FFT Analysis¶

The FFT Analysis feature provides real-time frequency domain analysis of waveforms. This guide covers how to use FFT analysis in the GUI.

Overview¶

Fast Fourier Transform (FFT) converts time-domain signals to frequency domain, revealing:

Frequency components
Harmonic content
Noise characteristics
Spectral purity
Modulation analysis

Key Features¶

Real-Time FFT: Updates with live view
Multiple Windows: Hanning, Hamming, Blackman, etc.
Peak Detection: Automatic peak finding
Dual Display: Time and frequency domain together
Export: Save FFT data and plots
Customizable: Scale, units, and display options

Opening FFT Analysis¶

Using Control Panel¶

Click FFT tab in left control panel
FFT panel opens
Configure settings
Enable FFT display

Quick Access¶

Toolbar → FFT button
Or View menu → FFT Analysis
Or Ctrl+F keyboard shortcut

FFT Panel Controls¶

Source Channel Selection¶

Select Input Channel:

Dropdown menu: C1, C2, C3, C4
Choose channel to analyze
Can also select Math channels

Requirements:

Channel must be enabled
Waveform must be captured
Sufficient data points

Window Function¶

Purpose: Reduces spectral leakage

Available Windows:

Rectangular: No windowing, best resolution
Hanning: General purpose, good for most signals
Hamming: Similar to Hanning
Blackman: Low sidelobes, good for weak signals
Bartlett: Triangular window
Flat-top: Best for amplitude accuracy

Selection:

Window dropdown → Choose function

When to Use:

Hanning: Default, good all-around
Blackman: Finding weak signals near strong ones
Flat-top: Accurate amplitude measurements
Rectangular: Maximum frequency resolution

Display Options¶

Magnitude Scale:

dB (Decibels): Logarithmic scale
Linear: Linear amplitude scale

Toggle:

☐ dB Scale
☑ Linear Scale

Frequency Scale:

Linear: Equal frequency spacing
Logarithmic: Equal spacing per decade

Phase Display:

☑ Show Phase

Displays phase spectrum
Shows phase angle vs frequency
Useful for filter analysis

FFT Display¶

Frequency Spectrum View¶

Main Display:

X-axis: Frequency (Hz, kHz, MHz)
Y-axis: Magnitude (dB or linear)
Plot shows spectral content

Features:

Grid overlay
Peak markers
Cursor measurements
Zoom and pan

Dual View Mode¶

Show both time and frequency domain:

Layout:

┌─────────────────┬─────────────────┐
│  Time Domain    │  Frequency      │
│  Waveform       │  Spectrum (FFT) │
└─────────────────┴─────────────────┘

Enable:

View menu → Dual View
Or checkbox in FFT panel

Split View Options¶

Vertical Split:

Time domain on left
FFT on right
Wide format

Horizontal Split:

Time domain on top
FFT on bottom
Tall format

Peak Detection¶

Automatic Peak Finding¶

Enable:

☑ Enable Peak Detection

Settings:

Number of peaks: 1-10
Minimum height threshold
Minimum peak separation

Peak Display¶

Peak Table:

Peak | Frequency    | Magnitude
-----|--------------|----------
1    | 1.000 kHz    | -10.5 dB
2    | 2.000 kHz    | -25.3 dB
3    | 3.000 kHz    | -35.1 dB

On Plot:

Markers at peak frequencies
Labels showing values
Color-coded

Peak Analysis¶

Fundamental Frequency:

Highest peak (usually)
Base frequency of signal

Harmonics:

Peaks at integer multiples
2f, 3f, 4f, etc.
Indicates distortion

THD (Total Harmonic Distortion):

THD = √(H2² + H3² + H4² + ...) / H1

Where:

H1 = fundamental
H2, H3, etc. = harmonics

Live FFT¶

Real-Time Analysis¶

Enable Live View + FFT:

Start Live View (Ctrl+L)
Open FFT tab
FFT updates in real-time
Watch spectral changes

