4
\$\begingroup\$

I am stuck at calculating output ripple voltage with non-ideal components for buck converter. As you can see from the picture, the output ripple voltage caused by each individual component is calculated.

However, how about the total output ripple voltage? Is it equal to sum of ripples caused by individual components?

Another point is that with given output ripple voltage specs, how do you choose inductor and capacitor along with their equivalent series resistance values to meet the output ripple specs?

(output ripple voltage specs is 1% of output voltage)

enter image description here

\$\endgroup\$
  • \$\begingroup\$ Yes. You choose as low ESR as you can afford. If you can move from DCM to CCM or from CCM to deeper CCM, you can increase the inductance too. \$\endgroup\$ – winny Dec 13 '16 at 17:23
  • \$\begingroup\$ @winny, careful about just choosing the lowest possible ESR. ESR probably contributes to the phase response of the control loop; changing it may require adjusting the loop filter to compensate. \$\endgroup\$ – The Photon Dec 13 '16 at 20:15
  • \$\begingroup\$ Thanks. That seems to be a trial and error method. I would like to know if there is a better method to meet the specs without having to use trial and error a lot. \$\endgroup\$ – anhnha Dec 13 '16 at 20:19
  • \$\begingroup\$ Sure! You make one giant excel-matrix and pick the best one in regards to your constraints. @ThePhoton Yes, but moving from DCM to CCM would have even bigger impact so I simplified a bit. Lower ESR will be cheaper than more inductance too. \$\endgroup\$ – winny Dec 13 '16 at 21:11
  • \$\begingroup\$ Well, I have just read the application note below from TI and as in the note, output voltage ripple is not sum of individual component at all. ti.com/lit/an/slva630a/slva630a.pdf \$\endgroup\$ – anhnha Dec 13 '16 at 21:16
1
\$\begingroup\$

However, how about the total output ripple voltage? Is it equal to sum of ripples caused by individual components?

No, of course not. The output voltage is across the capacitor, so only the capacitor's ripple voltage appears at the output.

The "ripple voltage" across the inductor's series resistance is entirely masked by the much larger voltage being applied to the inductor by the switch. It has no relevance other than to waste power that would otherwise increase the overall efficiency of the converter.

Another point is that with given output ripple voltage specs, how do you choose inductor and capacitor along with their equivalent series resistance values to meet the output ripple specs?

Choosing parts for a switchmode converter of any type is a multidimensional optimization process involving several interdependent parameters (input to output voltage ratio, switching frequency, transient response, etc.), of which output ripple is just one. There is no simple answer, but it's generally a good idea to pick capacitors with the lowest ESR available.

Sometimes it makes sense to use multiple capacitors in parallel in order to get a better value for the overall ESR. This may also be driven by the need to not exceed the capacitors' ripple current specifications, which is closely related.

\$\endgroup\$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.