[sdiy] Re: cap id question

[Happy Harry]

Hi Ingo (et al)

Here's my train (possibly derailed...) of thought.

Consider a parasitic effect like capacitor leakage... this
looks like a shunt resistor in parallel with the cap, usually
a nice high value.  If the cap is in a high impedance circuit,
the percentage error is greater. Just straight ohms law.

OTOH if you consider equivalent series resistance, then the higher
impedance may mask the effect... a low impedance would be more of a
problem here.

Often the parasitic that is most worrysome is the series inductance
of the capacitor (usually from the leads, etc.) which limits the
capacitor high frequency impedance. That's why decoupling (and according to 
Jung... coupling) caps are often made up of several
parallel units of different values. If the caps were perfect, just one
would be fine...

I always parallel decoupling caps. And I never parallel coupling caps.
Sorry... I just use a good cap to start with... and I'm not a bat!

H^) harry


>From: Ingo Debus <debus at cityweb.de>
>To: synth-diy <synth-diy at node12b53.a2000.nl>
>Subject: Re: [sdiy] Re: cap id question
>Date: Mon, 29 Jan 2001 15:33:31 +0100
>
>
>
>harry wrote:
> >
> > Capacitor parasitic elements (like inductance, resistance...) are 
>usually small
> > and low impedance circuits tend to swamp (overcome) them.
>
>But why?
>If I think of a coupling cap between two stages, the cap and the input
>impedance of the 2nd stage form a voltage divider. The higher the input
>impedance, the less a nonideal anything (resistance, inductance,
>nonlinearity) in series with the cap will affect the signal, right? Like
>with CMOS analog switches: the higher the impedance of the circuit, the
>less the quality of the switch matters.
>
>One could say, with higher impedances smaller caps are used. If the
>ratio of capacitive and parasitic impedance of a cap is about the same
>for different cap values the above effect is more or less compensated.
>But how can get things *worse* with higher impedances? Am I missing the
>point completely here? If the circuit impedance is so low that an
>electrolytic has to be used, things are certainly worse.
>
>(Using the term "worse" here regarding signal fidelity).
>
>Ok, for parasitic things in parallel with the cap higher circuit
>impedances are more critical. Are they important for good
>non-electrolytic caps?
>
>Ingo
>

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com