At 2:44 PM -0700 07/25/02, Tony Karavidas wrote:
>
>I quickly looked at that page and the reason they don't use DC blocking
>>caps is because those are all bipolar supply circuits. They just don't
>>happen to show the power...
I'm not sure what you mean here. I'm talking about a DC voltage offset at
the output due to the input bias current and input offset current (my
understanding is that the inputs must draw at least some current and that
they are not perfectly symmetrical). Afaik, whether or not the DC source
resistance seen by the input terminals is the same is not dependent on the
type of power supply used. Then again, I'm not an EE and I could be wrong.
>What they are talking about is adjusting for offset errors, not power DC
>offsets.
Right, although I don't think I know what a power DC offset is. I was
talking about adjusting the feedback and input resistors to minimize
offsets at the inputs. Yet, all the examples shown by AD and commercial
manufactures using the chip; either single sided, bipolar, or using a third
half of a 2135 to imitate a split supply; use large caps in series with the
output when used to drive headphones or small speakers. Paul also seems to
agree with the need for such a design. So perhaps the adjustments for
offset errors discussed on the website are insufficient in practice.
Btw, it looks like sdiy is back up.