Frame Rate:

Depends on FFT size
Typical: 10-100 updates/sec
Balances speed vs resolution

Dynamic Range¶

Auto-Scaling:

☑ Auto Scale Magnitude

Automatically adjusts Y-axis
Keeps peaks visible
Adapts to signal changes

Manual Range:

Set min/max magnitude
Fixed scale for comparison
Better for monitoring

FFT Settings¶

FFT Size¶

Number of Points:

256, 512, 1024, 2048, 4096, 8192
More points = better frequency resolution
Fewer points = faster updates

Trade-offs:

Larger FFT:
  ✓ Better frequency resolution
  ✗ Slower computation
  ✗ Less time resolution

Smaller FFT:
  ✓ Faster updates
  ✓ Better time resolution
  ✗ Coarser frequency resolution

Frequency Range¶

Nyquist Limit:

Max frequency = Sample Rate / 2

Example:

Sample rate: 1 GSa/s
Max frequency: 500 MHz

Display Range:

Full range: DC to Nyquist
Custom range: Zoom to region of interest

Averaging¶

Enable Averaging:

☑ Average FFT
Number of averages: 10

Benefits:

Reduces noise
Smooths spectrum
Better for weak signals

Types:

Linear: Simple average
Exponential: Weighted average, more recent data

Measurement Cursors¶

Frequency Cursors¶

Add Cursors:

Right-click FFT display
Select "Add Frequency Cursor"
Or use cursor panel

Features:

Measure specific frequencies
Read magnitude at cursor
Delta between cursors

Example:

Cursor 1: 1.000 kHz, -12.5 dB
Cursor 2: 2.000 kHz, -28.3 dB
Δf: 1.000 kHz
ΔMag: -15.8 dB

Peak Markers¶

Auto-Markers:

Automatically placed at peaks
Track peak movement
Show frequency and level

Manual Markers:

Place at any frequency
Measure harmonics
Compare levels

Exporting FFT Data¶

Export Spectrum¶

File Formats:

CSV: Frequency, magnitude, phase
NPZ: NumPy format with metadata
MAT: MATLAB format

To Export:

FFT tab → Export button
Choose format
Select file location
Save

CSV Format:

Frequency (Hz), Magnitude (dB), Phase (rad)
0.00, -80.5, 0.0
100.00, -65.2, 0.52
200.00, -55.1, 1.04
...

Export Plot¶

Save Image:

PNG, PDF, SVG formats
High resolution
Include or exclude annotations

To Export:

Right-click FFT display
"Export Image"
Choose format and settings
Save

Applications¶

Frequency Measurement¶

Measure Signal Frequency:

Capture waveform
Enable FFT
Enable peak detection
Read fundamental frequency

Accuracy:

Limited by FFT resolution
Resolution = Sample Rate / FFT Size
Example: 1 GSa/s / 8192 = 122 kHz resolution

Harmonic Analysis¶

Identify Harmonics:

Look for peaks at multiples of fundamental
Measure harmonic levels
Calculate THD

Example:

Fundamental (f₀): 1.00 kHz @ -10 dB
2nd Harmonic (2f₀): 2.00 kHz @ -30 dB
3rd Harmonic (3f₀): 3.00 kHz @ -40 dB

Noise Analysis¶

Measure Noise Floor:

Capture signal
Enable FFT
Look at regions without peaks
Noise floor = baseline level

Signal-to-Noise Ratio:

SNR = Peak Level - Noise Floor

Example:

Peak: -10 dB
Noise floor: -70 dB
SNR: 60 dB

Modulation Analysis¶

AM (Amplitude Modulation):

Carrier frequency peak
Sidebands at carrier ± modulation frequency

FM (Frequency Modulation):

Carrier and sidebands
Sideband spacing = modulation frequency
Bessel function pattern

Advanced FFT Features¶

Spectrogram View¶

Time-Frequency Display:

Shows how spectrum changes over time
Color map: frequency vs time vs magnitude
Good for transient signals

