ENS-76 VCO Option 2 question

Bob Zimmer bzimmer at voicenet.com
Tue Mar 25 18:38:52 CET 1997

At 09:02 AM 3/25/97 +0100, you wrote:
>Replacing the 556 with an 071 causes it to generate a low amplitude sine
>(+-2.5V) that does not change frequency regardless of the frequency of the
>Oha, Oha !
>This smells like self oscillating of the 071.
>This is STRANGE ! 

Agreed!  It is definitely oscillating independent of the VCO itself!

>Self oscillation is the big problem of feedback circuits
>like op-amps.
>The open loop gain characteristic of the op amp is very important for this.
>The tl082/071 might have a different open loop gain & phase
>characteristic, but when I took a look to the data sheets I got,
>I didn't notice that. Maybe I was too blind to see.
>I had however the feeling that the two datasheets are just copys
>of one or the other and that someone later tryed to put
>the differences in, but made some mistakes.
>However, tl071/082 have a much higher open loop gain than 741.
>The sluggish 741 maybe therefore not so sensitive for oscillation 
>problems in this appl.
>What do you mean by : can't drive headphones ?
>The headphone and cable may be  a capacitive load for the opamp
>this will cause delay and so the phase problem will be worse.
>More oscillation should be the result.
>This occurs though you have a 1k output resistor.

I can take an ordinary set of headphones and plug them into any of the other
waveform outputs and hear the output.  In the case of this one output, I
hear nothing.  If I remove the resistor as mentioned in my other post and
short the input to ground, it not only stops the independent oscillation and
starts following the VCO, I can also hear what I would expect through the

>This shows  how instable the circuit is !
>One more idea : since tl071 etc. have jfet inputs,
>you might have a capacitive coupling problem.
>Make shure that the (-) traces are very short and away from the output
>and supply traces.

For now, this is just on a breadboard, not on a PC Board.

>Hotfixes :
>Try to reduce gain by using a cap parallel to 39k feedback res.
>A bandwith of 20kHz is enough for audio.
>Try supply decoupling caps right at the opamp against goog, clean ground.
>Hope I could help !

  >=== Bob Zimmer -- Philadelphia PA ===<
  >=== bzimmer at voicenet.com          ===<

More information about the Synth-diy mailing list