Transient analysis allows you to check the signal level over time like an oscilloscope. If you want to check the frequency components contained in the signal, it is convenient to use the FFT function. Signal distortion and noise components can be checked by simulation.

This time, I will introduce the FFT function.

If you are just starting LTspice, we recommend that you look at the "basics" from the list below.

Let's use LTspice series list is here


Also, if you would like to see a video on how to write a basic circuit and how to execute it, there is an on-demand seminar that does not require you to enter personal information, so please take a look if you are interested. Detailed information about the seminar is also provided to those who fill in the questionnaire.

LTspice On-Demand Seminar - Function check with RC circuit -

What is FFT?

FFT (Fast Fourier Transform) stands for Fast Fourier Transform.

Like a spectrum analyzer, SPICE calculates and displays the frequency components and levels (power) contained in the signal. The FFT function is a built-in function of Waveform Viewer because it performs calculations based on the data obtained by transient analysis (time axis).

Let's use the FFT function!

work procedure

This time, as an example, we will use transient analysis to check a signal waveform that combines three sine waves with different frequencies and magnitudes, and use the FFT display function to check the frequency distribution.

Article header library 128921 pic01 1
Figure 1: Circuit that produces a composite waveform of sine waves (TRAN analysis)

The transient analysis results are shown in Figure 2.
The frequency component and magnitude cannot be determined just by looking at the waveform of this OUT terminal.

Article header library 128921 pic02 1
Figure 2: Composite waveform of sine waves (OUT terminal)

So let's use the FFT display function.
Select View→FFT from the menu bar.

Article header library 128921 pic03 1
Figure 3: FFT display method 1

Then, a Box like the one in Fig. 4 will appear, so normally just press "OK".

Article header library 128921 pic04 1
Figure 4: FFT display method 2

The FFT display results are shown in Figure 5.

A logarithmic graph with signal level (dB) on the vertical axis and frequency (Hz) on the horizontal axis was obtained.
There are peaks at frequencies of 1 kHz, 3 kHz, and 10 kHz, and we were able to grasp the components and magnitude of the waveform.

Article header library 128921 pic05 1
Figure 5: FFT display result

Usage points

When using the FFT function, it is recommended to set the command “.options plotwinsize=0” to disable compression of analysis data. A low noise floor is obtained using this command (Figure 5).

Also, setting the transient command “Maximum Time Step” to less than 1/100th of the signal period will give good FFT results. However, if it is set short, the simulation time will take a little longer, so please make adjustments while looking at the balance.

For comparison, we removed the plotwinsize specification in .options and ran the simulation with the default Maximum Time Step. Compared to the previous results in Figure 5, the noise floor at the base is larger, and peaks are visible at unintended frequencies (Figure 6).

From this result, you can understand that the command “.options plotwinsize=0” is important for FFT analysis.

Article header library 128921 pic06 1
Figure 6: FFT display result (no plotwinsize specified, MaxiumTimeStep default setting)

Let's look at the output voltage signal of the power supply IC with FFT!

Previously, let's use LTspice - Confirming the operation of the DC-DC converter confirmed the ripple voltage level of the output voltage, so I would like to confirm the frequency component with the FFT function.

The circuit uses the LT8640 Demo file. Since the FFT analysis requires steady-state results, check the ripple voltage after power-up. Therefore, the simulation time for transient analysis is set to 500u~700usec.

Article header library 128921 pic07 1
Figure 7: Simulating output voltage frequency analysis using the LT8640

The output voltage ripple waveform is shown in Figure 8.

FFT analysis confirms that not only the 1MHz switching frequency component but also 2x, 3x, even and odd frequency components are included (Fig. 9).

Article header library 128921 pic08 1
Figure 8: Output voltage waveform (time axis)
Article header library 128921 pic09 1
Figure 9: Output voltage waveform (frequency axis)

Noise from power supply circuits may affect peripheral circuits and EMI tests.
By using the FFT function to understand the frequency components, noise countermeasures such as filter design can be studied through simulation.

Please take this opportunity to try the FFT function!

LTspice demo file verified this time

FFT_Simulation__1.zip

The two simulation files that were performed this time are stored. Please try!

At the end

This time, we introduced the FFT function!

If you haven't used LTspice yet, please download LTspice from the link below!
Please try once.

Download LTspice here

We also hold regular LTspice seminars for beginners. You can learn the basic operation of LTspice, so please participate.

Click here for LTspice seminar information

Click here for recommended articles/materials

List of articles: Let's use LTspice Series

LTspice FAQ:  FAQ list

List of technical articles:  technical articles

Manufacturer introduction page:  Analog Devices, Inc.  

Click here for recommended seminars/workshops

Click here for analog technology seminar information

Inquiry

If you have any questions regarding this article, please contact us below.

Analog Devices Manufacturer Information Top

If you want to return to Analog Devices Manufacturer Information Top, please click the button below.