Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.
I don't use LTspice but if you run a noise analysis you should then be able to probe devices to find their contribution to the output noise. That way you can track down the main noise generating components.
Re: Need help to trace the voltage noise density with LTspic
Hello
here is the snap shot of the design stage and the noise ...
I don't see 1/f noise and the broadband noise. Do you have an idea ? I probably make a mistake on the spice command line.
Be careful about using zero for the input voltage - it will give erroneous results. I would suggest you use a small DC input voltage.
The real problem is that I don't think the LT1012 model doesn't model the noise correctly. Try changing the model to the AD743 and see what happens (you will need to change the biasing to ensure it is within the working range of the opamp). It should look more like what you expect, including 1/f noise.
You might find it easier to use a unity gain follower to avoid DC problems (leave off R2).
Re: Need help to trace the voltage noise density with LTspic
Hello
Thanks for your help, I tried whith the AD742, it doesn't seem to be good (2.9nV/sqrt(Hz) @ 10kHz specified in the datasheet).
I'm not sure I use correctly the spice command...
Re: Need help to trace the voltage noise density with LTspic
Hello
well you're right, the characteritic is better but not exactly the same into the datasheet for the same gain (=1).
I'm wondering if the simulation is relevant
what is your opinion ?
thanks a lot for your help
**broken link removed**[/img]
PS : the noise is about 4nV/sqrt(Hz) with a shunt for the feedback.
You should remove the 1k resistor - it will be increasing the noise. Use a short circuit. It should then be pretty close to the data sheet, if not exactly the same values.
My simulation gives 3.23nV/rt(Hz) at 10kHz. I think the other discrepancies are probably that the model isn't accurate. This is not uncommon. Some opamp models don't include noise at all. Some have huge amounts of noise (because they don't attempt to model noise correctly). Sometimes the header of the model will say which features of the opamp are accurately modelled.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
