Distortion with 3rd Octave Noise

Introduction

This sequence demonstrates a procedure to measure Total Harmonic Distortion (THD) using a band-limited pseudo-random noise stimulus.  This approach can be very useful when testing modern digital devices that do not efficiently transmit sine waves as a source signal, especially in telephony.  The general procedure for this test is outlined in IEEE 269, Annex J, part J.3. 

The sequence measures THD at a single frequency only.  This particular demonstration has been set to test at 315 Hz.  At the beginning of the sequence the user is prompted to enter a test level in Volts.  A white noise stimulus with a 1/3rd octave width centered around 315 Hz is played through the speaker and recorded through the reference microphone.  A full bandwidth spectrum with 1 Hz resolution is then analyzed.  Three power sums are calculated around the fundamental (315 Hz), the 2nd harmonic (630 Hz), and the 3rd harmonic (945 Hz).  These values are used to calculate the individual 2nd harmonic distortion, 3rd harmonic distortion, and the total harmonic distortion, which are displayed in terms of percentage. 

 

loudspeaker splice sequence

Final Display for Distortion 3rd Octave Noise sequence

 

Two other values are generated by this sequence: SDR (signal to distortion ratio) and SDNR (signal to distortion plus noise ratio).  SDR is the ratio of the fundamental power to the total distortion power.  For SDNR the total noise plus distortion must first be calculated.  This is done by applying an A-weighting to the spectrum and then applying a notch filter around the stimulus frequency range.  The remaining power spectrum after removing the stimulus is the total noise plus distortion spectrum.  This power spectrum is summed and the resulting power value divided from the stimulus band’s level, and the result is SDNR.  Both SDR and SDNR are displayed in dB.

Although there are limit steps at the end of the sequence, they have all been set at 999 or -999.  Limits for these metrics will vary quite a bit depending on the application, so the values have been left for the user to decide.

 

 

Required Equipment


Hardware

  • SCM-2 measurement microphone PN 4004 or equivalent
  • Listen AmpConnect PN 4040
    OR
  • SoundConnect Microphone power supply PN 4040 and
  • Power amplifier (Crown D45 PN 5600)

Software

  • SoundCheck 9.1 or higher
  • Limits PN 2000
  • Post-Processing PN 2004

*Hardware requirements may vary if testing telephones

 

 

System diagram

 

loudspeaker splice sequence setup

Setup & Calibration

  • Calibrate the amplifier as instructed in the SoundCheck manual.
  • Calibrate the reference microphone or ear simulator as instructed in the SoundCheck manual.
  • Connect the channel 1 (or Left) output of your soundcard to the input of the amplifier.
  • Connect the output of the amplifier to your loudspeaker. 
  • Connect the reference microphone or ear simulator to the microphone power supply.
  • Connect the output of the mic supply to the channel 1 (or Left) input of your soundcard.

You are ready to start the sequence.

 

Further sequence development

Ways in which you could modify or further develop the sequence include:

  • The sequence could certainly be expanded to measure at a number of frequency points.  Doing so would likely result in a rather long sequence with a large number of steps.
  • The use of the equation editor (Module 2012) would allow many of the post processing steps to be combined into a single step, greatly simplifying the overall sequence.  This would also make adding additional frequencies much easier.

 

Download Sequence Documentation (PDF)

Download Sequence (Zip file)

To run the sequence in SoundCheck, please save the zip file to your hard drive, then follow the instructions in "Sequence Editor: Exporting/Importing a Sequence' section of the manual. Download Manual.