If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

• Resolved

# Op Amp specification for conductor thermal noise measurement application

Dear Engineers,

I want to asking about Op Amp specification to realize thermal noise measurement on conductor.

What the most important Op Amp specification for that application?
What the specification for measure noise voltage and noise current is different?
Can you give me example the Op Amps type which are appropriate for that application?

Thanks,

Whenever you use an op amp to measure the noise of something, the most important paramaters are:

1. 1.      Voltage noise spectral density
2. 2.      Current noise spectral density
3. 3.      Bandwidth

The OPA211 is an example of a device with very low voltage spectral density (1.1.nV/rtHz). The current noise for this device is not optimized (3.2pA/rtHz).

The OPA827 has good voltage noise 4nV/rtHz, and good current noise 2.2fA/rtHz.

I highly suggest going through the noise precision labs videos. There are 9 videos and they cover a lot of detail on this subject.

https://training.ti.com/ti-precision-labs-op-amps-noise-1?cu=14685

Art Kay Senior Applications Engineer High Performance Linear

• In reply to Art Kay:

Art,
I have read all of the slide in training.ti.com/ti-precision-labs-op-amps-noise-1, it really helped me to understand the noise in op amps. As a first step I want try to design 10uV amplifier using lm4562. Noise on the input according to my calculations is about 0.4uV with gain 100 and bandwidth of 100Hz. I think it would be good if we compare the noise in the calculation and the actual circuit. Do you know the circuit to test the noise on my amplifier design?

Thanks,

Art,
For your information, I want to compose the noise measurement circuit using protoboard.
Thanks,

A common method used to measure low noise circuits is discribed in the TI design given in the link below.  The key is to follow your amplifier with low noise amplifier that will boost the noise signal to a level you can measure with a scope.  If you provide a schmatic of your circuit we can simulate this.  You should definately sim it before you build the measurement circuit.

http://www.ti.com/tool/TIPD147?CMP=AFC-conv_octorefdesigns

Art Kay Senior Applications Engineer High Performance Linear

• In reply to Art Kay:

I want to ask about the ground plane and ground chassis connection which I think is the important thing in low voltage measurement.
If I am using double supply on Op amp with VGND is half of supply voltage, what voltage should I connect to the ground plane and chassis ground for common voltage? The lowest voltage (the same value as op-amp negative supply) or the VGND which is half of op-amp positive supply?
I hope you can explain the reason why we choose that.
Thanks,

Attached is a good presentation that covers the subjects of chassis grounding, and PCB grounding.  I think slides 39-41 are most pertinent to your question. I suggest looking through the slides and see if they help. Henry Ott has a very good book on the subject also “/Electromagnetic Compatibility Engineering”.

http://www.amazon.com/Electromagnetic-Compatibility-Engineering-Henry-Ott/dp/0470189304/ref=sr_1_1?ie=UTF8&qid=1436386611&sr=8-1&keywords=henry+ott

Art Kay Senior Applications Engineer High Performance Linear

• In reply to Art Kay:

Art,

I read some of your OP Amp noise presentation. and learn that we use some gain and post gain amp to get the Amps noise plot in 0.1-10hz.

but, I'm confuse that, such like in OPA333 datasheet page1, there is a noise plot, the resolution is about 500nV/div, how we can get such low resolution of oscilloscope?

Regards,

We use a low-noise post amplification stage to increase the output noise of the device under test to a level we can easily measure using the oscilloscope. So the measurement in the datasheet is taken with a 1mV or 5mV vertical division on the oscilloscope and then we edit the axis for the datasheet plot to correct for the gain of the amplification stage.

Regards,
Collin Wells
General Purpose Amplifier Applications

• In reply to Collin Wells:

Just to add to what Collin said.  We have a TI Precision design reference design that covers this.  See link below

Art Kay Senior Applications Engineer High Performance Linear

Three major factors to affect Noise：Voltage，Current，Resistor；
Voltage (Amp Bandwidth Gain)；
Current (Amp Bandwidth Gain Req)；
Resistor (Bandwidth Temp Gain Req)；
The OPA211 OPA188 and OPA277 are good example of a device to understand noise theory.
oscilloscope setup: 1.Full bandwidth; 2. 1s/div if there is only 10 divs at your oscilloscope screen; 3. BNC adapter but not oscilloprobe;
4. first noise floor test in shielding box; 5. Secondly noise total test in shielding box; 6. standard deviation to real noise;

I suggest going through the noise precision labs explanation:
www.ti.com.cn/.../index.shtml