Incorrect result of noise calculation in Multisim (AD)

[Dimkaf]

During evaluation Multisim ( Analog) was detect following troubles:
When we using SPICE model AD in conventional circuit for noise estimation then our result matched to values in chart of datasheet (see file TestAmp.ms12).In the case of noise estimation for circuit like differential amplifier (see file LDR_AMp2n.ms12), value of noise from two OpAmp less then single.How it's possible?

[Tayyab_R]

Hi Dimkaf,

I looked at the result from the noise analysis and it seems like the one from the differential amplifier is higher as expected. The only thing which might have caused confusion is that in the second, the Stop frequeny was 1 MHz while in the first file, it is 10 GHz. Change the Stop Frequency to 10 GHz to see a comparable result. I get a peak of around 18p for the noise in the first file and about 23p in the second file.

Hope this helps.

[Dmitry_Fedotov]

Is so strange, I change frequency setup, please see attachment. Analyze in frequency domain from 1Hz up to 1MHz. My interest to region of  frecuency from 1kHz to 10kHz. As you could see from these pictures, the value of noise on sigle OpAmp Vn=8.3nV @10KHz that matched to datasheet. In the case of two OpAmp this value change to 12nV. we achive +3dB in noise, but it's only half of circuit. 

How to analyze total noise factor in differintial circuit when two points for input and two outputs?

[Tayyab_R]

Hi,

Please open the Noise Analysis and click on Help. This explains basically the types of noise that Multisim can simulate. As far as the opAmps are concerned, the models for the components are made by the manufacturers to accurately simulate the part. The noise analysis for the part will also depend on how extensive the noise related information is for the part.

Hope this helps.

[Dmitry_Fedotov]

Thank you for support, but maybe I not explained exactly the core of issue. My problem is noise calculation in differencial circuits. As you could see in the circuit from attached file, Bode Plotter plugged to inputs V2 and V1 and outputs Probe1&3 of amplifier. Results of measure gain in band is correct. In the case of noise measure, I cant use same approach, because on the folder of noise analyze  possible to select only single pin for input and output that not allow to analyze distribution of noise in circuit.

[Tayyab_R]

Hi,

I don't think you will be able to do this using a single Noise analysis. You will need to run seperate analysis for both voltage sources. What you can try is using a LabVIEW instrument to do this. Please take a look in the link below:

https://decibel.ni.com/content/groups/multisim-custom-simulation-analyses-and-instruments-powered-by...

Particularly, take a look at this instrument:

https://decibel.ni.com/content/docs/DOC-21601

This is not doing exactly what you need but it gives you an idea of what can be done.

Hope this helps.