Film capacitor polarity

I just read an article in elector(don't laugh, i got a stack of them from a friend) where they claimed that even film capacitors will work differently depending on how you place them in acircuit. Especially in high impedance circuits.

Their reasoning was that because film capacitors are rolled up stacks of film. One side will be connected to the outside-film and form some kind of shield. So if you use a capacitor to ground, it will be better to connect the outer film to ground to reduce rf interference and stuff like that. They even had a circuit to find out what leg the outer film was connected to.

Is this effect noticable in audio applications? Do the ceramic capacitors most people use to decouple op-amps also have this 'feature'?

Is this effect noticable in audio applications?

Click to expand...

There is only one way to find out, right?

/Anders

Just remember that not all film caps are wound - some of them are stacked.

But why would the outside foil be marked on many film caps, if it didn't matter?

Best regards,

Mikkel C. Simonsen

Hi Mendelt,

[quote author="Mendelt"]I just read an article in elector(don't laugh, i got a stack of them from a friend) where they claimed that even film capacitors will work differently depending on how you place them in acircuit. Especially in high impedance circuits.

Their reasoning was that because film capacitors are rolled up stacks of film. One side will be connected to the outside-film and form some kind of shield. So if you use a capacitor to ground, it will be better to connect the outer film to ground to reduce rf interference and stuff like that. They even had a circuit to find out what leg the outer film was connected to.[/quote]
I've read that one as well and I think it makes sense. Actually fairly sure about it, really. And it's being mentioned for a long time already, that Elektuur-article isn't the first and only one.
Many caps have/had a marking (een zwarte streep), indicating the side of the cap that'd best be connected to the most low-Z point of the circuit. IIC certain Philips caps had the 'Philips'-name 90 degrees rotated in bold, forming that black line.

And believe it or not, that polarity thing came to my mind when stuffing the caps on the Pultec-EQ-boards last week - if one has to insert them this or that way, then better according to that rule, why not ? It'll not matter much (if at all) in lots of cases, but it doesn't cost any additional effort either.

Is this effect noticable in audio applications? Do the ceramic capacitors most people use to decouple op-amps also have this 'feature'?

Click to expand...

I haven't ever tried both ways, but can imagine it can make a difference w.r.t. disturbances - that's actually what this is all about.
Because of their different construction I think this isn't an issue for ceramics. And in addition: decoupling is usually between two low-Z nodes, so determining the relative low-Z & hi-Z nodes is not relevant and will also be challenging :wink:

Bye,

Peter

Wow. Thanks for all the replies :grin:

It's nice to know things like this. Just flipping one capacitor probably won't make a difference but the nice thing about DIY is that you have time to do all kinds of little tricks to make your gear just a little better.

old news. Look in the advanced section.

http://www.aikenamps.com/

[quote author="Gus"]old news. Look in the advanced section.
[/quote]

Thanks, I didn't know that site. Lot's of info there.