# Noise voltage spectral density from datasheet

1. Mar 16, 2014

### melanie707

Having found a datasheet for a particular component,
I have to find the noise voltage and current spectral density.
In the datasheet I have values like RTI for a gain of 1000 and Rs=0, but I need to find values for when gain G=100 and for different values of frequency.

Does anyone know how to do this?
Would really appreciate any help!

Melanie

2. Mar 16, 2014

### rude man

What is "RTI"? Not a standard abbreviation.

3. Mar 16, 2014

### melanie707

Relative to input

4. Mar 16, 2014

### rude man

This problem needs to be better defined.

If the noise data is rti then the gain is immaterial by definition - but we should still see the circuit and the data. Which device is it so we can look it up ourselves.

5. Mar 16, 2014

### melanie707

Thanks rude man,

What we are actually told to do is:
"For Rs = 0 and gain =100, enter the input noise voltage density and input noise current density for the following frequencies:
10 Hz 100 Hz 1 kHz 10 kHz 100 kHz 1 MHz
(Assume that the manufacturers’ input noise data are for Rs = 0; in many cases, this will be stated explicitly.)"
So I found a datasheet like this, (we are supposed to look at 4)
http://pdf1.alldatasheet.com/datasheet-pdf/view/70328/LINER/LT1168.html
but I don't know how to find these values :/. Really stuck. Our notes cover theory, but nothing on this. I've spent hours and I am clueless.

6. Mar 16, 2014

### rude man

Look at the "Voltage noise density (vnd) vs. frequency" graph in your data sheet, then just read off the vnd from the graph for the different frequencies.

Same for the "Current noise density vs. frequency" graph.

I will add that this is an instrumentation amplifier which is really a concatenation of two amplifiers. For low gains the second amplifier adds significant noise to the output while for high gains the first amplifier noise predominates. Which is why the gain has to be specified even for rti noise, unlike for an op amp.