0

Capacitor differences - ceramic or not
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

Capacitor differences - ceramic or not

by uhe on Fri Nov 26, 2010 9:48 am

Hi,

for some capacitors there exists a electrolytic version an a ceramic version e.g. for 470nF.
So if I have a schema with a capacitor rated at 470nF how do I determine what kind of cap I need to use?
What is the difference between a electrolytic and a ceramic cap with the same farad value?
"I have not failed. I've just found 10,000 ways that won't work." - Thomas Edison
uhe
 
Posts: 178
Joined: Mon Sep 03, 2007 4:50 pm
Location: metric world

Re: Capacitor differences - ceramic or not

by adafruit_support_bill on Fri Nov 26, 2010 10:04 am

This page describes various types and their uses. The first page is a summary, subsequent pages go into more detail: http://www.radio-electronics.com/info/data/capacitor/capacitor_types.php

adafruit_support_bill
 
Posts: 76239
Joined: Sat Feb 07, 2009 10:11 am

Re: Capacitor differences - ceramic or not

by stinkbutt on Fri Nov 26, 2010 3:00 pm

The differences between types of capacitors lie in their divergence from an ideal capacitor. That is to say, there are several ways a real capacitor can be described in a circuit that is not ideal:

1. ESL: Equivalent series resistance is the effective resistance of the capacitor to current running through it from the plates to the other end of the circuit, and can be modeled with a resistor (Resr) placed in front of the capacitor.

2. Leakage: Leakage is the tendency for the capacitor to self-discharge. It can be modeled with a resistor places in parallel with the capacitor.

3. Self-inductance: Every part has a small inductance, from surface effects, if nothing else.

Put together, you get a model that looks more like this:

Image

Why am I telling you this? Because Electrolytic capacitors suck in this model.

* For an ideal capacitor R(esr) = 0, a short. Unfortunately R(esr) is very high for Electrolytics.
* For an ideal capacitor R(leakage) = infinity, an open circuit. R(l) is very low for Electrolytics.

In fact, electrolytics have only two virtues which recommend them:

* They're cheap.
* They're big. (in capacity, not physical size, though they are large physically, as well.)

Now, ceramics are also cheap, though not as large, but if you've already access to a ceramic of the appropriate size, then you've just eliminated the one remaining reason to choose an electrolytic over a ceramic; size. You're going to be hard-pressed to find a capacitor CRAPPIER than electrolytics. Sure, ceramics are no great shakes when it comes to some of these qualities, but in the land of the blind (electrolytics) the one-eyed man (ceramics) is king.
Last edited by stinkbutt on Fri Nov 26, 2010 5:12 pm, edited 1 time in total.
Red M&M, Blue M&M: They all wind up the same color

stinkbutt
 
Posts: 593
Joined: Wed Feb 17, 2010 2:40 am

Re: Capacitor differences - ceramic or not

by ImaginaryAxis on Fri Nov 26, 2010 4:28 pm

Stinkbutt and Arduino posted some very good tips on the non-ideal parameters of capacitors.

It also depends on what you are trying to achieve with the capacitor and what kind of application they will be placed in. For example, the capacitor selection for an LDO (linear drop-out regulator) is very important in regards to ESR. The capacitance and the value of the ESR can change the Bode plots to affect stability. This is due to where the poles and zeros are in the Bode plot. In this case a higher ESR can ensure the LDO is stable.

For bypassing in power electronics, aluminum electrolytic are often used as tank capacitors. They provide low frequency bypass and serve as a charge reservoir to help maintain the voltage during high current demand. Low ESR is important here because of the ripple current experienced in the capacitors. A ripple current and a high ESR means power dissipation and can dry them out.

I had to replace the caps in my Viewsonic LCD monitor because of this - but I think that was due to the capacitor plague several years ago.

ImaginaryAxis
 
Posts: 267
Joined: Sat Nov 13, 2010 3:16 pm
Location: Phoenix, AZ

Re: Capacitor differences - ceramic or not

by tinsmith on Fri Nov 26, 2010 11:22 pm

  • Electrolytic: It gets the job done and it comes in big sizes.
  • Ceramic: It does a better job with what it has, but you can't get a big one.
  • Tantalum: When electrolytic won't do and ceramic can't, there's tantalum. However, there may be geopoltical concerns. Low breakdown voltage.
  • Film: Edging towards better quality, but at the cost of compactness. It also suffers at high frequencies (> 100 kHz).

That's probably what you'll use 99.5% of the time, and almost all of that will be spent in the "ceramic if it comes in that value, else electrolytic" regime.
tinsmith
 
Posts: 62
Joined: Thu Nov 18, 2010 9:18 pm

Re: Capacitor differences - ceramic or not

by uhe on Sat Nov 27, 2010 9:05 am

WOW I didn't expect such great comments :)
Okay what this is about? I'd like to build a filtercircuit with an output stage that looks like this:
out.PNG
out.PNG (12.92 KiB) Viewed 2975 times
For C13 the BOM says *electrolytic cap* but local stores only carries the ceramic version.
The ceramic is rated at 50v and the filter should use 12v. So after reading all your comments it shouldn't matter or did I mis sth.?
"I have not failed. I've just found 10,000 ways that won't work." - Thomas Edison
uhe
 
Posts: 178
Joined: Mon Sep 03, 2007 4:50 pm
Location: metric world

Re: Capacitor differences - ceramic or not

by stinkbutt on Sat Nov 27, 2010 2:10 pm

In this circuit you'll be fine with a ceramic. In fact you'll be ever so slightly better off than with an electrolytic, since the ESR and leakage will be better.

