Op Amp distortion - the resolution

[Mike Faithfull]

Thanks to the help received here, my stage monitor is now working properly.
As recommended, power pins decoupled at the ICs, 100 ohm resistors fitted to
isolate the output pin from stray capacitance to ground. I did manage to
borrow a 'scope, and even after applying the remedies it was clear that
crossover distortion was still present - it looked on the screen just like
the drawings in the text books. Well, I s'pose it would really, wouldn't
it!!

It all went away after replacing the LM358 with a TL082 in both parts of the
circuit. It now sounds sweeeeeeeet !!!! Thanks to all - but especially
Phil Allison and Eeyore - for the good advice.

 

[Winfield]

Thanks to the help received here, my stage monitor is now working
properly. As recommended, power pins decoupled at the ICs, 100 ohm
resistors fitted to isolate the output pin from stray capacitance
to ground. I did manage to borrow a 'scope, and even after applying
the remedies it was clear that crossover distortion was still present
- it looked on the screen just like the drawings in the text books.
Well, I s'pose it would really, wouldn't it!!

It all went away after replacing the LM358 with a TL082 in both
parts of the circuit. It now sounds sweeeeeeeet !!!! Thanks to
all - but especially Phil Allison and Eeyore - for the good advice.

It's possible the simple opamp replacement would have
worked, without the new supply bypass capacitors and
the output-capacitive-load decoupling resistor, given
the superior characteristics of even a low-end JFET
opamp, such as a TI's TL082, with its class-AB output
stage, higher loop-gain-bandwidth f_T, and good phase
margin with capacitive loading.

Do you wanna try removing the added parts to see if it
still works? Understand, we recommend that you put 'em
all back again, even if it does work OK. But still, it
would be an interesting experiment. Are you game?

 

[Phil Allison]

"Winfield"

It's possible the simple opamp replacement would have
worked, without the new supply bypass capacitors and
the output-capacitive-load decoupling resistor, given
the superior characteristics of even a low-end JFET
opamp, such as a TI's TL082, with its class-AB output
stage, higher loop-gain-bandwidth f_T, and good phase
margin with capacitive loading.

** The TL072 /082 series are in fact *notorious* for having poor ability to
remain stable with capacitive loads - making an additional resistor in
series with the output almost essential.

Go look up the published schematic.

You may notice the cause.



....... Phil

 

[Mike Faithfull]

Mike Faithfull wrote:

(snip ..)

Do you wanna try removing the added parts to see if it
still works? Understand, we recommend that you put 'em
all back again, even if it does work OK. But still, it
would be an interesting experiment. Are you game?

Not a chance! It's working now and that's how it stays !!

 

[Winfield Hill]

"Winfield"



** The TL072 /082 series are in fact *notorious* for having poor ability to
remain stable with capacitive loads - making an additional resistor in
series with the output almost essential.

Go look up the published schematic.

You may notice the cause.

...... Phil

Yes I see the extra output resistance, good point.
Still, it's gotta be better than the LM324, LM358,
etc., so it'd be interesting if Mike were to play
the game. He could test the added output resistor
aspect by temporarily shorting it out.