SLVA882

Hi Nestor,

You seem to have everything you need to calculate the output ripple current. Just use n, m, and D to get the normalized value and then multiply by your max output current for the worst case current. Using your numbers I'm getting the normalized ripple current to be 0.927 which makes sense because D is so low (Figure 6) there isn't a lot of cancellation between phases. With a max current of 45A this gives an current through the caps of 41.7Arms. 

Please note, if you're going through these calculations to determine the output capacitance you'll need picking a number to handle just the RMS current alone isn't enough. You'll likely underestimate how much is actually needed. Typically transient response dominates your performance and so going through the steps in Section 5.5 will give you a much more realistic picture of how much COUT you'll actually need.

Cheers,

Carmen

Hi Carmen,

Thanks for your reminder and reply.

According to equation 2, we can estimate the current through the output capacitors is 41.7Arms.

But I was confused, why the calculation result is some big.
As I know, the ripple current of the output capacitor should be similar to the inductor current.

https://www.richtek.com/Design%20Support/Technical%20Document/~/media/AN%20PDF/AN041_EN.ashx(Figure 3)

Or I misunderstand the meaning of this formula.

I try to use the other formula to calculate the output ripple current of the capacitor and it looks like only 7.29Arms. (mic2155@page21)

http://ww1.microchip.com/downloads/en/DeviceDoc/mic2155.pdf

Btw, I also use webench to do the rough simulation.

The “Cout IRMS” isn’t big even the inductor or switch frequency is a little different.

Could you explain more about the calculate result and how to use it to select the output capacitor? (permissible ripple current?!)

Thank you.

Hi Carmen, 

Thank very much for your clarification and sharing more information with me. ^^