11
\$\begingroup\$

Now I know when it comes to electrolytic capacitors it's usually good practice to use a voltage rating 1.5x - 2.5x the maximum rating you ever expect the capacitor to be exposed to, as the lifetime can be significantly reduced by running it near its limits.

But does the same protocol apply to ceramic capacitors? For example, assume a 25V ceramic capacitor - will running it at 24V reduce its lifespan significantly?

I'm currently designing a PSU which needs to be reliable, so at the moment I'm using 50V caps for a maximum 20V input (35V caps aren't available), but I'd like to switch to 25V caps if possible.

\$\endgroup\$
9
\$\begingroup\$

You do not have the same lifespan issues with ceramic caps. You will lose capacitance as the voltage increases. The amount you lose is dependant on the dielectric.

When choosing the electrolytics you also need to look at the ESR, temperature rating and the maximum operating temperature. Since the lifetime is predicted using the Arrhenius activation energy law running using a 105°C rated cap will give you a significant increase over running a 85°C cap (assuming all other conditions are the same). There are some 5000 hour rated electrolytics.

\$\endgroup\$
  • \$\begingroup\$ OOC, are you the same "luciani" who develops the luciani library for gEDA? If so, thanks! I use those footprints lots. \$\endgroup\$ – Thomas O Oct 16 '10 at 18:49
  • 1
    \$\begingroup\$ That's me. Kemet (ceramic cap mfg) has some excellent app notes that explains the tolerancing of their caps. For power supplies I like the Nichicon PM series of electrolytics. I would check their website because they may have a new low ESR series. \$\endgroup\$ – jluciani Oct 16 '10 at 21:34
  • \$\begingroup\$ I missed the buck regulator statement. If you require the low ESR of ceramic and small size I would go with an X5R or X7R dielectric. The tolerancing of the Y5V is quite bad over the voltage and temperature range. \$\endgroup\$ – jluciani Oct 16 '10 at 21:37
  • \$\begingroup\$ @Thomas O - Really, unless you need a LOT of capacitance in a tiny package, you should never use anything less than X(7/5)R anyways. Z5U Etc... are really not worth it unless you need to cut every cent. \$\endgroup\$ – Connor Wolf Jun 13 '11 at 21:18
7
\$\begingroup\$

ceramic caps are really durable, MLCC's are often tested at rated to 200% rated voltage for ~1000 hours before failure.

That being said, I often find that higher voltage MLCC's can actually be cheaper than the lower voltage for a given capacitance due to higher industry volume for the higher voltage ratings. For instance finding 16v 0.1uf MLCC is never a problem, but something like 100pF may be cheaper at 50V than it is at 16V

\$\endgroup\$
3
\$\begingroup\$

You should actually not be comparing ceramics to electrolytics in terms of lifespan at all, they are actually intended for totally different purposes. Most beginners (I did untill recently) think they are just higher values, but they are as different as inductors are to resistors. Mainly because electrolytic values are more approximate, and their value much more temperature dependant than ceramic, polycarbonates or tantalum. Note: Tantalum are very voltage sensitive.

You answer is that I would also use a 50V cap for a 20V supply, in fact if you glance at the charge curve for electrolytics you will notice that they work better (more farads) at about 1/3 of their rating anyway. As for being for different purpose, you will get heating in an electrolytic if you allow it to completely charge and discharge on a cycle, for instance using a 220uF to service a 10 amp load on a rectifier will tax the cap and cause heating in the leads, and as the other answers alread indicate, heating equals less farads.

I'm just an amateur, so no clever math, sorry.

\$\endgroup\$
  • \$\begingroup\$ I have to use ceramics, as it is required by the buck regulator I am using. But I want to know if using a 25V cap is okay for a 20V supply. \$\endgroup\$ – Thomas O Oct 16 '10 at 19:08
  • 2
    \$\begingroup\$ Ah, ok, buck regulator is different story, definite ceramic needed, I believe you want a 50V component, LOW K if possible. Parallel them to get up to the needed value if necessary. \$\endgroup\$ – Conrad B Oct 16 '10 at 19:18
  • \$\begingroup\$ Size is critical, which is why I'm looking to reduce voltage. \$\endgroup\$ – Thomas O Oct 16 '10 at 19:21
  • \$\begingroup\$ the frequency response of each type of capacitor is different also. \$\endgroup\$ – Kellenjb Oct 18 '10 at 4:02

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.