FLDIGI Users Manual  4.0
Spectrum Viewer

Table of Contents

Spectrum Scope Display

Spectrum Viewer Description

The Spectrum Viewer is an analysis tool used to evaluate received signals. It can be used to measure signal integrity, audio-codec noise floor, transceiver passband characteristics, Rx signal spectrum, etc. The time domain audio stream is converted to the frequency domain using a Fast Fourier Transform, FFT. The FFT uses a snapshot of 8192 audio samples obtained from a sliding window. At any instance in time the FFT is computed from the immediate past 8192 samples. A 8192 sample set in the time domain results in a 4096 set of complex values in the frequency domain.

The fldigi modems use a sampling rate of 8000, 11025, 12000 and 16000 samples per second. The respective FFT's then can display values up to 1/2 of the sample rate. The viewed frequency range is constrained to 0 to 4000 Hertz.

The display is signal magnitude in dB versus frequency.

Operating Controls

The Spectrum Viewer operating controls are used to set:


When the Spectrum Dialog first opens, or when the # scans is changed that control is displayed in the gray background color. When the scan count has reached or exceeds the # scans then that control changes to green as shown above.

The scope graticule is always 10 dB per division on the vertical scale. The horizontal scale is dependent on the the selected frequency range:

Signal Characteristics


Basic Signal Measurement

A measurement of amplitude and frequency is made by left clicking on the display with the center of the cross-hair cursor placed on the data point of interest. The FFT display can be paused to take a snapshot of the values.

The label will change to indicate the state of the FFT processing; "Running" or "Paused".

The above image shows a psk-31 idle (no modulation) test signal whose characteristics are

a signal that no one wants to put on the air, but is too often observed. The frequency / dB readout to the right of the "Paused" indicator was obtained by left clicking on the signal just to right of center. The 3rd order intermodulation frequency is clearly 20 dB below the fundamental phase modulation signal. A clean psk-31 should be better than minus 25 dB IMD. Modern transmitters can achieve IMD levels at or better than -30 dB when not overdriven. No ALC is the rule. Set the transmitter power level to full power and then reduce the audio drive to no more than 25% of full power. This will usually result in a PSK signal of which you can be proud.

Here is the same signal during modulation:


Differential Signal Measurement

You can measure the dB difference between two points:

The signal level and frequency differences will be shown in the right most top status control. Entry of points 1 and 2 need not be in order and either the #1 or #2 point can be selected multiple times to extract dB differences over a range of point pairs. Selecting measurement points will be easier if you pause the FFT display.


Agreement with main fldigi dialog

fldigi measures the S/N and IMD of a psk signal on it's main dialog. These measurements are made only during idle. During modulation the values are either blanked or dimmed at the user's discretion. This illustrates the test signal display both on the spectrum viewer and on the main waterfall.




Select whether the Signal Measurement points are shown with amplitude, frequency or both as linear graticules.

You can capture the current full FFT spectrum at any time, both when running or paused. Select the "Save to CSV" menu item and a comma separated value file will be save to the fldigi temp folder.

Close the spectrum viewer by either selecting the "Close" menu item or clicking the dialog decoration "X" button.

Filter Evaluation


This illustrates using the Spectrum Viewer to view and evaluate a transceiver filter. This is 40 meter noise on an IC-7200 with the filter set to 3000 Hertz.

CSV file usage

The comma separated value file can be opened using any spread sheet program. My personal preference is the program Gnumeric (available for Unix, Linux and Windows). The file contains columnar data for frequency and magnitude (linear). You can use the spread sheet program mathematics to generate a column of 20log10(mag) values.

Return to Top of Page
Return to Main Page