Capacitance versus voltage for X7S caps?

Tried the major mfgs and the typical datasheet looks like this:

formatting link

Quote "Capacitance for X7S varies under the influence of electrical operating conditions such as voltage and frequency."

Then under diagrams ... nada, zip, zilch. Great.

One paper listed X7S with the same voltage coefficient as X7R but that doesn't sound right. Anyone have a link to some hard data, with a graph in there and preferably no marketing hype?

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg
Loading thread data ...

Joerg,

Not too long ago I was also looking for C(V) dependencies for different capacitor types; didn't find much useful information either. One of the reasons for that is C(V) behavior of particular cap is strongly affected by electrostatic mechanical (!) effects; thus, at low frequencies, it is too much of dependency from everything.

Vladimir Vassilevsky DSP and Mixed Signal Design Consultant

formatting link

Reply to
Vladimir Vassilevsky

Joerg a écrit :

I looked at that recently but for X7R/X5R. Data sheets often have nothing, but the good manufacturers offer some 'simulation' program with lots of curve fitting,...

IIRC TDK has some web base one too. I think Kemet, AVX, Murata, Taiyo have what you're after.

--
Thanks,
Fred.
Reply to
Fred Bartoli

I seem to recall reading that C0G and NP0 dielectrics have the lowest cap-vs-temp dependance and that pretty much everyone "understands" that X7R dielectrics have large coefficients and are pretty much unsuitable where it matters. They pack a lot of capacitance into a small space and that makes them great for decoupling jobs and not so much else.

But I don't know remember reading anything about X7S, specifically. If they are the same as X7R, they are crap if what else I read was right.

Jon

Reply to
Jon Kirwan

I've been through those. My impression was that a lot of docs I used to see there have been "cleaned out". I need something in the form of a document, not a simulator. Also, other than SPICE and beam field simulators I don't trust them. For example, National has flagged all my first switchmode converter ideas as "can't be done". And all went into mass production without a hitch ...

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

The lousy ones are Y5V and Z5U. X7R is actually pretty good.

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

For low values of "pretty good". I suppose losing 20% of capacitance due to voltage and/or 5% due to temperature and/or a few more percent due to aging is better than -80% or whatever..

NP0 are now available (though a bit pricey and bulky) even in fairly large values like 0.1uF 1206.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

X7R is bad enough that it distorts like hell in an audio amplifier (used as the Miller cap) and I know I certainly can't even come close to using them in integrators, from actual (hilarious, for a moment) experience. Decoupling is what they are good for.

Jon

Reply to
Jon Kirwan

Unsurprisingly, the integration curve (step input) looks just like a high permeability, ungapped ferrite inductor's.

You're just tracing out the D-E curve instead of the B-H curve.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams

formatting link
formatting link
formatting link
formatting link

Reply to
Ken

Digikey has 244 items listed in stock with X5S, X6S, X7S and Y5S temperature coefficients.

Reply to
Muzaffer Kal

formatting link

formatting link

formatting link

formatting link

Well, I know what Z5U does. But it doesn't tell me what X7S does :-)

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

Because distributors sell them and I'd like to use this one:

formatting link

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

Exactly. We only go to about 50% of rated voltage under normal operating conditions and I'd be perfectly happy if the capacitance loss there would be 20%.

But not 4.7uF/100V :-)

Oh, and it can't be taller than 0.100".

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

That's what I am needing them for :-)

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

If you can't find data, send me a sample and I can run it on our HP4194 impedance analyzer.

--
Mark

On Thu, 25 Mar 2010 08:02:15 -0700, Joerg 
 Click to see the full signature
Reply to
qrk

Thanks for the offer, Mark. I have a HP4191 here so I could also do that. But this project is too fast-track. Needs to go into layout ideally by tonight or tomorrow morning, no time to get samples here.

It blows my mind that maufacturers do not furnish such data. This time that could cost the sales since w may go hi-rel electrolytic, to be on the safe side.

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

I think there's an important point being missed here: capacitance is a term in a linear equation, and the behavior of these capacitor materials is NONLINEAR. The 'capacitance' change is undefined unless you specify a full test setup.

The nonlinearity of these materials includes hysteresis and memory effects and acoustic emission and sensitivity. They don't have 'capacitance versus voltage', they have nonlinearity. The manufacturer has good reasons to only specify a few limits.

Reply to
whit3rd

Well, in fact that's even worse because some relaxation time factors in too.

--
Thanks,
Fred.
Reply to
Fred Bartoli

Those factors are all minor when using capacitors for bypassing. For all traditional ceramics such as X7R, Z5U or Y5V (the latter two having much higher non-linearity) the manufacturers give capacitance versus voltage spec. For X7S they don't. Somehow that smells like a "Marketing didn't like the graphs" situation.

All I need to know is how much C will be left at 50% voltage, roughly. I don't care whether it's non-linear.

--
Regards, Joerg

http://www.analogconsultants.com/
 Click to see the full signature
Reply to
Joerg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.