Trying to get lower noise with an op-amp based amplifier

OBW0549

Joined Mar 2, 2015
3,566
This is the noise "look" on an average result using batteries.
noise batteries metal box.png
That's kind of disturbing. I see a lot of shot noise, aka popcorn noise-- the sudden apparent jump from one level to another. I also see a fair amount of flicker noise (1/f noise), which I would expect with this op amp. There's also a hint of external interference; it looks a bit like some radio station in the background. Altogether, it's definitely not pure "white" noise.

Attached is some reading material on noise; I think you're going to have to do some digging, here.:(
 

Attachments

Hi everyone.

I’m trying to desing a simple low noise amplifier but I’m not able to reach low enough noise, and I suspect I’m missing something. I would appreciate some help to track the problem down.

The circuit is built around a TL972 (this opamp is 4 nV/sqrt(Hz)).
I used a non-inverting 101x voltage gain, with two poles.
View attachment 215755
C2 (with R2) sets a high pole at about 40 kHz.
C1 (with R1) sets a low frequency pole at about 7 Hz.
Both poles set the passband.
View attachment 215756

Circuit works perfectly (built and tested), except for the noise expectations.
When I short input to ground, I measure 1 mV – 5 mV (rms) at the output, suggesting an EIN of 10-50 uV rms (due to the approx 100x voltage amplification).

There’re two problems:
1)- Probably 10 uV is not that bad, considering the circuit is not inside a metal enclosure (I was aiming for 5 uV max).
2)- The real issue is that noise level varies with apparently irrelevant things. Sometime the noise is 10-20 uV (quite stable), but then I move some meters away or closer, or move some wires around, and the noise gets 40-50 uV (stable again). And it’s possible to revert and partially control these changes, suggesting there’s a very concise reason behind which I couldn't find after several hours of trial and error.

Of course I did not use a protoboard. I soldered all components as tight as I could (no flying wires), and used 1000 uF close to each supply pin and ground. I also tried to short C1, which (maybe) improves very little. The power supply is linear, and no close SMPS was powered on.

Any tip? Anything I could try to achive lower noise with the same op-amp?
Thank you!
I would suggest step down transformer at the input and coupled inductor at the output...i did this before and it works
 

Thread Starter

Elerion

Joined Sep 11, 2017
125
Attached is some reading material on noise; I think you're going to have to do some digging, here.
Thank you. I'm currently reading at a good speed :)

3) The TL972 lists its input equivalent noise voltage density as 4 nV/√Hz at 100 kHz, but it says nothing about input current noise density. And if this circuit is going to be driven from a high-impedance source such as 10 kΩ, it's likely the current noise that will cause the biggest problems rather than the voltage noise.
I tried my setup with a 220 kohm resistor between input and ground (so the source impedance is high), to test how the TL972 performs.
It seems like the thermal noise of the resistor is even stronger than the contribution of the opamp's noise current.
I measured 1.2 mV rms noise at the output (100x gain). Thermal noise of 220 kohm at 300 ºK and 20 kHz is about 8.5 uV, which would be amplified to 0.85 mV at the output. Very close. Substituting the resistor by a 100 ohm one, the output noise drops to 350 uV. So the test seems to be valid.

So, why would a very low noise current opamp be essential (with high source impedance) if the thermal noise of the source resistor is much higher and dominates??
 

OBW0549

Joined Mar 2, 2015
3,566
So, why would a very low noise current opamp be essential (with high source impedance) if the thermal noise of the source resistor is much higher and dominates??
If the thermal noise of your source resistor is dominating, and the op amp's current noise is not causing any significant added voltage noise by flowing through it, then your op amp's current noise is not an issue. My initial guess that it would be very high, was apparently off the mark.
 

Thread Starter

Elerion

Joined Sep 11, 2017
125
I managed to get very good results. In this post I'm sharing my findings.

After adding some low value caps in the supply rails, adding a high pass filter at the input and improving the shielding as much as I could, I managed to get values very close to those in the datasheets for some opamps.

