• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 2 answers
  • Subscribers 62 subscribers
  • Views 280 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.

Original question:

Changing quantity of output capacitors in Webench

TPS544C25: WEBENCH tool

Kiran Kerai
Prodigy 110 points
Part Number: TPS544C25
Other Parts Discussed in Thread: LM10011

I have used WEBENCH to create a design and edited component values (capacitors, resistors, inductor) to align with a prototype we have built and tested.

Component changes to resistors and inductors have been done by selecting models from the library however some of the capacitors were not available therefore I have opted to create a custom part and enters capacitor value and ESR.

Will the performance of my design reflect the manual changes to schematic?

I am also looking to run load step analyses using the simulation tool. Will manual schematic changes be accounted for with this simulation?

  • 0
    TI__Genius 11315 points

    Hey Kiran,

    Yes, as long as you have filled in the parameters of the part correctly, the WEBENCH designer will account for it during simulation.

    Thanks,
    Caleb

  • 0 in reply to Caleb Perez
    Prodigy 110 points

    Thanks Caleb

    I am struggling to validate the output ripple voltage for my design. Could you provide insight on how this is derived?

  • 0 in reply to Kiran Kerai
    TI__Genius 11315 points

    Hey Kiran,

    Vout_Ripple =SQRT((ESR_Ohm*Iripple_max)^2+(Iripple_max/(8*Cout_F*0.000001*fsw_Hz))^2)*10^-3

    Where ESR_Ohm is the effective ESR of your output caps in Ohms, and

    Iripple_Max =1000*((Vin_max-Vout)*Vout)/(Lout*(1-Tol_L)*Vin_max*fsw(Hz))

    Where Tol_L is the tolerance of inductance for the selected inductor.

    May I suggest that you use the PSPICE average model or the excel calculator sheet available on the product page to calculate the Vout ripple instead?

    Thanks,
    Caleb

  • 0 in reply to Caleb Perez
    Prodigy 110 points

    I have a matching formula for Iripple max so that makes sense.

    I am having trouble with Vout_Ripple. If I put values into the excel calculator sheet:

    Vout_Ripple =SQRT((0.000285*5.02364)^2+(5.02364/(8*0.00056*0.000001*500000))^2)*10^-3 = 2.24

    I dont think the 0.000001 term should be in the formula as Cout_F is in farads

    I believe the formaula is taken from equation 25 of the following application report from TI:

    Output Ripple Voltage for Buck Switching Regulator (Rev. A) (ti.com)

    Here, the 0.000001 term is not featured in the equation. Can assume the equation in the excel calculator sheet is incorrect?
    It should be:
    =SQRT((0.000285*5.02364)^2+(5.02364/(8*0.00056*500000))^2)  for ripple in Volts
    or:
    =SQRT((0.000285*5.02364)^2+(5.02364/(8*0.00056*500000))^2)*10^3  for ripple in milli Volts
    Please confirm.

    Also, if we intend to use adaptive voltage scaling with LM10011 DAC, much like the EVM board here:

    K2E_EVM_SCH_16_00175_03.opj (einfochips.com)
    See sheet 31.

    Do we need to change the compensation resistors from the WEBENCH model or can the bottom resistor be removed and replaced with connection to the DAC output.
    Datasheet states: "When VOUT_SCALE_LOOP = 1 (default), no bottom resistor RBIAS is needed. The reference voltage is equal to the output voltage, which allows tighter system DC accuracy by removing the resistor divider tolerance."
    Will stability change from the WEBENCH report if we add the DAC?
    Do I need to simulate with the LM10011 in place?

  • 0 in reply to Kiran Kerai
    Prodigy 110 points

    TPS544B25_TPS544C25_Calculator_Checklist_1p0.xlsxTPS544C25 WEBENCH design report.pdf

    Using the calculator, I have entered the value of capacitors in my feedback network taken from the Webench report (both attached)

    Here,
    R1=Rfbt=10k
    R2=Rcomp2=1k
    R3=Rcomp=49.9k
    Rbias=Rfbb=10M
    C1= Ccomp3=330p
    C2=CComp=1n
    C3=CComp2=10p

    The cross over frequency and phase margin is different in the Excel calculator compared with WEBENCH. Which is correct? Why is there a difference?

  • 0 in reply to Kiran Kerai
    TI__Genius 11315 points

    Hey Kiran,

    The equation I gave you is for ripple in mV. You should simulate the voltage scaling using a program like PSPICE to see if the device stays stable with the compensation network you have. I wouldn't recommend using a DAC to control the compensation loop.

    I believe the difference comes from device parasitics. WeBench generally imports these parasitics automatically from it's internal database of components (in this case I believe you entered them manually) while the excel sheet does not consider these at all for the feedback network.

    Thanks,
    Caleb

  • 0 in reply to Caleb Perez
    Prodigy 110 points

    Hi

    The equation you gave me for Vo_ripple contains *0.000001 term which does not align with the link. Also the 10^-3 should be 10^3 to convert from V to mV.
    Vout_Ripple =SQRT((0.000285*5.02364)^2+(5.02364/(8*0.00056*0.000001*500000))^2)*10^-3 = 2.24 mV

    I think the formula should be 
    =SQRT((0.000285*5.02364)^2+(5.02364/(8*0.00056*500000))^2)*10^3 = 2.66 mV
    Please confirm.

    The DAC is a requirement for our design as one of our processors needs to dynamically control the output voltage. This is the purpose of LM10011.
    I will try and install PSPICE for TI.

  • +1 in reply to Kiran Kerai
    TI__Genius 11315 points

    Hey Kiran,

    I've dug into the excel sheet a bit and it looks like you're right. The 0.000001 term should not be there, as all of the values in that group are already in standard format. The 10^-3 should also be 10^3 to convert to mV. 

    I will document this and have it released on our next refresh. Please create a separate thread for any issues with the LM10011.

    Thank you,
    Caleb