High capacitance ceramics

[Stonent]

I got to thinking, now that I've been ordering parts for projects things are a little more clear.

Typical ceramics seem to top off just under 1uF and that's about where electrolytic capactors start. (I know there are exceptions, I'm just going on most common stuff) Is the polarization in part because the capacitance is so high you can't discharge it fast enough in an oscillating circuit or is it still part of the physics?

I was reading an article today that Murata has a 10uF SMD ceramic, but it didn't say if it was polarized or not.

Tantalums are a different story I guess, they have a very wide range but are polarized.

[John Coloccia]

Non-polarized electrolytic are just bigger than polarized.  Large value ceramics are HUGE, so they're generally unavailable beyond about 1uf.  It's just a matter of size.  If you really want to use ceramics, you can just put them in parallel.  I'm doing this in one of my designs now because I can't tolerate the poor performance of the electrolytic.

Tantalum are electrolytic too, and are generally polarized as well.  They're VERY exciting when they go into thermal runaway.

[free_electron]

we regularly use 22uf and 33uf in ceramic. but that is about how high they go. those are 1210 footprints ...

[John Coloccia]

Actually, I was saying and typing ceramic, but thinking "film".  MLCCs are tiny, indeed

[nicknails]

we regularly use 22uf and 33uf in ceramic. but that is about how high they go. those are 1210 footprints ...

Out of curiosity, why 1210 footprint?

We use 10uF and 2.2uF all the time.  Pricing is decent so they can't be that rare.

[Rufus]

Is the polarization in part because the capacitance is so high you can't discharge it fast enough in an oscillating circuit or is it still part of the physics?

Small high value capacitors require very thin dielectric films. It is easy to form thin dielectric films and maintain electrical 'contact' with them by using metal as one plate of the capacitor and a conductive liquid as the other. The liquid is chosen to react electrolytically with the metal to form the dielectric film on its surface. The reaction is somewhat reversible so such capacitors can not stand reverse polarity. The films are self-healing. A hole in the dielectric allowing current to flow grows more dielectric to fill it.

Tantalums have the problem that if enough current is available holes in the dielectric can produce catastrophic damage rather than repair.

[Monkeh]

I've got a load of 4u7 ceramics in 1206. They're not terribly hard to find, I don't know where you've been looking..

[walshms]

More and higher value ceramics will continue to appear, I think.  The dielectrics are getting better, and they are getting better at making them.

[Monkeh]

http://uk.farnell.com/avx/st203c227maj10/capacitor-ceramic-220uf-25v-x7r/dp/1829121

When you really, really want a ceramic capacitor.

[Whales]

http://uk.farnell.com/avx/st203c227maj10/capacitor-ceramic-220uf-25v-x7r/dp/1829121

When you really, really want a ceramic capacitor.

I saw these on Farnell once before and thought about how awesome it would be to be paid to make these.  How many actually sell?

[ovnr]

Well, a recent project of mine used a 47u 6.3v 0805 (X5R) ceramic. Just beware of the usual drop in capacitance - this one is down 80% at rated voltage. Since I'm just using it at 2-3 volts, it's at around 20uF.

(And it didn't work out as I had hoped, but that's another story of leaky diodes and shit.)

[amyk]

http://uk.farnell.com/avx/st203c227maj10/capacitor-ceramic-220uf-25v-x7r/dp/1829121

When you really, really want a ceramic capacitor.

And it costs more than just buying 10 x 22uf ones...

Ceramic caps have a strong voltage-dependence on their actual capacitance:
https://www.eevblog.com/forum/projects/t22291/

[DutchGert]

http://uk.farnell.com/avx/st203c227maj10/capacitor-ceramic-220uf-25v-x7r/dp/1829121

When you really, really want a ceramic capacitor.

And it costs more than just buying 10 x 22uf ones...

Ceramic caps have a strong voltage-dependence on their actual capacitance:
https://www.eevblog.com/forum/projects/t22291/

But 10x22uF could screw up your total ESR.....