.NOISE: Noise Analysis

General form:

.noise v(output <,ref>) src ( dec | lin | oct ) pts fstart fstop
+ <pts_per_summary>

Examples:

.noise v(5) VIN dec 10 1kHz 100MEG
.noise v(5,3) V1 oct 8 1.0 1.0e6 1

The .noise line does a noise analysis of the circuit. output is the node at which the total output noise is desired; if ref is specified, then the noise voltage v(output) - v(ref) is calculated. By default, ref is assumed to be ground. src is the name of an independent source to which input noise is referred. pts, fstart and fstop are .ac type parameters that specify the frequency range over which plots are desired. pts_per_summary is an optional integer; if specified, the noise contributions of each noise generator is produced every pts_per_summary frequency points. The .noise control line produces two plots, which can selected by setplot command:

  • one for the Voltage or Current Noise Spectral Density (in (\frac{V}{\sqrt{Hz}}) or (\frac{A}{\sqrt{Hz}}) respective the input is a voltage or current source) curves (e.g. after setplot noise1). There are two vectors over frequency:
    • onoise_spectrum: This is the output noise voltage or current divided by (\sqrt{Hz}).
    • inoise_spectrum: This the equivalent input noise = output noise divided by the gain of the circuit.
  • one for the Total Integrated Noise (in (V) or (A)) over the specified frequency range (e.g. after setplot noise2). There are two vectors which are in reality scalars:
    • onoise_total: This is the output noise voltage over the specified frequency range
    • inoise_total: This the equivalent input noise over the specified frequency range = output noise divided by the gain of the circuit.

The units of all result vectors can be changed by using control variable sqrnoise:

  • set sqrnoise: will deliver results in squared form, means the unit is (\frac{V^{2}}{Hz}) or (\frac{A^{2}}{Hz}) . This value refers more to the convinient Power Spectral Density.

Default setting of ngspice is unset sqrnoise, which delivers Voltage or Current Noise Spectral Density. This is more practical from designers point of view.