Recently, I designed a signal chain and calculated the noise, but met some problems:
The hand calculation result doesn’t match the simulation one, the detail is as attached;
How to calculate the equivalent noise resistor? When there are some capacitances in the circuit, how can I calculate the noise?
Could you please help me out? Thank you!
Noise calculation of OPA2333.docx
Hello Cera,
I have attached an updated TINA-TI schematic regarding the mismatch between the simulations and the datasheet.
By inserting the noise voltage source at the non-inverting input of the OPA2333, the square root sum of squares noise is added to Broadband noise of the amplifier. This simulated value should now match with the data sheet. Feel free to use this method in further calculations.
Please note that we are in the process of updating the simulation model to a more realistic model.
I also see a few errors in your total noise hand calculations and that could lead to the mismatch in your simulation results.
You made errors when:
I encourage you to look at the TI Precision Labs short answer exercises and segments on noise—specifically in sections 3 through 5 for calculations and section 2 for theory—that can be found here: www.ti.com/precisionlabs
I also encourage you to look at the presentation with a full noise calculation of an instrumentation amplifier linked here: /cfs-file/__key/communityserver-discussions-components-files/14/8637.Instrumentation-Amplifier-Noise-Analysis_2D00_clean.ppt
If you haven't already, download the Analog Engineer's Pocket Reference guide, it has all of the equations on noise: https://www.ti.com/seclit/sl/slyw038b/slyw038b.pdf
Finally, thermal noise on capacitors is called kT/C noise. If you have a resistor in your circuit, most of thermal noise created by the capacitor won’t matter compared to the resistor. So in this context, capacitors don’t contribute to the noise calculations unless there is a pronounced filtering effect like described in the TI PL Noise pt. 4 section.
However, there is a phenomenon called capacitor-switched noise that can lead significant noise added in your circuit. I invite you to look at this presentation: http://www.eecg.toronto.edu/~johns/ece1371/slides/ECE1371_NoiseSC-2.pdf , check out the applications of TI’s delta sigma group, and do more research on the topic if you are interested.
Best,
Cole Macias
P.S. I've also uploaded a PDF of the hand calculations, please take a look at them.