A simple TL072 does a better job than the TL972, not only in a high impedance situation, but also with input shorted. The EIN voltage improvement is about 2x. To be fair, the TL972 based amplifier has a greater bandwidth (about 2x), but still, I don't get even close to the datasheet's specs. I do get very close for the TL072 (measured 25 nV/sqrt(Hz) vs datasheet's 18 nV/sqrt(Hz)).

These are some captures of TL072 noise at the amplifier's output.

Rs_0_shielded.pngRs_0_shielded_140ms.png


That's kind of disturbing.
These look much better, don't they?

Even a LM358A works fairly well and I get very close to the datasheet's specs! I measured 65 nV/sqrt(Hz) vs datasheet's 55 nV/sqrt(Hz). Even though noise current is not published in the datasheet, it's performance with high impedance sources is astonishing. Very close to TL072. Surprisingly, IMHO the LM358 does better than the TL972, but the lower bandwidth makes me discard it as an alternative. To be fair, I used a LM358A (from ST): an enhanced LM385 (mainly lower offsets and bias current, but same frequency and noise performance).

For the TL972, using the exact same setup, I measure 55 nV/sqrt(Hz) vs datasheet's 4 nV/sqrt(Hz)) !!! I repeated the tests several times. By the way, I tried three different TL972. All opamps used in these tests are 100% originals bought from trusted sellers. So, personally, I'll avoid TL972. To make matters worse, it's current consumption is higher than any of the other alternatives.

Hope this info is of some help. Thanks to everyone who participated in this thread.
 

ci139

Joined Jul 11, 2016
1,898
or the TL972, using the exact same setup,
. . .
actually if you plot the \(I_{V_{CC}}\) and / or \(I_{V_{EE}}\) against the \(V_{OUT}\) . . . for different Op.-Amp.-s you get different results . . .
. . . that means that the decoupling schemes should be speciffic for the Op.-Amp. being tested
(just a note :: EDIT : the above was brought out coz i still speculate that you may be able to reach your TL972 spec.-s if you browse Ti-s AN-s and user forums ...)
EDIT.2 : the OPA1692 d/s Fig.12 shows p-p noise voltage of about ±50nV ? is about ≥25mV.RMS . . . ?? corresponds to 33Hz bandwidth for the mean of the specified 4.2 4,5 nV/√[Hz] ??
 
Last edited:

Audioguru again

Joined Oct 21, 2019
6,674
The noise images look different because the 'scope timebase and voltage settings are completely different.

The lousy LM358 was mentioned. Some have ordinary noise level and others have very high noise.
in addition to the low bandwidth they produce horrible crossover distortion like this:
 

Attachments

atferrari

Joined Jan 6, 2004
4,764
The noise images look different because the 'scope timebase and voltage settings are completely different.

The lousy LM358 was mentioned. Some have ordinary noise level and others have very high noise.
in addition to the low bandwidth they produce horrible crossover distortion like this:
IIRC the datasheet by Nat Semi detailed a cure for that distortion. IIRC again, it was a resistor from somewhere going to the negative rail.
 

Audioguru again

Joined Oct 21, 2019
6,674
The output transistors in most opamps are biased with a low current in class-AB so they do not produce crossover distortion.
But the LM324 quad and LM358 dual were designed for very low supply current then their output transistors have no bias current and operate in class-B which produces crossover distortion.

If course, you can add a DC output resistor so that one output transistor conducts a fairly high current all the time in class-A.
But the resistor does not fix the high noise level and poor high frequency response, Use a modern audio opamp instead.
 

Thread Starter

Elerion

Joined Sep 11, 2017
125
If course, you can add a DC output resistor so that one output transistor conducts a fairly high current all the time in class-A.
But the resistor does not fix the high noise level and poor high frequency response, Use a modern audio opamp instead.
I do like the LM358 for some applications: it is able to go down to 0 V using single supply, current consumption is very low, it's cheap, and some applications benefit from the low bandwidth (from a stability point of view). OP07 is also an old low bandwidth opamp, but some specs are still great for me, like its low offset voltage and drift.
 
Top