 mA

    Iout max 100 mA

    Prefer to stay in CCM down to Iout min

  • 0 in reply to Martin Risso
    TI__Genius 15810 points

    Hello,

    If I may ask, why do you insist on CCM in light load? Along with what I mentioned before, your inductor would potentially become very large. The RHP zero will be at lower frequencies, causing the design to become more sluggish and slow because the bandwidth would need to be lower than that. If you are okay with this then we can give you a BOM for you to prototype.

    As far as Webench goes, I will split your thread and send it to the appropriate forum where the right people can answer the question.

    Regards,
    Akshay

  • 0 in reply to Akshay Mehta
    TI__Genius 15810 points

    Hello,

    Thank you for the inputs. I am attaching a schematic which I have marked up with yellow sticky notes in PDF. Please go through that. The inductor is sized for a 100% ripple at 15mA and therefore just barely stays in CCM at light load and at full load it should be definitely in CCM. Please prototype this and we can take it further.

    Regards,
    Akshay

    webench_Schematic_neg15Vo 100mA.pdf
  • 0 in reply to Akshay Mehta
    TI__Genius 15810 points

    For some reason I am not able to split your thread and instead I am ending up deleting stuff. I'm going to create a new post in the webench design forum and then you could reply there if you'd like to.

    Regards,
    Akshay

  • 0 in reply to Akshay Mehta
    TI__Genius 15810 points

    Hello,

    I apologize for multiple replies, but I just realized that you could use a much newer device for your design which reduces complexity by a huge margin. Please go through the app note: http://www.ti.com/lit/an/snva722a/snva722a.pdf

    This family is also of the SIMPLE SWITCHER brand with internal compensation and wide Vin. Another advantage they have is that they are synchronous and therefore will be more efficient. I hope this helps.

    Regards,
    Akshay

  • 0 in reply to Akshay Mehta
    Intellectual 470 points

    Akshay,

    Thank you for the BOM.

    I'm using an LDO post regulator to filter out the ripple. 1 MHz is outside the post regulator (TPS7A3001) ripple rejection range.

    I have seen inverting switchers that look like a modified SEPIC, two coils and a coupling capacitor.  Are there advantages to using this kind of configuration? Does T.I. have any Web Bench enabled parts like this?

    Marty

  • 0 in reply to Martin Risso
    TI__Genius 15810 points

    The reason I chose high frequency is that since you wanted to run in CCM near light loads, the inductor would be a bit bigger. At lower frequencies the inductor is going to becomes a little more big. For 750kHz, the following changes could be made:
    L = 330uH rated for 200mA
    Cout = 68uF
    Rcomp = 82.5kohms
    Ccomp = 6.8nF
    Ccomp2 = 68pF
    Cramp = 220pF

    You are probably talking about Cuk. It is much more difficult to design than the regular buck boost. For starters it would take two inductors or two coils on the same core. I do not think Cuk as a topology is enabled on Webench. The easiest approach for your requirement is using LM46001.

    Regards,
    Akshay