5
\$\begingroup\$

I need to choose a ceramic capacitor (ESR<10mOhms, 5VDC) for my buck converter application, whose value should be around 300uF. The reason for going for a ceramic capacitor is the low ESR they have. Also my converter would be operating at 100kHz. Off course these capacitors can be placed in parallel (for eg three 100uF caps in parallel) to give equivalent capacitance as well as low total ESR.

Now the confusion is that the capacitors data sheets are very very vague. Vague in a sense that all of them don't give the ESR directly. While searching for it on the Google, I read a post on this website where this issue has been discussed that some caps data sheets give DF instead of direct ESR, and from that D.F we can calculate the ESR knowing the frequency at which the capacitor is operated.

Also some data sheets give the max ripple current and mostly do not. And this is the most confusing for me. How can i select the capacitor for my buck converter application (max 40A current)? I mean is this current ripple rating of the capacitor of my concern or not? As my thought earlier about selecting a capacitor was just keeping the capacitor value, the voltage rating and the ESR value in mind, but I'm not sure about that current ripple thing (which is not mentioned in most of the ceramic capacitors data sheet), if it is to be considered or not? Its been almost 3 hours that I am trying to find something suitable for my application, but unexpectedly I haven't found anything. Either the price is too high (more than 5$ for e.g), or the data sheet isn't providing enough information (some don't give ESR as well as D.F) and now I am confused about current rating (as this parameter also, is not mentioned in most of the data sheets).

Please help me out, you comments would be appreciated.

Thank you!

\$\endgroup\$
9
  • \$\begingroup\$ I am not really sure what you are asking her. If you exceed its SOA you can damage the cap, so if you care about that, don't do it. If the datasheet doesn't provide any SOA information, and if you care about it, then consider looking at the part as already exceeding the SOA. \$\endgroup\$
    – PlasmaHH
    Commented Aug 10, 2015 at 15:33
  • \$\begingroup\$ What i am trying to ask is that if current ripple information provided in the data sheet is of our concern or not (when using it on the buck converter output)? As current ripple definition in capacitor is " AC current flowing in the capacitor", and i am confused that we are using this capacitor in a dc application, so as we know that no DC current passes through a capacitor, should this current ripple be considered as a selection parameter while choosing the capacitor ? \$\endgroup\$
    – yiipmann
    Commented Aug 10, 2015 at 15:50
  • \$\begingroup\$ You are using the cap to smooth out ripple of your DC/DC aren't you? So it is subject to ripple current... \$\endgroup\$
    – PlasmaHH
    Commented Aug 10, 2015 at 15:52
  • \$\begingroup\$ Yes, you must absolutely consider the ripple current in the cap. In a buck converter the input caps are subjected to significant ripple, the output caps will see the inductor PP ripple current and potentially additional ripple due to load transients. \$\endgroup\$
    – John D
    Commented Aug 10, 2015 at 15:56
  • \$\begingroup\$ Yes i am using it to smooth the o/p voltage ripple. So it means that i should consider the current ripple parameter. As i was confused about AC and DC current phenomena (knowing that DC current doesnt pass through a capacitor and) that i am using it on the output of a DC DC converter, so it might be that this current ripple rating isnt of my concern. \$\endgroup\$
    – yiipmann
    Commented Aug 10, 2015 at 15:56

1 Answer 1

1
\$\begingroup\$

The ripple current in the output caps is a significant compared to the output current in a buck convertor running in current mode with a not too large output inductor In fact the peak to peak ripple current could be say 30% of the output current in this orthodox setup SO at 40 amps dc you need caps to handle almost 7 amps peak now the approx. current waveform is triangle so you are up for multiple caps in parallel Use big smd packages and remember that your control loop will change so you may get instability which mill mean changing compensation component values OR leaving the output electro in parallel with your new ceramic caps

\$\endgroup\$
3
  • \$\begingroup\$ Thanks!, will keep that thing in mind while designing the layout. Can you please please refer me some ? i have been looking for them for over a day now and it just feels so exhausting to go through the data sheets over and over again and not finding (all) the concerned information (for e.g esr, or current ripple in most of the cases). I have been looking for them on mouser.de and digikey.de, but no success so far. There are thousands of results even if i try to filter my search, and it seems impossible to go through all of them. \$\endgroup\$
    – yiipmann
    Commented Aug 11, 2015 at 10:29
  • \$\begingroup\$ i need capacitance esr<<15mOhms, as i would be charging my battery cell with this buck converter, and in the battery cell's data sheet it is given that the battery's internal esr <=15mOhms (AC 1 KHz, after 50% charge). As ive read that the output capacitor's ESR should be less than that of the battery cell being charged. \$\endgroup\$
    – yiipmann
    Commented Aug 11, 2015 at 10:50
  • \$\begingroup\$ This answer would really benefit from the use of punctuation: as is, it is barely understandable. \$\endgroup\$ Commented May 2, 2019 at 14:14

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