# Op_Amp

Discussion in 'Homework Help' started by Maysoon, Apr 15, 2015.

1. ### Maysoon Thread Starter New Member

Apr 15, 2015
2
0
hey!
please,can anyone tell me the difference between TL084, LM318 and LM741?!
I wanna know which one is better for the circuit shown in the figure if the input signal is noisy and the same amount of noise exists on the ground signal also wanna know why .

Last edited: Apr 15, 2015
2. ### OBW0549 Well-Known Member

Mar 2, 2015
1,881
1,538
The datasheets for those three opamps will list their input-referred voltage noise, either in terms of noise voltage density as a function of frequency (or at one or several frequencies) or as total RMS noise voltage in a specific frequency band. This provides some basis for comparison. R1 and R2 also are sources of noise, as determined by the equation for Johnson noise in resistive components. The opamp noise and the noise from each of the resistors are uncorrelated, so the total noise will be the vector sum of the sources.

Hope this helps a bit; at least, it should give you a starting point.

3. ### Papabravo Expert

Feb 24, 2006
11,156
2,184
Yes, by all means have a look at the datasheet(s). Hint: If the 741 was one day older it could have gone across the Red Sea with Moses.

4. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,071
Digikey lists 36572 opamps in their data base. Their listing goes on for 1463 pages. There will be a quiz about all of them tomorrow...

5. ### WBahn Moderator

Mar 31, 2012
20,247
5,758
What is the metric being used by which one is to be considered "better" than the others?

6. ### Maysoon Thread Starter New Member

Apr 15, 2015
2
0
to neglect the noise and get a clean signal

7. ### MikeML AAC Fanatic!

Oct 2, 2009
5,451
1,071
Are you sure that the noise is originating inside the opamp, or could the input signal be contaminated with noise before it goes into the opamp?

8. ### shteii01 AAC Fanatic!

Feb 19, 2010
3,975
617
The circuit you showed is a simple inverting amplifier with a gain of 2. Examples: 1 volt goes in, -2 volts come out; -1 volt goes in, 2 volts comes out. This circuit is not capable of removing noise.

#12 likes this.
9. ### WBahn Moderator

Mar 31, 2012
20,247
5,758
It may not be able of removing noise, but it is capable of adding noise.

Look for information in the data sheet related to input referred noise and total harmonic distortion and any other noise-related parameters.

10. ### kdillinger Active Member

Jul 26, 2009
141
3
The device that has higher CMRR will be better given your description. If the "noise" is present both in the signal and ground then that would be a common mode signal. An op-amp with high CMRR will reduce the noise the most.
We can help more if we understood what is generating the input signal and where the noise is coming from; keep in mind "noise" is a catch all term. Generally, noise is any unwanted signal, but to be specific there is pink noise, white noise, EMI - whether conducted or radiated from sources like switching power supplies, mobile phones, screaming children...
Using an op-amp to create an active filter may also help depending on the nature of this "noise."

11. ### crutschow Expert

Mar 14, 2008
16,554
4,465
If you have ground noise then you may be able to reduce that by going to a differential instrumentation amp IC and connecting one of the differential inputs directly to the common point of the signal source.

12. ### WBahn Moderator

Mar 31, 2012
20,247
5,758
I don't think that is an option for the TS. I think they have been given a specific circuit and then asked which opamp they would choose, from a noise perspective, for THAT circuit. I'm not saying it's a good problem, only that I think it is the problem the TS has been given.

13. ### crutschow Expert

Mar 14, 2008
16,554
4,465
That certainly may be the case.
But he mentioned something about ground noise which seems odd for a homework problem.

14. ### WBahn Moderator

Mar 31, 2012
20,247
5,758
I think the key, pulling bits from several different comments together, is when they say that the same amount of noise is in the signal and on the ground. Combined with the implication that they are covering CMRR (recently, anyway), that implies that the idea is that the circuit has to work in the presence of common mode noise (particularly since one input is the signal and the other is ground).

15. ### MrAl Distinguished Member

Jun 17, 2014
3,749
791
Hi,

I am not sure if this applies but, FET's usually have lower noise than bipolars, if used correctly of course.
Also, there are some op amps that are made specifically to be low noise, and they state that in the docs. You might be able to find one that works for your application. I am pretty sure they make low noise op amps for use in audio pre-amps, if that helps at all. I was looking for one a long time ago and ran into that.
FET's dont have as much of a certain kind of noise that bipolars have, as long as they are used right, so they became popular for amplifier front ends a long time ago.

LM358's have an output stage that requires the output current to switch between sourcing and sinking and this creates plenty of crossover distortion. If biased correctly however this can be totally eliminated. The idea is to bias it such that the upper output transistor is always passing current.

16. ### kdillinger Active Member

Jul 26, 2009
141
3
Regarding voltage noise density performance in general:

JFET<Bipolar<CMOS

17. ### crutschow Expert

Mar 14, 2008
16,554
4,465
That's generally true for high impedance circuits.
But, if you working at the Johnson–Nyquist thermal noise limit with low impedance circuits, a low noise bipolar transistor or op amp can be better than a FET (some bipolar devices have noise lower than a 100 ohm resistor).

MrAl and OBW0549 like this.
18. ### OBW0549 Well-Known Member

Mar 2, 2015
1,881
1,538
I can't avoid the feeling that we aren't being told the real problem; that something is getting lost (or something extraneous is being added) in translation, or the problem has been misunderstood and we are trying to solve a garbled version of it. I'm having particular difficulty with the "the input signal is noisy and the same amount of noise exists on the ground signal" part. What exactly is meant, here? It would help to see an equivalent circuit containing those noise sources. And by the way, how much noise is on the signal, and what are its spectral characteristics? Put enough noise on the signal, and it's not going to matter which opamp is used; and in any case, no opamp circuit is going to "neglect the noise and get a clean signal" except by filtering, and I don't see any frequency response-shaping components in that circuit.

I suspect the original problem was simply to calculate the noise voltage contribution from the opamp and RSS it with the noise voltage contributions from the two resistors, then choose the lowest overall noise; but I can't be sure from the information we've been given.

We need a better problem statement.

19. ### MrAl Distinguished Member

Jun 17, 2014
3,749
791
Hi,

Maybe he is thinking about common mode rejection.

20. ### crutschow Expert

Mar 14, 2008
16,554
4,465
There's too many "maybe's" in this problem.

Last edited: Apr 18, 2015