Buying capacitors, difference between same value&same sice but different voltage.

[David_]

Hello.

I have done a PCB layout and am putting a order for the components I need, I probably should have picked a SMD size and stuck with that but I have used both 1206 and 0805 since 0805 does save space but 1206 allows to be used as a jumper so I can route a or two signals in between its pads.

I am buying ceramic capacitors and I have two different voltage levels that they are to be used, so I need 0805 0,1µF on a ~40V rail and 0805 0,1µF on a 5V rail.

First of all the prices vary with a almost ridiculous range, for one cap I can choose to pay 0,86kr or 7,38kr and it vary more with other components sometime. 1 USD ~7kr
While the price differ the material in the cap does not both is X7R but I have to think that the more expensive should have less parasitics and such, no?

Back to the 0805 0,1µF caps, I was going to order them all as 50V caps but I'm just taking a easier path.
Am I wrong doing this or not? I mean is there any reason to buy two kinds, I am only buying small quontetyes anyway so bulk price does not come in to it.

There datasheets offer no info at all.

 

[biff44]

Re: Buying capacitors, difference between same value&same sice but different voltage.

well, it might help if you stated what you were using them for.

Some large value capacitors cost a lo if you want a very small package.

In general, higher breakdown voltage capaictors have higher series inductance, so they would not be desired if you were making RF circuits.

In some rare cases with microwave circuits, too big of a capacitor can actually resonate and look like and inductor or very lossy resistor, instead of a capacitance. So for microwave projects, you choose the lowest breakdown voltage you can safely use.

 

[ads-ee]

Re: Buying capacitors, difference between same value&same sice but different voltage.

Using a 50V rating capacitor for a ~40V rail is likely to be a reliability problem. The normal recommendations you'll find are you should de-rate capacitors between 2-3x the expected working voltage. So the 5V would be at least a 10V capacitor and the ~40V one would be rated for at least 80V.

- - - Updated - - -

Something like this from Murata (assuming these are just high frequency bypass caps so you can use ceramics)

 

[FvM]

Re: Buying capacitors, difference between same value&same sice but different voltage.

0.1 µF/50V X7R 0805 shouldn't be more than a few euro- or US cent when bought in hundreds quantities. There's no advantage of using smaller voltage ratings in this case. If you want low inductance, you'll primarly go for smaller packages, 0603 or even 0402.

For 40 V node, you'll better chose 100V capacitors or calculate less than 50% of nominal capacitance.

 

[ads-ee]

Re: Buying capacitors, difference between same value&same sice but different voltage.

For 40 V node, you'll better chose 100V capacitors or calculate less than 50% of nominal capacitance.

I think you meant to say 50% of nominal voltage?

 

[FvM]

Re: Buying capacitors, difference between same value&same sice but different voltage.

No, I was talking about expectable capacitance. Using X5R/X7R capacitoirs at 50% of rated voltage maximal is in fact a rule of thumb.

 

[ads-ee]

Re: Buying capacitors, difference between same value&same sice but different voltage.

I was referring to what you wrote in post #4, which was "less than 50% of nominal capacitance".

I know about the 50% rule of thumb for voltage I mentioned it in post #3, for hi-rel you might even want to use 30% of rated voltage maximal.

 

[David_]

Re: Buying capacitors, difference between same value&same sice but different voltage.

They are to be used as decoupling caps next to ICs, in some cases I have found caps where I get 100ps for one or two dollars. But this particular purchase is for one PCB, I have just gotten going with PCB etching and for the moment my wallet is very light but I intend to identify suitable components for future projects and buy a couple of hundreds each, I am mostly doing things ether with 40V max for power supply circuits or things with 12V-3,3V. So when I can afford I'll choose a decoupling cap with a as high voltage rating as I can find/afford and stick to that in my circuits.

I am buying from Elfa distrelect, where would you suggest buying from?
Private person, low quantity(1-100ps perhaps).

When I started looking at capacitors I had thought that I would get ceramics for decoupling and aluminum/tantalum electrolytic for bypass purposes but I found that I could find ceramic at the same capacitance and voltage rating cheaper than electrolytic and I begun to think that maybe the information on caps that I had read is beginning to get outdated, this might sound as a lazy un-informed question but I have done reading but I can't be satisfied being told "for that purpose, use electrolytic not ceramic" and follow that advise without knowing why.

I have very little real experience and I am just now starting to have to decide on what kind of capacitor to use where. I wonder, in regard to power supply filtering and decoupling. Does a ceramic of the same value as a electrolyte have the same effect?

