Hello, In this video we represent each component as a source of noise,but in real life the traces also play a role.
How can we simulate noise with the contribution of the traces connecting the components?
Thanks.

 

Noise is an unknown, which depends upon the location and length of the traces, and what EMI noise sources are nearby.
So you design with the sensitive node traces short and not near any traces with any high amplitude signals.

 

Hello,the first way is to do via shielding for eliminating noise from traces ?
2.
Also regarding trace length , there is this recommendation to make traces as close as possible to the opamp .because long trace create parasitic capacitance for high speed opamp . But I heard otherwise that long trace create inductance in signal integrity theory .so why long trace creates parasitic capacitance for input of an opamp?
Thanks.

 

Hello,the first way is to do via shielding for eliminating noise from traces ?
2.
Also regarding trace length , there is this recommendation to make traces as close as possible to the opamp .because long trace create parasitic capacitance for high speed opamp . But I heard otherwise that long trace create inductance in signal integrity theory .so why long trace creates parasitic capacitance for input of an opamp?
Thanks.

Long traces contribute BOTH!! They increase SERIES inductance and they increase SHUNT capacitance.
THAT is why the application literature always gives examples of best practice. They are aware of the problem.

 

Hello Mr. Bill , I can do simulations of noise and calculations .
But how do I see the source of the noise in real life ?

for example my power supply is a source of noise and it meat decoupling capacitor to eliminate noise from power supply .
How can I know that the decoupling network works ?
That is such a thing called EMI probe where .
If it’s a good idea how can I test my power supply noise before and after decoupling ?
Thanks .

 

The common way of checking for power supply noise is to use an oscilloscope with AC coupling and look for signals on top of the DC level. This may require providing a low noise resistor so that the supply is delivering current similar to the intended circuit requirement. Power supply noise is often dependent on the load current. After that level is known, a similar check with the actual load can be done to see if the actual load is contributing to the noise level.
To evaluate the actual noise contribution from individual components is quite tedious, as it would require individual testing of each component with a current similar to the current in the actual circuit. (That can be very tedious indeed.) What would remain is actual measurement of the op-amp noise, which many makers od devices claimed to be low noise will show the arrangement they use for that test.
So how low a noise level are you actually seeking to have??
My experience has not needed to go beyond common noise reduction precautions. Evidently I have been lucky in that respect.