Subject: Re: Help needed with audio mute circuit
X-Newsreader: Microsoft Outlook Express 6.00.2800.1106
Date: Wed, 18 Sep 2002 05:10:26 GMT
NNTP-Posting-Date: Wed, 18 Sep 2002 07:10:26 MET DST
"LY" schrieb im Newsbeitrag
> > DC biases and AC coupling gets to be such a mess. The problem is that
> > you have multiple derived grounds and they aren't tracking the power
> > supply voltage at the same rate. Just trying to keep the power supply
> > ripple and intermodulation distortion out of the audio can become a
> > nightmare. Muting power on/off spikes is non-trivial.
> Tell me about it! Sounds like you've been down this road before.
> > All of these problems disappear entirely with a split power supply. You
> > get a real ground as a single reference for all signals, end to end. It
> > also eliminates heaps of capacitors, resistors, and regulators.
> Is it a common practice to invert the supply to generate a negative rail?
> What about noise introduced in the inversion process? I know what you
> can't walk two steps before needing another cap. Then there's the issue
> the opamp selection. Finding a single supply, low noise & distortion, and
> CHEAP isn't easy :-) I've often think about throwing in an inverter, but
> CODECs themselves are DC biased already. What inverter IC do you use?
a self-conducting jfet can shorten the output to gnd, but you will need a
negative voltage bigger than the neg.output swing + Upinch-off to reliably
put it into high impedance. Since there is not much current involved, you
can do that with a voltage doubler circuit after your transformer. With this
arrangement the quality will not suffer, as the signal is not passing
through the fet.
Really the relays are not an expensive option, you need only one relays for
multiple channels if there are enough contacts and again you can power it
with a separate rectifier from your transformer to switch before the
regulator starts to drop out.
electronic hardware designer