[sdiy] Electrolytic Caps for Audio Decoupling

[Seb Francis]

Grant Richter wrote:

>>>
>> In the THAT datasheets and design notes they show polarized caps  
>> with the negative end connected via a resistor to the IC inputs  
>> which are at virtual ground.  And in fact many times I have seen  
>> polarized caps used in places like this where in all likelyhood the  
>> polarity will be reversed about half of the time.
>>
>
> Once again, there is no voltage difference between the plates of the  
> cap because of the high impedance.
> Consider it as a voltage divider.
> Any resistor value into a very high impedance will have no  
> significant voltage drop across it, because there is no current flow  
> to cause a voltage drop.
>
>
The THAT IC inputs are current inputs at virtual ground, so the input 
resistor (20K in this case) is the impedance - which I guess is high 
enough not to allow significant voltage drop across the capacitor, 
unless there is a DC component to the input signal (which there won't be 
unless one of my sound sources is not working correctly).  But I can see 
that in a modular synth situation it could very likely be the case that 
an input had a significant DC component.

Anyway, thanks to everyone who responded about this.  I now have a much 
clearer idea about the ins and outs of AC-coupling (oops, almost said 
DEcoupling then ;)

On the breadboard I'm going to use whatever cheap electrolytics I have 
lying around, but maybe I'll invest in some solid aluminium ones as 
recommended by Tony and see if I can hear any difference..

Seb