Quad Opamps? (Was Re: ASM-1 PCB availability?)

[sasami at blaze.net.au]

Phil wrote:
>hi... why avoid quad ops?  is this only for audio path, or 
in general?

Intra-chip cross talk would be a crude way of describing the 
problem. When you have four opamps sharing the same tiny 
power connections to the common silicon wafer, there will be 
a degree of interaction between the devices on that wafer. A 
similar example is quite easilly demonstrated using a 40206 
hex schmitt inverter. Wire all of the inverters as 
oscillators with the same values for capacitors and 
resistors. Theoretically, the differences in tollerance 
would have the oscillators all drifting apart at slightly 
different frequencies, but in reality, they tend to stay in 
perfect sync.
Bypass capacitors can help, but only so much. The limit is 
the connection between the chip pin and the wafer itself.

To answer the question, this is more of a problem when there 
is a lot of activity on the opamps in question. If they are 
all acting as buffers for fixed voltages, I doubt there 
would be an issue.

Ken

>On Tue, 7 Nov 2000, Jim Patchell wrote:
>
>> Rule #3, avoid quad op-amps.  Duals are OK, singles are 
better.
>> There are some complicated rules I use for choosing an 
op-amp package, and it
>> appears that other people here have similar rules, but to 
keep it simple, I
>
>hi... why avoid quad ops?  is this only for audio path, or 
in general?
>i'm in the middle of building a superseq using tl074's...
>
>thanks,
>Phil
>
>
>