What kind of capacitor did they reccomend for C10? I'm curious about this circuit. It's in the signal path, and the signal looks at time to be an AC signal (cap in the signal path) and at times a DC signal (the diode, the polarized capacitors...)
Red M&M, Blue M&M: They all wind up the same color

stinkbutt
 
Posts: 593
Joined: Wed Feb 17, 2010 2:40 am

Re: Capacitor differences - ceramic or not

by uhe on Sat Nov 27, 2010 3:26 pm

The schema / BOM lists an electrolytic cap.
"I have not failed. I've just found 10,000 ways that won't work." - Thomas Edison
uhe
 
Posts: 178
Joined: Mon Sep 03, 2007 4:50 pm
Location: metric world

Re: Capacitor differences - ceramic or not

by stinkbutt on Sat Nov 27, 2010 3:48 pm

Oh, OK. I'm not sure why they think they can get away with polarized caps when they're seeing an AC signal, but it's not like using unpolarized caps is BAD in this circuit.

Any time you see an electrolytic cap, you can substitute an unpolarized ceramic without too much trouble. You can also substitute a polarized tantalum as well. The Polystyrene caps, however, are mission critical and can't be replaced quite so cavalierly.
Red M&M, Blue M&M: They all wind up the same color

stinkbutt
 
Posts: 593
Joined: Wed Feb 17, 2010 2:40 am

Re: Capacitor differences - ceramic or not

by zener on Sat Nov 27, 2010 4:01 pm

That is an interesting circuit. Yes .47uF should be ceramic. I don't think anyone makes an electrolytic that small. I don't think polystyrene is too important in that circuit either.
Last edited by zener on Sat Nov 27, 2010 4:11 pm, edited 1 time in total.

zener
 
Posts: 4558
Joined: Sat Feb 21, 2009 2:38 am

Re: Capacitor differences - ceramic or not

by marzetti on Sat Nov 27, 2010 4:02 pm

Wow! Talk about serendipity, I logged on to ask the very same question about caps and found the answers all laid out for me. Thanks guys.

marzetti
 
Posts: 3
Joined: Wed Nov 03, 2010 11:06 pm

Re: Capacitor differences - ceramic or not

by stinkbutt on Sat Nov 27, 2010 4:32 pm

Zener wrote:That is an interesting circuit. Yes .47uF should be ceramic. I don't think anyone makes an electrolytic that small. I don't think polystyrene is too important in that circuit either.


Yeah, I wouldn't be putting an electrolytic anywhere in the signal path at all - You can basically get better and very nearly as cheap from a ceramic.

I really cannot evaluate how critical the caps that they've designated polystyrene are in the circuit.

marzetti: The bullet from this thread is "electrolytics suck."
Red M&M, Blue M&M: They all wind up the same color

stinkbutt
 
Posts: 593
Joined: Wed Feb 17, 2010 2:40 am

Re: Capacitor differences - ceramic or not

by ImaginaryAxis on Sat Nov 27, 2010 4:39 pm

stinkbutt wrote:Oh, OK. I'm not sure why they think they can get away with polarized caps when they're seeing an AC signal, but it's not like using unpolarized caps is BAD in this circuit.


It does not matter. The node at R17 pulls the cap to ground. The DC bias is assumed higher at the input so even with an AC signal, C10 does not become reversed biased.

ImaginaryAxis
 
Posts: 267
Joined: Sat Nov 13, 2010 3:16 pm
Location: Phoenix, AZ

Re: Capacitor differences - ceramic or not

by stinkbutt on Sat Nov 27, 2010 6:56 pm

OK, I suppose that makes sense. I still wouldn't ever put an electrolytic in a signal path, except perhaps in this particular case, where you've basically downsampling each signal to a 3 bit number, and you're not even using the full bandwidth of a 3 bit number, only using 5 or 6 of the possible 8 values.
Red M&M, Blue M&M: They all wind up the same color

stinkbutt
 
Posts: 593
Joined: Wed Feb 17, 2010 2:40 am

Re: Capacitor differences - ceramic or not

by uhe on Sat Nov 27, 2010 8:17 pm

@marzetti you're welcome :)

Maybe the reason why no ceramic is used is historically as in when this was designed a ceramic version was not available?
Another question I'm thinking about: what would happen if I set VCC to 5v instead of 12v?
I don't intend to use the bargraph output..
"I have not failed. I've just found 10,000 ways that won't work." - Thomas Edison
uhe
 
Posts: 178
Joined: Mon Sep 03, 2007 4:50 pm
Location: metric world

Please be positive and constructive with your questions and comments.