Author Topic: feeding opamp out straight back into negative input?  (Read 503 times)

11-90-an

feeding opamp out straight back into negative input?
« on: February 07, 2021, 08:21:43 AM »
Well I'm analyzing my amp right now, as I have found a sound I would like to reproduce in a pedal, but that's not much of the topic now...

In the amp schematic I found this,



and it feeds the output into the negative input without any resistance whatsoever?

I think I'm missing something... any ideas?

P.S. here's the full schematic...



EDIT:
Well while researching I found this website... https://wiki.analog.com/university/courses/electronics/text/chapter-4
See example 4.7... capacitance multiplier? I think the application in my amp is just some sort of fixed all-pass filter just to use an extra opamp...

As you all probably know (and can see from my reactions :icon_lol:) this is my first time seeing this topology... it *is* quite unusual for me to see this kind of feedback loop, as (from my reasoning) the signal would get cancelled out... any help for my remaining 3 braincells would be greatly appreciated...  :P
« Last Edit: February 07, 2021, 08:31:48 AM by 11-90-an »
flip flop flip flop flip

anotherjim

Re: feeding opamp out straight back into negative input?
« Reply #1 on: February 07, 2021, 09:22:50 AM »
It is common. It gives unity gain (pass signal at x1) because you have signal on the +input and this always gets to the output with x1 gain.
The simplest opamp circuit does it...

In your example, the feedback does two jobs. Setting gain at unity while sending an inverted phase "cancellation" signal back to the input.
If you draw it differently. One wire connected output to -input only and then another wire from output to C11. It's the same, but you can see the two functions.
Croeso i Diystompboxes.

Rob Strand

Re: feeding opamp out straight back into negative input?
« Reply #2 on: February 07, 2021, 12:30:36 PM »
Quote
I think the application in my amp is just some sort of fixed all-pass filter just to use an extra opamp...
It's a Sallen and Key low-pass filter,

https://en.wikipedia.org/wiki/Sallen%E2%80%93Key_topology

In your case,  f0 = 5kHz and Q = 1.6.   Q=1.6 means there's a peak in the response,



I suspect it's way of making the small speaker sound more like a 12".    It will remove a lot of the high-end fizz from the distortion/overdrive.   The circuit U2A is bass boost to EQ the small speaker and cabinet into sounding like a larger 12" cab.  However, I'm not 100% sure the U2A circuit is drawn correctly as C10 doesn't do much.  If possible find another Laney schematic with this circuit and check, or check/trace your actual PCB.
The internet:  answers without the need for understanding.

PRR

Re: feeding opamp out straight back into negative input?
« Reply #3 on: February 07, 2021, 06:43:35 PM »

  • SUPPORTER

Rob Strand

Re: feeding opamp out straight back into negative input?
« Reply #4 on: February 07, 2021, 08:11:02 PM »
I looked at the circuits for the LG-20R and LG-35R and they don't have a filter like the LG-12.

The LG-20R and LG-35R use Laney's standard solid state preamp: split channels, CMOS 4069 the overdrive., the MAX button which I think is boost + mid scoop.
« Last Edit: February 07, 2021, 09:20:04 PM by Rob Strand »
The internet:  answers without the need for understanding.

Rob Strand

Re: feeding opamp out straight back into negative input?
« Reply #5 on: February 08, 2021, 01:13:51 AM »
The HCM-10 has a similar preamp to the LG-12.  The bass boost EQ and low-pass filter are the same.   So C10 still doesn't make sense.
The internet:  answers without the need for understanding.

11-90-an

Re: feeding opamp out straight back into negative input?
« Reply #6 on: February 08, 2021, 02:44:43 AM »
So C10 still doesn't make sense.

I'm obviously not sure, but to me it *seems* like some sort of 2-pole filter?
flip flop flip flop flip

Rob Strand

Re: feeding opamp out straight back into negative input?
« Reply #7 on: February 08, 2021, 03:33:36 AM »
Quote
I'm obviously not sure, but to me it *seems* like some sort of 2-pole filter?
In theory it's a two pole filter but  in practice it acts like a one pole-filter.

On it's own, the 33n + 22k + 47k part would affect say 100Hz to 300Hz.  However the impedance of the 470nF cap is 3k ohm at 100Hz and 1 kohm at 300Hz, which is a lot lower than 22k and 47k.    So the 470nF cap bypasses the 33n + 22k + 47k part altogether.  If we replaced the  33n + 22k + 47k part with a resistor I suspect the circuit would behave the same.   It's not doing any harm, I just can see it doing anything at all  ;D.
The internet:  answers without the need for understanding.

PRR

Re: feeding opamp out straight back into negative input?
« Reply #8 on: February 08, 2021, 09:38:31 PM »
> I just can see it doing anything at all

Seems to give a 26dB boost at 4Hz.  Subsonic thump. I wonder if it is a transcription error. 0.047uFd?
  • SUPPORTER

Rob Strand

Re: feeding opamp out straight back into negative input?
« Reply #9 on: February 08, 2021, 11:17:49 PM »
Quote
Seems to give a 26dB boost at 4Hz.  Subsonic thump. I wonder if it is a transcription error. 0.047uFd?
It starts boosting at about 75Hz, that in itself might be OK.   It then keeps rising as we go lower to that crazy figure.

A 47n might explain it.   That would be the correction for the small baffle size then the 33nF could be correction for the roll-off of the speaker itself.

Best resolved with by looking at the PCB.
The internet:  answers without the need for understanding.