Capacitor questions

Magnus Danielson magnus at analogue.org
Mon Aug 3 00:43:43 CEST 1998


>>>>> "c" == cmacdon  <cmacdon at ix.netcom.com> writes:

Hi!

 c> I apologize if this is a worn-out retread of an old topic, I'm a bit new
 c> at this.

Well, capacitors MAY seem trivial but that is not really the case with
it's real life function.

 c> I am breadboarding a diode ladder filter and have bought polystyrene and
 c> polypropylene capacitors for the project based on what I have been
 c> reading on various diy web sites (thanks a million to those of you who
 c> put those up BTW!).

 c> I erroneously assumed that all low-value capacitors are small and paid
 c> no attention to the dimensions listed for the polypropylene caps in the
 c> Mouser catalog.  Well the .22uf items are about as big as my thumb!

 c> So my questions are, am I making a silly error here?  Do others use
 c> these giant caps in their filters?  The dimensions for metallized
 c> polypropylene caps are somewhat smaller, are those okay instead?  Does
 c> the capacitor type affect the audio qualities of a filter circuit or is
 c> it mainly for stability or some other reason?

Well, there is many reasons for capacitors being of this and that
size.

For instance, one of the parameters of the dielectrum, the relative
dielectric constant may vary over a range of 1 to 1000. A large value
means you can get smaller physical dimension for the same capacitance
(so you where not totally out in the blue). The downside with most of
the higher values is that they have bad properties in other fields,
such as high leackage and bad linearity. If you want really good
leackage you would turn to Teflon/PTFE which is the best material in
that respect. Many materials tend to absorb water and this will alter
their conductivity/leakage. The trouble is that Teflon has a
dielectric constant of about 2.1 so a capacitor made out of Teflon
will not be very small. If you pick Polystyrene which is about 2.55
then the size would go down somewhat. If you would go with Barium
titane (BaTiO3) you could see the constant to go up to 3600 in one of
the directions and given that you do not put physical strain on it.
The directionallity and strain/preassure sensitivity is certainly an
issue for some applications and you would not like your filters to be
microphonic since you usually have good dynamics in a system.
The temperature stability is another. Since the inner mechanics of the
dielectric comes into play it is not very supprising that temperature
can affect the dielectric constant. On the other side you can see how
the physical properties of a capacitor comes into play. A larger
capacitor will have larger series resistance and also larger series
inductance, both of which are parasitic properties which may be
considered as part of a design. Most of the materials with high
dielectric constant has high internal losses and leakage. Poor
temperature stability on top of that and you know that they don't fit
in audio.

One factor which will control the size is the breakthrough voltage,
which will be linear with the dielectrum thickness and thus requireing
an dielectrum area increase which is also linear.

So, the caps that you have is not at all very of scale, there is
several reasons for caps being picked for audio but sadly is the full
knowledge not very well known. Personally I would go with
polypropylene.

Cheers,
Magnus



More information about the Synth-diy mailing list