Buffer/Clipping gain stage on same dual opamp?

[MrStab]

Hi guys,

I'm testing the waters of rudimentary PCB design, for a high-gain distortion circuit that works well on vero over 3 "prototypes", and i'm curious about something:

assuming all the appropriate anti-oscillation methods are employed, would it be advisable to have an input buffer and (feedback loop) diode clipping stage on the same dual opamp? up until now, the input & output buffers have been on the same IC, and the actual effect on another, but this is proving to be a challenge when trying to make a board with all the offboard wiring on the same side. too many sections winding round each other - they're low impedance, and the high-impedance input is fairly isolated, but still. i can picture a much better layout if i could just go from left to right, as it were, without "designating" each dual op-amp as such.

i've read some threads with advice on how to avoid unwanted effects with high-gain circuits lately, but can't really find an answer to this specific question.

cheers for any advice!

[anchovie]

I've never had a problem doing that.

[Mark Hammer]

Keeping stages/paths separate by using physically separate chips can be a useful strategy when there is risk of something that stage X does to the power line which could affect stage Y.  That's part of why it's a good idea to use a separate chip for an LFO, so that any power-line spikes can be avoided or kept to just that chip, rather than contaminating other chips.  The textbook scenario would be something like the Zombie Chorus where two op-amps are used for the LFO, and another pair are used for the audio path.  There is a temptation to simply use a quad to simplify things, but now you're stuck with the audio path and LFO sharing the same power, with no option to isolate them from each other.

If there is anything about the two stages you're working with, where the power-line artifacts created would interfere with the optimum functioning of the other stage, then yeah, keep 'em separated and decoupled.  But that's a pretty rare set of circumstances fro my experience here.  If they are both high-gain stages, then the bigger risk is simply having any wires from the two stages too darn close to each other.  I.E., good layout is called for, not separate chips.

[thelonious]

I've never had a problem doing that.

+1.

But Mark has a good point in that sometimes you gotta

[MrStab]

thanks for the vouching, the insights and the tune, guys! much appreciated., you've put my mind at ease and now i can get to work on trying it out. i'm paranoid about LFO's and decoupling etc, but i can't think of any power-related issues in this case as they're just passing signal and not pulsing or anything. hell, in anal-retentive theory terms, maybe it's even "better" to even out current consumption by mixing up the buffers with the gain stages. somehow. maybe.

thanks again