• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 538 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.

LM3481: Queries in application note of LM3481

Mustafa Mahammad
Prodigy 30 points
Part Number: LM3481

We are referring LM3481 application note for designing Non-Isolated Fly-back converter. For that we are going through the design formulas given in application note.

In the application note of LM3481, there is one equation to calculate Primary Inductance Lm in page 5, and a graph is given to choose the primary inductance value with respect to Vin in page 10. We tried to calculate primary inductance  with equation in page 5 which is not matching with the graph. Example specifications are given in page 9 , with these specifications the calculated primary inductance is not matching with graph or selected Lm=12uH

Please let us know how the graph has been achieved for the range of input voltage specifications given in section 4 (page 9) of application note.

We are trying to design Fly-back converter with below specifications

Vin(min)=3.16V

Vin(max)=25.66V

Vout=12V

Iout=1.556A

Fsw=100KHz

Turns ratio=1

TI_Application note_LM3481 Boost Controller.pdfTI_Application note_LM3481 Boost Controller.pdf

  • 0
    TI__Guru* 78070 points
    Hi Mustafa,

    Thank you for considering the LM3481. The app note offers a design guideline, and the equations and selection curve for the inductor is the minimum required to stay CCM. Actually you should choose a part not smaller than that, but also not too big to avoid higher cost.

    I don't think your design with transformer turns ratio 1:1 is correct for this 3.16V to 12V conversion. It will ask a duty cycle of about 80%, hitting the limit of the LM3481. You should follow choose a turns ratio like 1:2 or something similar.

    Please also refer to www.ti.com/.../flybuck-flyback-design-calc for your flyback design.

    Best Regards,
    Youhao Xi, Applications Engineering
  • 0 in reply to Youhao Xi
    Prodigy 30 points
    Hi Youhao Xi,
    Thanks for your quick response.
    I have a question in application note regarding the selection of Primary Inductance Lm.
    I considered the specifications given in section 4 (page 9) of application note and calculated Primary Inductance (Lm) value using equation in Page 5 and the results are Lm=17uH at Vin=5V, and Lm=104uH at Vin=32V. But the selected Lm is 12uH which is not in the limits with calculated results. This 12uH is selected from graph given in page 10.

    So, please let me know how the graph has been derived and why the selected Lm is not matching with the calculated results.

    Regards
    Mustafa Mahammad
  • 0 in reply to Mustafa Mahammad
    TI__Guru* 78070 points

    Hi Mustafa,
    You are very welcome. I see the confusion. Note that there is a parameter KL, which is the factor for inductor ripple current limitation. In practical design, designers generally choose KL between 20% to 60%. Since a CCM flyback transformer worst case is at min Vin, under which the primary winding peak current is the highest, the transformer is thus selected by Vin_min. In the example that you referred to, it seems KL was chosen to be 25% by the author. Note that smaller the KL, the larger inductance is, which usually means (but not necessarily always the case) a larger core. So trade-off should be evaluation between larger core and larger ripple current.

    Thanks,
    Youhao

  • 0 in reply to Youhao Xi
    Prodigy 30 points

    Hi Youhao,

    Thanks for your quick response.

    Even though if we consider KL=0.25,  the inductance value is 13.6uH at Vin min=5V.

    Here, my point of concern is, the graph is given for complete input voltage range and selected Lm is 12uH.

    But when we calculate primary inductance using the equation, the results are not matching with graph.

    Which equation was used to plot the graph or which equation was used to select 12uH at Vin max=32V?

    Attaching the caluculations we have done for your quick reference.

    Also in application note clearly mentioning to consider complete voltage range to design inductor. (Page 5: Magnetizing inductor should be designed at highest input voltage to ensure the converter could run at CCM during all the input voltage range)

    Regards

    Mustafa Mahammad

    Inductance calculation.pdf

  • 0 in reply to Mustafa Mahammad
    TI__Guru* 78070 points
    Hi Mustafa,

    You calculation is correct. The selection curve is not correct. The writer seemingly pasted a wrong figure, and we will update the datasheet soon. Here I would like to thank you for catching an error that may exist for many years without being noticed. And, we apologize for any inconvenience this mistake may cause you.

    Regarding the inductor selection, in my opinion KL=0.2 is not always favorable because this means the CCM and DCM boundary is at about 10% of the full load. As I mentioned, most designer would select KL between 0.2 and 0.6, or even 0.8, and larger the number leads to smaller inductor, which usually cost less. The design is always a trade-off between the ripple current and cost (or size). For this example, 12uH seems being selected by choosing KL=0.3 and Vin=5V.

    For practical design, I would recommend not to use an unnecessarily large magnetics. I would select the inductor by Vin_min, with KL between 0.2 and 0.6. This is because the peak current occurs at Vin_min, and the transformer should not saturate at that peak current. Obviously in this example the circuit will enter DCM at light load, or under higher Vin . DCM is not the end of world and there are many Flybacks are designed in DCM. I believe the test results shown in the app note were obtained on the EVM and you can tell that the circuit operated well.

    Thank you again.

    Best Regards,
    Youhao Xi