Understanding the Circuit: Help with Op-Amp Theory

[davenn]

hi guys
My op-amp theory is a bit weak ... I have gone to a dozen sites trying to identify the layout
in the circuit below ( this was from some one in another forum asking Q's and I got to a point where I couldn't describe any further)

he was first asking about the back to back zeners
so told him it was a symmetrical clipper
The right hand op-amp (U24) is a LPF with feedback to the first op-amp (U27)
U27 appears to be a comparator and going by a bunch of www sites it is
R135/R123 are providing a voltage divider ref to the + pin. but its not a fixed ref because of the feedback link to U24.
So that would imply some sort of ALC ?

EXCEPT

that the - and + pins are reversed ... that is in all examples I could find signal was inputted on the + pin and the ref/feedback was on the - pin
I can't find any reference to the setup as shown here

cheers
Dave

 

[NascentOxygen]

It looks like the pair make a triangle wave generator, or you can call it an astable multivibrator, depending on where you wish to take the output.

The output of the inverting integrator, when it reaches a pre-defined level, trips the comparator and causes an about-turn in the integrator's output ramping.

 

[Baluncore]

This circuit is actually a "slew rate limiter" with a gain of two, a form of low pass filter.

If you follow around the feedback loop it has a net negative feedback so it is a stable follower rather than an oscillator.

U27 is a sort of comparator, but unlike a Schmitt-trigger it operates without hysteresis from positive feedback. The output voltage is symmetrically clamped by the zener diodes VR6 and VR7.

The clamped symmetrical voltage is integrated by the miller integrator, U24, R120 & C39. The integrator slew rate is therefore limited to a maximum. The output from the integrator is fed back through the 50% voltage divider to the comparator, U27, here best seen as an error amplifier looking at the sign of the difference between input voltage and half the output voltage.

The output voltage is therefore twice the input voltage on pin 2 of U27. The slew rate of the output is limited.

 

[NascentOxygen]

You're right, the absence of positive feedback around the comparator changes everything.

 

[jim hardy]

The right hand op-amp (U24) is a LPF with feedback to the first op-amp (U27)
U27 appears to be a comparator and going by a bunch of www sites it is
R135/R123 are providing a voltage divider ref to the + pin. but its not a fixed ref because of the feedback link to U24.
So that would imply some sort of ALC ?

Balun nailed it.

It is helpful to talk to ones self when figuring out circuits. And give your parts personalities.

Back to opamp basics - it's only 'operational' if it is surrounded by a feedback network that allows it to keep its inputs equal. Balance them, i call it.

Can the left hand opamp U27 do that ?
Sure, if there's no interference from that outside input at pin 32.
Input signal comes in on pin 44 and is presumably not shorted to common by Q16.
U27's output gets clipped and handed to the U24 integrator who has time constant (R132+R120)C30.
Integrator's output gets divided by 2 and applied back to U27's +input...
when integrator drives U27's +input = his -input, U27 is satisfied and reduces his output to zero so integrator stops integrating.
U27 can balance his inputs through the integrator.
It's a clever two opamp voltage follower whose output can't move very fast..
Not often do you see a follower with signal applied to + input oops make that -input pin . Since they inverted the feedback to U27 they have to swap his input pins , else feedback around the loop would be positive..

Q16 let's them override input by shorting it to common, which will make U27 drive output of integrator very near zero in order to balance his inputs.
I suspect they limited slew rate to work around some stability problem with the rest of the gizmo. Handy trick, could have even made R120 adjustable.

I'd name that circuit "The Well Tempered Follower" , with apologies to Herr Bach.

old jim

 

[davenn]

thanks guys
appreciated :)

the way U27 was wired was what was annoying me.
As I said earlier, I couldn't find any examples of a comparator wired that way, so I was at a loss to explain its operation

 

[Baluncore]

The error amplifier U27 operates as a comparator only for fast changing or noisy input signals. When the input voltages are reasonably stable it is quite possible for the error voltage to be in the linear mode, to sit close to zero volts and so not be clamped by the zener diodes.

 

[jim hardy]

fixed my blooper.

Not often do you see a follower with signal applied to + input oops make that -input pin .

A comparator usually has a little bit of positive feedback to prevent it from "dithering" when input is near its switch point.
In instrument world we call that hysteresis or 'deadband'.
Positive feedback pushes the inputs farther apart , negative pushes them closer together.

That first 741 stage by itself could certainly be classified a comparator with zero hysteresis.
Sometimes the distinction between linear and logical circuits becomes fuzzy - is PWM digital or analog ?

For a much faster comparator see LM710.
https://www.engineering.uiowa.edu/sites/default/files/ees/files/NI/pdfs/01/04/DS010410.pdf

neat circuit dave.

Now - about them cookies ...

 

[davenn]

help yourself :)