AW: Re: external input (again)

[Don Tillman]

   Date: Sun, 26 May 1996 11:03:55 -0800
   From: haines at apc.net (Matt Haines)
   >
   >And lose the gratuitous polarity inversion.  (That's considered tacky
   >in both audiophile and engineering circles.)

   As in, add another op amp? 
Heavens, no.

   Or should I be considering using the + input?
Yes, exactly!

   >In general, when proposing a circuit and asking for advice it's a good
   >idea to give us as many specifics as possible about what you want and
   >the context the circuit is going to be used in.  Input impedance,
   >output impedance, gain, noise, overload issues, input/output
   >protection issues, low/high end frequency reponse, that sort of thing.

   Very true...assuming I know even to ask those questions! :) As far as
   impedance goes, I'd assumed I would be dealing with the typical 100k high
   input impedance and 1k low output impedance. Not so? I'm most likely going
   to take the output of a sampler or synth and feed it into the amplifier. So
   what does that make it?

Those are certainly reasonable numbers.  Pro-quality mixers usually
have a 10k to 30k input impedance for line-level inputs.  100k makes
it a little more adaptable to other stuff.

   Re: gain, I *think* I covered that, in that I'm trying to bring a synth's
   output up to a practical modular-synth level. +/-0.5v (which is what audio
   signals seem to be around, for synths anyway), up to +/-5v. Which is a gain
   of 10.

Well if you're not sure of the input voltage, or if you want to make
the device a little more flexible, a switch-selectable gain dealie
would be the right thing.  For example, check out the preamp on the
ARP2600.

   Overload...I'd like the option to overload the amp if I *wanted* to, but
   that's not something I want happening all the time.

Opamps by themselves don't overload gracefullly.  Limiting zeners in
the feedback loop would be an improvement.  If you're convinced it's
not going to overdrive, it's not an issue.

   Protection...yes, please! But seriously, I really don't know what to ask
   here. I'm not going to be sticking any unusual signal sources in the thing.

Folks in my band with advanced degrees in electronical engineering
seem incapable of getting the grounding right.  If your input signal
finds itself riding on on top of a 115v 60Hz sine wave it's a good
idea to have reverse-biased diodes to the power +/- supplies as well
as in input resistor.

   Freq response...not crucial, as in not audiophile. I'm going to be running
   it through horribly mind-twisting processors anyway. 20-20k not necessary,
   but 200-5k in not good enough.

With opamp circuits it's a good idea to roll of the high end in the
feedback network to limit the gain in the region where the opamp is
less able to handle it.

   As far as getting a book on op-amps, yes it's something I need to do. I'm
   much more likely to get answers specific to my problem here on DIY though.

It's important... at the very least it will give you the warm fuzzies
about how these things work.

  -- Don