# Op-amp feedback

Discussion in 'Homework Help' started by hebryn, Dec 9, 2013.

1. ### hebryn Thread Starter New Member

Dec 9, 2013
1
0
Hi

I'm having troubles understanding the attached op-amp circuit. This op-amp should according to the one making the answers, have negative feedback. I would say the feedback was positive.

What is correct, negative or positive feedback? and if its negative why is that?

File size:
39 KB
Views:
41
2. ### Jony130 AAC Fanatic!

Feb 17, 2009
4,166
1,188
Let as assume that Io increase his value, so Vf voltage drop across Rf resistor also increase. So the voltage at non-inverting input also increase its value.
If so, the op amp output voltage also have to increase. This increase in output voltage will reduce Vsg voltage and P_MOS will reduce his drain (Io) current. As you can see we have a negative feedback circuit. And it is all thanks to P-MOS which inverts the phase by 180 degrees.

Last edited: Dec 9, 2013
hebryn likes this.
3. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,070
And this circuit will make a better oscillator than an amplifer.

4. ### Jony130 AAC Fanatic!

Feb 17, 2009
4,166
1,188
Can you explain why this circuit will oscillate?

5. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,070
The opamp may intrinsically be unconditionally stable with 100% feedback, but when you add the outboard inverter stage, the added phase shift makes the combined circuit unstable.

6. ### LvW Active Member

Jun 13, 2013
674
100
No - I don`t think that it is certain that it will oscillate.
The only thing you can say is that the phase margin of this circuit will be smaller if compared with the classical negative feeedback scheme (resistive divider connected to the inv. opamp terminal) without any additional transistor in the feedback loop.

Of course, there is a "chance" for instability (oscillations) - however it depends on the frequency characteristic of both gain functions (opamp and FET) as well as the resistor ratio.

7. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,070
I have built this. Take my word for it, it oscillates...

8. ### LvW Active Member

Jun 13, 2013
674
100
For any resistor values and ratios?

Example:
FET amplifier with a gain of -5 (both resistors in series) and a resistor ratio of 4:1.
Thus the feedback factor will be unity. A universal-compensated opamp will NOT oscillate. Agreed?

Last edited: Dec 9, 2013
9. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,070
Not agreed. This sim shows typical oscillatory results of the OP's circuit, the one I said makes a better oscillator. The components are the same as the one I built. I was able to stabilize it, but I had to add some feedback to the inverting inpu. It will take me a while to find the final, compensated circuit that worked...

• ###### DF138.jpg
File size:
91.7 KB
Views:
38
Last edited: Dec 9, 2013
10. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,070
I found it. Here is what it took to tame this beast. Note that by changing C1 I am able to get anything from under-damped to critically-damped to over-damped.

• ###### DF138a.jpg
File size:
283.2 KB
Views:
28

Aug 17, 2013
858
40
I also thought that positive feedback was when the loop was connected to the non-inverting input, and negative feedback when connected to the inverting input.

So, Is the sign of the feedback dependent on the sign of the feedback voltage, whatever input the loop is connected to? I mean, if the feedback is a negative voltage connected to the non-inverting input, do I have negative feedback; and viceversa?

12. ### WBahn Moderator

Mar 31, 2012
20,057
5,644
Negative feedback means that if the output voltage goes up, say due to noise, that the feedback action results in the output voltage being driven down. Positive feedback means the opposite -- a positive voltage excursion in the output results in feedback action that results in the output voltage being driven more positive.

Imagine making a break somewhere in the feedback loop. Often times making this break right at the output of the opamp is most convenient, but it really doesn't matter. Call the voltage at the upstream part of the break Vo' and the voltage at the downstream part of the break Vo. Now, if you change Vo (and nothing else), the circuit will respond by changing Vo'. In general, you will have:

Vo' = K*Vo

Vo'/Vo = K

K is the loop gain. If K is positive, you have positive feedback. If K is negative, you have negative feedback.

13. ### LvW Active Member

Jun 13, 2013
674
100
MikeML - I think it is a nice and helpful exercise to proof if and why a particular circuit will oscillate. And the circuit under discussion - supplemented by your simulation results - is a very good example.

Therefore, back to your example circuit: I think, it is no surprise that it oscillates.
There are two reasons:
1.) As can be seen by comparing (in your simulation results) the gate ac voltage with the ac voltage at the opamp input, the overall feedback factor is larger than unity (that means: The gain of the FET amplifier is larger than the max. allowed value of 5, given a 4:1 resistor ratio in the drain path).
2.) You have selected a POWER MOSFET (by accident or intentionally?) with very bad high-frequency characteristics. For example, the input capacitance is in the order of 0.5nF. As a consequence, the gain drops already in the lower kHz range.
________
Thus, it is really no surprise that the circuit oscillates because the FET circuit acts as a lowpass with corresponding additional phase shift.
I think, the situation will be different if a small-signal MOSFET with the correct gain value is used.

Therefore, I think my general comments as given in former posts still hold:

The circuit under discussion may oscillate or not - depending on the frequency response of both active elements (opamp and FET) as well as on the overall feedback factor which is set by the two drain resistors.

Final comment: Anybody who is not familiar with the concept of stability and oscillations could use this feedback example to see and understand why oscillations might occur (or not).

EDIT:
I have replaced the Power mosfet by a small-signal P-Jfet (2N2608) - and did some simulations (TRAN). As a result, the circuit was absolutely stable for a varity of opamp models (741, LM328, AD712, LT1007,...).

Last edited: Dec 10, 2013
14. ### LvW Active Member

Jun 13, 2013
674
100
We have negative feedback if anywhere in the feedback loop we have a signal inversion (minus sign - identical to 180 deg phase shift).

15. ### WBahn Moderator

Mar 31, 2012
20,057
5,644
Unless we have two!

16. ### LvW Active Member

Jun 13, 2013
674
100
Yes - correct. Thus, I should say: ...one single inversion only (or 3, 5,...).

Aug 17, 2013
858
40
Thanks,

So, would it be right to put it this way: the sign of the feedback depends on whether the output -or the gain- is influenced positively or negatively; no matter what input it's connected to?

18. ### LvW Active Member

Jun 13, 2013
674
100
To start with your last sentence: Yes - it does not matter if the feedback signal is connected to the inv. or non-inv. input. What matters is the number of phase inversions (factor "-1") within the feedback loop. In 99.5% of all cases this inversion is realized using the inverting input terminal - however, your example shows that this is not necessary.

Regarding the first sentence, I think the wording is somewhat confusing.
What means "...the ouput is influenced positively..." . Do you know what you mean? For my opinion, it is quite sufficient to require an odd number of phase inversions.

By the way: Do you know why negative feedback can cause oscillations? And more specifically, do you know why in your circuit oscillations might occur under certain conditions - and how you could stabilize the system?

Aug 17, 2013
858
40
What I meant is: if the feedback makes the output drop, then it's negative; and if it makes it raise, then it's positive.

Someone else opened this thread with that circuit. For now, that circuit, why it oscillates and phase inversions are too advanced for me; I just began with op-amps.

20. ### Jony130 AAC Fanatic!

Feb 17, 2009
4,166
1,188
Can you tell me how do you get this number without knowing op amp type?