Stats

  • Locked Locked
  • Replies 1
  • Subscribers 125
  • Views 18327
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

noise analysis in cadence

srihari18

Hi i am trying to do noise analysis for a four-quadrant multiplier with two inputs i1 and i2. The output is obtained by driving it through load resistor of 1K. Now when i do noise analysis both the input and output noise are the same with no change. The steps i did were
1) Chose noise in analysis in EDA and set the frequency range. For input noise i chose the option "current" since the inputs are of the current form and chose one of the inputs (i2).

2) For output noise i only had the option to choose between "probe" and "voltage". Didn't select voltage becoz the multiplier output is of the current form. So i chose probe and selected the resistor. The input and output noise is same in waveform except the y axis of input noise is in nano scale while the output noise is of the form pA/sqrt(Hz). Can anyone kindly tell me what exactly is the unit of noise say for current its nA,pA etc. Similarly what is the unit for noise??? 

3) Also in "edit object properties" i have for the resistor component to choose "generate noise" yes or no. I selected the option no. Is my procedure right.

But when i choose i1 as input noise i can notice some change in waveform between input and output noise.  Can anyone please help what is the correct way to do it. Because my lab mate showed his noise analysis waveform wer i saw the output waveform was reduced in value in comparison with the input noise. Output noise was suppressed.

The above is for noise vs frequency. Can anyone please tell me how i can plot output noise vs (one of the input current). Plssss help!!!

Parents
  • Andrew Beckett

    The noise units (both input and output referred noise) will be in A/sqrt(Hz) or V/sqrt(Hz) or maybe A^2/Hz or V^2/Hz (depending on how you plot it it). The simulator only really computes output noise, and whether it is a current or a voltage noise density depends on how you measure it. The input referred noise is just the output referred noise divided by the gain from the input to the output.

    So when you are plotting the input noise you're not plotting the input current - you're plotting the input-referred noise (so this is as if all the noise was at the input of the circuit, and the circuit was noiseless) - it's a useful metric for comparing blocks.

    If you want to see how the noise varies with different DC input currents, you could choose to do the noise analysis versus the DC current rather than frequency (e.g. set the sweep variable to be component parameter, pick the current source at the input, and the "dc" parameter as the parameter to sweep). Then the noise will be computed at a single frequency.

    Whether this is what you want depends on what  you're trying to analyse. If you are using the multiplier as a frequency multiplier, you may need to use hb or pss together with hbnoise or pnoise to measure the time-averaged noise.

    Also, whether the load should be treated as noiseless also depends on what you're actually trying to measure.

    I suggest you understand (and explain) your goal first before trying to set up the simulator - unless you know what you're trying to measure, it's going to be hard to advise you. Perhaps you should consult your supervisor?

    Regards,

    Andrew

Reply
  • Andrew Beckett

    The noise units (both input and output referred noise) will be in A/sqrt(Hz) or V/sqrt(Hz) or maybe A^2/Hz or V^2/Hz (depending on how you plot it it). The simulator only really computes output noise, and whether it is a current or a voltage noise density depends on how you measure it. The input referred noise is just the output referred noise divided by the gain from the input to the output.

    So when you are plotting the input noise you're not plotting the input current - you're plotting the input-referred noise (so this is as if all the noise was at the input of the circuit, and the circuit was noiseless) - it's a useful metric for comparing blocks.

    If you want to see how the noise varies with different DC input currents, you could choose to do the noise analysis versus the DC current rather than frequency (e.g. set the sweep variable to be component parameter, pick the current source at the input, and the "dc" parameter as the parameter to sweep). Then the noise will be computed at a single frequency.

    Whether this is what you want depends on what  you're trying to analyse. If you are using the multiplier as a frequency multiplier, you may need to use hb or pss together with hbnoise or pnoise to measure the time-averaged noise.

    Also, whether the load should be treated as noiseless also depends on what you're actually trying to measure.

    I suggest you understand (and explain) your goal first before trying to set up the simulator - unless you know what you're trying to measure, it's going to be hard to advise you. Perhaps you should consult your supervisor?

    Regards,

    Andrew

Children
No Data

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.