Is it okay to bend the output electrolytic filter capacitor right next to my SMPS inductor?
\$\begingroup\$
\$\endgroup\$
8
-
\$\begingroup\$ I don't see why it should be a problem. Ideally the cap shouldn't get too warm, but that's optional. Make sure to buy a fully encapsulated inductor also. \$\endgroup\$– LundinNov 20 at 15:54
-
\$\begingroup\$ What do you mean with "bend" though, is this an "axial" or "radial" cap? Only one pin is visible on the pic, I'm assuming axial. \$\endgroup\$– LundinNov 20 at 15:56
-
1\$\begingroup\$ Could you please edit your question to include a schematic. It looks like your board layout already includes proper chip caps at the output of your SMPS -- do you? Why do you think you need an electrolytic cap there? What do you think that it will contribute? \$\endgroup\$– TimWescottNov 20 at 16:19
-
2\$\begingroup\$ Putting an exorbitantly large capacitor on the output of a switching regulator is generally not a good idea. At the input, maybe, depending on what the source can handle, and say if hold-up time is required. "Instantly satisfy demands" is just transient response, which should be handled by nominal output capacitance, and the response (and compensation) of the regulator. | Adding part numbers would be welcome. I'm a bit perplexed by the combination of a ~3x3mm SON combined with what might be a 5-10A schottky and who knows what inductor. \$\endgroup\$– Tim WilliamsNov 20 at 19:00
-
1\$\begingroup\$ I was assuming this sits on the input to the regulator to counter voltage surges (pretty common practice), but on the inside... hmm. How does the regulator deal with the inrush current? \$\endgroup\$– LundinNov 21 at 7:37
|
Show 3 more comments
1 Answer
\$\begingroup\$
\$\endgroup\$
2
It looks like you have a high voltage buck converter here. This layout looks good by eye to me, as long as you have a solid ground connection.
If you're soldering these by hand and the capacitor isn't physically compromised by this, it should work just fine electrically. You have ceramic bypass caps locally, and the low ESR/ESL of those should meet loop stability and inductive ringing under control, as long as they are chosen correctly. The electrolytic capacitor is basically just to store more charge as a reservoir, so its placement on the board is not as critical.
Also note that for a buck, the input capacitor is the most critical in terms of board layout, since it's the input current that is discontinuous (trapezoidal wave with hard edges), while I(out) is the same as inductor current (triangle wave).
-
1\$\begingroup\$ There's no way to tell that the layout is good without viewing the copper planes, including the ground underneath the part. \$\endgroup\$– LundinNov 21 at 7:34
-
\$\begingroup\$ @Lundin that's why I said "if you have a solid ground connection", meaning you're supplying adequate ground plane to whichever pin(s) are your ground connection. I meant that component placement looks adequately close for anything at a reasonable fsw. \$\endgroup\$ Nov 21 at 8:25