TPS40201-Q1 Equation of L, C

Toshio,

There is no TI device with that part number (TPS40201-Q1). Can you please confirm which device you have questions about? Thanks.

Hello Alex-san,

TPS40201-Q1 is typo and correct parts number is TPS40210-Q1.

Best Regards.

Q1:

Dmin can be calculated from equation 32. Yes, this equation (34) is correct but the wording above is not complete (i believe it is just short-hand since the correct equation is just underneith)

Q2:

Please use the same constants for all capacitors. I belive that these constants are approximations of some internal structure of the device. I do not know why they weren't described in the datasheet.

Hello Alex-san,

Thank you for your answer, but it is no enough to answer to customer.

Could you or someone answer all of my questions especially followings.

Q1: Equation(34)

- Please advise why 30% is used.

- According to your answer that equation(34) is correct, explanation is not complete.

   Please advise what dose “Iout(max)/(1-Dmin)” express.

Q2: Equation(45) to (48)

-  Please explain meanings of each constant.

Beast Regards.

DCDC team,

Can you please comment on why these constants were used in the TPS40210 datasheet formulas stated above?

Hi Toshio,

The statement of the text saying the peak-to-peak ripple is limited to 30% of the maximum output current is likely a copy and paste error from a buck converter design methodology. Equation 34 calculates it correctly because in a boost the inductor conducts the input current. Iout(max)/(1-Dmin) calculates the maximum average inptu current and the ripple is designed to be 30% of this as a rule of thumb.

The constants in equations 45 to 48 are estimations for using mixed type capacitors as you originally guessed. Peter Miller gives a good explanation of these equations in the following post: http://e2e.ti.com/support/power_management/non-isolated_dcdc/f/196/p/23656/116383.aspx#116383

Hope this helps.

Best Regards,
Anthony

Hello Anthony-san,

Thank you very much for your answer.

It is little difficult to understand Peter’s post so I’m asking our power engineer to explain it.

Could you please answer additional question about equation(49) and (6),(7)

<Equation(49)>

Equation(49) use “IDrive” to calculate Risns.

Is this mean that Gate Charge Current(IDrive) when FET is turn-on should be add to the inductor peak current?

Doesn’t TPS40210 has blank period to avoid the Influence of Gate Charge current?

<Equation(6),(7)>

Equation(6),(7) also calculate Risns but doesn’t use “IDrive”.

Please advise why Equation(6),(7) doesn’t use  “IDrive”.

Best Regards.

Hello Anthony-san,

Please also answer following questions about Equation(47) and (48) that I have already posted on Jul 04 2014.

These equations have constant 4 and 2.

Please explain meaning of each constant.

Please advise if these constant should be changed depend on capacitor type.

Best Regards.

Hello Anthony-san,

Please answer for my previous question about Equation (49),(6),(7),(47),(48).

Best Regards.

Ushikubo-san,

Sorry for the long delay. I am not the original writer of these equations so I do not know the answer right away. I have been travelling so I have not been able to dig into their derivation.

I will help as soon as I can.

Best Regards,
Anthony

Hello Anthony-san,

Thank you for your reply.

Customer is designing their board using TPS40210, so they need your answer.

Best Regards.

Equations 6 and 7 do not include IDrive because they are made to be more generic. The TPS40210 does have leading edge blanking so the IDrive term really isn't needed. In Equation 49 this was likely done just to add some more margin to avoid tripping the hiccup current limit early.

For the input capacitor selection, these constants fall straight out from the derivation assuming 50% duty cycle where the ripple is largest. The term of 2 in the ESR equation assumes half of the input voltage ripple is caused by the ESR of the capacitor.

I should also note the equations for the capacitors are more estimations to give a good starting point and hold very well when the same combination of capacitors are used. Once the capacitor type is mixed the equations become very complicated because the current is shared differently. It really needs to be tuned by testing the steady state and transient voltage ripple on the bench to make sure it meets the requirements of the design. Alternatively PSPICE can be used for a more accurate estimation.

Hello Anthony-san,

Thank you for your answer.

You explained term of 2 in the ESR equation (48).

Please explain terms 4 in Cin equation (47) also.

I case of assuming half of the input voltage ripple is caused by the capacitance, I suppose equation (47) should be

   Cin > (2 x Il ripple)/(Vin ripple x fsw)

Best Regards.

The equation assumes the other half of the voltage ripple comes from the capacitance. The equation starts from the same one used for the output capacitor in a buck converter shown below. The current waveform by the input capacitor in a boost is the same as the output capacitor in a buck. Adding the one half term to the voltage ripple results in simplyfing 2/8 to 1/4.

C > Il ripple / (8 * Vin ripple * fsw)

Hello Anthony-san,

Thank you for your answer.

I could understand the equation.

Best Regards.