Enable:

FFT tab → Display Mode → Spectrogram

Features:

Time on X-axis
Frequency on Y-axis
Color shows magnitude
Waterfall effect

Power Spectral Density¶

PSD Mode:

Power per frequency bin
Units: V²/Hz or dBm/Hz
Better for noise analysis

Enable:

FFT tab → Mode → PSD

Overlap Processing¶

Windowed Overlap:

Consecutive FFTs overlap
Smoother spectrum
Better time resolution

Settings:

0%: No overlap
50%: Common choice
75%: Maximum smoothing

Troubleshooting¶

No FFT Display¶

Problem: FFT panel empty

Solutions:

Ensure channel is enabled
Capture waveform first
Check FFT is enabled (checkbox)
Verify sufficient data points

Noisy Spectrum¶

Problem: Spectrum very noisy

Solutions:

Enable averaging
Increase number of averages
Use Blackman window
Check for electromagnetic interference

Poor Frequency Resolution¶

Problem: Can't resolve close frequencies

Solutions:

Increase FFT size (e.g., 8192 points)
Use longer capture time
Use rectangular window for best resolution
Reduce frequency span (zoom in)

Spectral Leakage¶

Problem: Energy spreading to adjacent bins

Solutions:

Use windowing function (Hanning, Blackman)
Increase FFT size
Adjust sample rate for integer periods
Use flat-top window for amplitude accuracy

Tips and Best Practices¶

Window Selection¶

!!! tip "Choosing Windows" - General use: Hanning window - Amplitude accuracy: Flat-top window - Weak signals: Blackman window - Best resolution: Rectangular (no window)

FFT Size¶

!!! tip "Optimizing FFT Size" - Larger size: Better frequency resolution - Smaller size: Faster updates, better for live view - Common sizes: 1024, 2048, 4096 - Use power of 2 for fastest computation

Averaging¶

!!! tip "Using Averaging" - Always average for noise reduction - 10-100 averages typical - More averages = smoother spectrum - Exponential averaging for tracking changes

Measurement¶

!!! tip "Accurate Measurements" - Use peak detection for automatic measurement - Cursors for manual measurement - Enable averaging for stability - Check Nyquist limit (max freq = sample rate / 2)

Keyboard Shortcuts¶

Shortcut	Action
`Ctrl+F`	Open FFT panel
`Ctrl+P`	Toggle peak detection
`Ctrl+A`	Toggle averaging
`Ctrl+D`	Toggle dual view
`Ctrl+E`	Export FFT data

Example Workflows¶

Example 1: Measure Fundamental Frequency¶

Objective: Accurately measure signal frequency

Steps:
1. Capture waveform (Ctrl+C)
2. Open FFT tab
3. Select source channel
4. Enable peak detection
5. Read fundamental frequency from peak table

Result: f₀ = 1.0234 kHz

Example 2: Analyze Harmonics¶

Objective: Measure harmonic distortion

Steps:
1. Capture waveform
2. Enable FFT with Hanning window
3. Enable peak detection (find 5 peaks)
4. Identify fundamental and harmonics
5. Calculate THD from harmonic levels

Result:
  f₀ = 1.00 kHz @ -10 dB (fundamental)
  2f₀ = 2.00 kHz @ -35 dB (2nd harmonic)
  3f₀ = 3.00 kHz @ -45 dB (3rd harmonic)
  THD = 2.8%

Example 3: Find Interference¶

Objective: Identify unwanted frequency components

Steps:
1. Capture signal in FFT mode
2. Use Blackman window (good for weak signals)
3. Enable averaging (100 averages)
4. Look for unexpected peaks
5. Use cursors to measure interference frequency

Result: Interference at 50 Hz (power line)

Next Steps¶

Visual Measurements - Measure FFT peaks with markers
User Guide: Advanced Features - Programmatic FFT analysis
Interface Guide - Learn all GUI controls
Live View - Real-time FFT updates