Bipolar electrolytics in xformerless input stages...?

[volki]

many xformerless pre's obviously use electrolytic caps for decoupling at the input. you keep reading that electrolytics, unless they're the bi-polar type, can exhibit non-linear distortion behaviour if not dc-polarized from one side. with op amp designs, no dc voltage than the ic's offset will be present at the cap. so this will mean nonlinear distortion UNLESS phantom is engaged, which polarizes the cap from the other side. this will also mean that the pre sounds different when phantom is engaged.

i'm not familiar with schem's of all those xformerless pre's around, some might not use lytics at all (film caps, rather), some might indeed use bi-polar lytics... just felt like bringing this up, maybe anyone would like to share their thoughts / experiences on this.

cheers,
volker

 

[PRR]

The large electrolytic on the gain-pot of most transformerless amps sits with zero bias and gives very little trouble.

Generally, with ground-referenced inputs, you can use standard electrolytics with the + side to the Phantom. Even if you never use Phantom.

On one design I was playing with, the inputs were biased to about +12V. That means the caps would be polarized one way for dynamic and low-current Phantom mikes, the other way for high-current Phantom and the odd grounded-center-tapped source (old RCA ribbons, a few DIs). And input current noise was too high to afford big enough film-caps. So that would have to use bi-polar caps.