Fast Fourier Transform (FFT) Analysis will cause DR. SPICE to perform a FFT against the current transient simulation results.
| Entry | Description |
| Number of Points | The number of points to sample each waveform. This is typically specified as a power of 2. |
| Output Variables | A FFT is performed against the variables listed. By checking the All Variables box, a FFT is performed against all output variables in the transient simulation. |
| Time Interval | The FFT is performed on only the time interval as specified by From and To. |
| Title | Tile which will appear within the Simulation menu. |
The accuracy of the FFT is a function of the number of data points and cycles within the waveform of interest. As a simple example, consider fft.cir found under the examples directory:
Simple FFT test circuit v1 1 0 sin(0 1 1meg) r1 1 0 1 .tran 0.01u 1u .end
Run 3 transient simulations of fft.cir while varying these simulation controls:
| Run Number | Final Analysis Time | Maximum Time Step |
| 1 | 1us | Default |
| 2 | 10us | Default |
| 3 | 10us | 5ns |
Against the results of each of the 3 simulation, perform an FFT against the V(1) node voltage. The FFTs of simulations 2 and 3 are better representations of a FFT as compared to the FFT from simulation 1. This is due to the fact that V(1) has more cycles in simulations 2 and 3. The FFT of V(1) from simulation 3 is a better representation of a FFT as compared to the FFT from simulation 2. This is due to the fact that the maximum time step is smaller in simulation 3, so there are more data points.
FFT Analysis Results