I have a existing buck regulator design. Converter part number: TPS54340Q1.

Input voltage = 5 V. Output voltage = 3.3 V

RMS ripple current is calculated as = 108 mA. (Don't worry too much about value. Conider the value for mathematical analysis)

I have two decoupling capacitors at the output:

a) Cout1 = 1 uF, ESR1 - 110 mOhm

b) Cout2 = 4.7 uF, ESR2 - 120 mOhm

Switching frequency = 500 kHz.

Question 1

I just need to calculate how ripple current will be split between Cout1 and Cout 2. I mean how will be ratio.

Can we do the current splitting just as we do with two parallel resistors for ESR1 and ESR2?

Question 2

Assume we got the value for each capacitor ripple current. How can make sure that the existing decoupling capacitor is enough to withstand the ripple current passing through it. ( I checked for rated current value in capacitors datasheet, there is no such parameters with ceramic capacitors but yes, there is rated current for tantalum capacitors. My output capacitor is ceramic type)

Done some research on internet from my end,but there was not much I could not find.


Cout1 Capacitor link

Cout2 Capacitor link

  • 1
    \$\begingroup\$ Can compute apparent Impedance of RC at 500kHz and harmonics? Then divide V(f) / Zc(f) for each cap and load R. \$\endgroup\$ – Sunnyskyguy EE75 May 14 '18 at 12:29
  • 1
    \$\begingroup\$ What is your ripple spec and what RLCRCR filter is needed to achieve it. Be sure you measure ripple into an impedance matched 50 ohm AC coupled scope. No I mean square root of square R^2+X(f)^2 \$\endgroup\$ – Sunnyskyguy EE75 May 14 '18 at 12:34
  • 1
    \$\begingroup\$ These letters suggest possible equivalent circuit components for your filter analysis in series and shunt. Beware of parallel resonance.model it accurately and simulate it in f domain eg. Falstad and measure accurately \$\endgroup\$ – Sunnyskyguy EE75 May 14 '18 at 12:45
  • 1
    \$\begingroup\$ Also and cap worth using must have a low ESR specified \$\endgroup\$ – Sunnyskyguy EE75 May 14 '18 at 12:49
  • 2
    \$\begingroup\$ @vt673 those values will be probably about 50 times too high. It might be prudent to find a capacitor that has a data sheet that indicated the graph of its resonant series frequency and look at the bottom of the curve to understand what the practical ESR will be. The Murata data sheet looks like it was written by a "committee" who kind of missed the point of what engineers need to study!! \$\endgroup\$ – Andy aka May 14 '18 at 12:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.