I know that for decoupling tantalum electrolytic as well as aluminum have a limited frequency useability as opposed to ceramic which are useable at far higher frequency's, this project is an adjustable power supply so all I am thinking about is the capacitors in that kind of circuit, rectifier filter, bypass, decoupling for ICs(regulator, ADC, DAC, voltage reference, op-amps), lowpass filter in reference buffer... so in this limited area of application is there anything wrong with using 10µF multilayer ceramic where one usual see and are told to use tantalum or such?

I find that there are very nice hybrid electrolytic, very nice indeed but the cost is considerably higher. But when looking at the specifications the are far superior to the cheap ones, in the end I will have to learn by doing but its never a bed idea to ask around.

Right know there is no switching regulator but that will come.
I feel there is to much stuff rumbling on in my head so I apologize if I might be some

 

[BradtheRad]

Re: Buying capacitors, difference between same value&same sice but different voltage.

Twhere would you suggest buying from?
Private person, low quantity(1-100ps perhaps).

I was happy with my purchase from the Cap King. Low prices.

He states $5 international shipping.

"top quality aluminum electrolytic ultra low ESR capacitors from Rubycon and Nichicon at the best prices online."

 

[David_]

Re: Buying capacitors, difference between same value&same sice but different voltage.

That seems to be a good source, but I really want to keep to SMD as much as possible.
Since I do the etching at home vias can be tough, I have bought a bag of those vias that is inserted in a drilled hole with one end having a slew. We shall see how they turn out but the slew makes them really weird to use, for a 0,8mm via the slew is just about 2mm.

 

[David_]

Re: Buying capacitors, difference between same value&same sice but different voltage.

On digikey I have found some good deals, but something keeps coming back to me cous I just don#t know.
When buying electrolytics I look at the specs in order to only buy caps with low ESR since my main use for them will be related to power supplies, by the way. When is a high ESR electrolytic a viable option?
The ones that I buy has ~0.03Ohm ESR but some caps have 1-5Ohms ESR and that seems worthless in comparison...?

Anyway, when going through and choosing ceramic caps the only specifications given(haven't checked all datasheets but a few) is for example:
voltage rating = 100V
dielectric material = X7R
tolerance = +-10%
and some temperature rating.
But what about stuff like ESR and ESL?
All sources I can find online tell me to use ceramic in certain situations because they have low ESR and ESL, but are they so low that I don't need to care about which of them all I choose?
With electrolytics the specs can vary considerably.

 

[FvM]

Re: Buying capacitors, difference between same value&same sice but different voltage.

You can expect that MLCC have low ESR and ESL, which doesn't vary very much between same size capacitors from different vendors. In most applications you don't need to bother about it.

Some vendors, e.g. Murata have individual data sheets of each type with detailed specification.

 

[BradtheRad]

Re: Buying capacitors, difference between same value&same sice but different voltage.

When is a high ESR electrolytic a viable option?
The ones that I buy has ~0.03Ohm ESR but some caps have 1-5Ohms ESR and that seems worthless in comparison...?

I believe high esr is okay when you have:

* low Amperes
* low voltage swings
* low frequency
* when neighboring impedances are high in comparison.

So just off the top of my head:

Example, for DC blocking a 32 ohm speaker, I suppose esr of 1 or 2 ohms is okay.

To filter a power supply for a 200 ohm load, I suppose esr of 5 ohms is okay.

To power a piezo beeper for 3 seconds after power goes off, I suppose esr of 10 ohms is okay.

Etc.

 

[Dan Mills]

Re: Buying capacitors, difference between same value&same sice but different voltage.

Sometimes you actually want a high ESR type, for example a few ohms of esr can be most helpful to damp the tendency to ring that some power planes can end up with (The bulk caps ESL does not help here).

Lower is not always better, and in fact some of the very low ESR polymer parts can cause real stability issues when hung on the output of an LDO.

Regards, Dan.

 

[David_]

Re: Buying capacitors, difference between same value&same sice but different voltage.

Oh thats interesting.
Especially since the main focus of my search for caps have been for use in a lab power supply made up of a mains transformer - SMPS pre-regulator - linear post-regulator.
In the SMPS circuit I have a switching regulator and the datasheets of those kind of regulator is very clear about the effects the ESR of both the input and output capacitors have, this is my first SMPS circuit and the main focus of the design have been low noise(and dam the expense)
and from what I have gathered I should then use as big a inductor as possible and the ESR of the input/output caps should be as low as possible since the ESR have a direct relation to the level of noise, or ripple, can't remember which.
And I had thought to use the same ones with the linear regulator but perhaps I should not.

Do you know anything about or have some information about how the instability in LDOs occur?