Audio System Evaluation with Music Signals Stefan Irrgang, Wolfgang Klippel GmbH Audio System Evaluation with Music Signals, 1 Motivation Field rejects are $$$ Reproduce + analyse the problem before repair Audio System Evaluation with Music Signals, 2
Audio System Evaluation over the product life cycle Standard Measurement Condition Production Product Specification Development (components, system) Physical Evaluation EOL Testing Definition Target Performance Physical + Perceptual Service Assessment Target Application Condition Field Monitoring Audio System Evaluation with Music Signals, 3 After Sales / Field Measurements ~ audio signal amplifier u, i u, i DSP amplifier woofer microfone parasitic vibration internal sound sources Acoustical measurements using (available) microphones reveal - Frequency Response, SPL - Overall transfer function - Coherence - Residual ambient noise Objectives - evaluation of the performance as seen by the end user - long term testing under real condition - Prevention of a total failure - Service for customer complaint Challenges - audio signal used - influence of other sound sources - use of compromised test elements - measurement shall be realized at low cost Electrical measurements based on voltage and current monitoring (control technology) reveal - Voice coil and magnet temperature - Peak displacement and offset in coil rest position - Change of suspension stiffness indicating fatigue - anticipation of a motor or suspension defect Audio System Evaluation with Music Signals, 4
What is a critical defect? Related to customer complaints Observable in in-situ condition Impulsive distortion (panel buzzing, loose particles, loose electrical connection) Significant air noise caused by a leakage of the enclosure (Subwoofer) Excessive nonlinear distortion caused by motor instability and severe asymmetries Audio System Evaluation with Music Signals, 5 Reproduced Sound Quality Generation of Signal in an Audio System Stimulus Measured Signal Input Signal u(t) p(t) Output Signal Desired Small Signal Performance Model H(s)-1 dl(t) n(t) Noise Large Signal Performance (motor, suspension) Nonlinear Model dn(t) Undesired Defects Rubbing coil, Buzzing parts Wire beat Loose particles, air leak noise Parasitic vibration of other components Nonlinear Irregular Residual Audio System Evaluation with Music Signals, 6
Properties of Music Stimulus Dense stimulus / Most complex Non persistant excitation Non stationary excitation (Unknown time structure) Defects occur quasi randomly How to separate defect from accepted response? Audio System Evaluation with Music Signals, 7 System Identification: Correlation Technique Requires Stationary Signals H( f, n) x, y x, x Incoherence: Power metric Good for Noise problems Moderate for regular non-linearities Poor for time varying TRF Poor for impulsive distortion x, x State of the art Correlation Technique x, y % y, y Stimulus X(f,n) HW FFT ()* x(t) Sxx(f,n) MA DUT MA Sxy(f,n) xx ( f, n) ˆ ( f, n) S xy Y*(f,n) Calculation IC(f,n) H(f,n) y(t) MA ()* Syy(f,n) ˆ ( f, n) S yy HW FFT Y(f,n) IC db log 1 db % Audio System Evaluation with Music Signals, 8
IC [%] State of the art Correlation Technique Incoherence: Requires long (statistical!) time for accurate estimation (see paper) for low energy symptoms. IC( f ) % e IC ( f ) IC( f ) K K T tot T 2 T is defined by required resolution (Woofer 1 Hz / Midrange 5 Hz) K (number of processed blocks) = f(error) Example: df = 1Hz, total time =5s error: 30%, Incoherence too low by 1.5dB Assumption: stationary excitation When using non-stationary signals: Symptoms may be not stable! No Auralization of separated distortion (power measure!) Audio System Evaluation with Music Signals, 9 State of the art Correlation Technique / Measurement Results Normal Setup Incoherenc e Defect Setup K LIP PE L 1 2 3 4 from regular nonlinearities mask symptoms from rattling Incoherence measurement is not sensitive for impulsive distortion Audio System Evaluation with Music Signals,
Acoustical In-Situ Measurement Properties: Adaptive identification Fast initial learning Copes with non stationary input Follows changes of linear system Improved separation Typical ANC method Audio Source DUT Adaptive Modelling p - p lin Adaptive Identification of System Physical Characteristics Audio System Evaluation with Music Signals, 11 Compression of SPL Output 130 125 db 120 115 1 5 95 90 85 80 Sound Pressure Response Short Term Response (1 s) Long Term Response (1 min) 20 50 200 500 2k Short term response measured within 1 s (without voice coil heating) Long term response measured after applying the sinusoidal stimulus for 1 min Audio System Evaluation with Music Signals, 12
Kms [N/mm] Delta Tv [K] P [W] Influence of Ambient Conditions 150 125 Delta Tv Tambient P real P Re Pnom 12 11 9 8 75 50 25 0 Winter Sommer 4 3 Ambient temperature 2 1 0 0 500 0 1500 2000 2500 3000 3500 t [sec] 7 6 5 Influence to the Loudspeaker System Alignment 80 C resonance frequency fs (t) at rest position X=0 170 fs (X=0) 160 Hz 150 140 130 120 1 90 80 70 60 More than one octave shift -30 C 50 0 500 0 1500 2000 2500 3000 3500 t [sec] Properties of the mechanical suspension depend on humidity, temperature voice coil displacement cannot be predicted by time-invariant parameters adaptive learning process required Audio System Evaluation with Music Signals, 13 Acoustical In-Situ Measurement Residual Properties: Model identifies linear system only Separation of residual from linear distortion Exploit Residual: Non-linear Defect Noise Audio Source DUT Adaptive Modelling - p p lin d r Residual Physical Characteristics Audio System Evaluation with Music Signals, 14
Bl [N/A] PDF [1/mm] Assessing Regular Nonlinear and Irregular Proabilitiy densitiy function (pdf) Impact on sound quality: Parasitic resonances are excited by sufficient (kinetic) energy Defect occurs rarely Depends on Level (not highest!) Phase relationship require high state (e.g. Bl(x)) More comprehensive then sine sweep! 1.25 1.00 0.75 0.50 0.25 0.00 3.5 3.0 2.5 2.0 1.5 1.0 0.5 sine tone music -3-2 -1 0 1 2 3 Displacement [mm] Nonlinear Force Factor 0.0-5 -4-3 -2-1 -0 1 2 3 4 5 << Coil in X [mm] coil out >> Audio System Evaluation with Music Signals, 15 Audio system defect: Parasitic Vibration problem Most defects behave as a nonlinear oscillator active above a critical amplitude new mode of vibration powered and synchronized by stimulus Externally excited mass parasitic resonator spring Loose joint (Nonlinearity) vibration distortion signal one period time Audio System Evaluation with Music Signals, 16
Freq. [Hz] 4 3 2 Mic. Voltage [V] V Freq. [Hz] 4 3 2 Residuum of the Modeling (defect in car) tambourine 0. measured signal p(t) 0.05 0.00-0.05 residuum e(t) +db (!) -0. modelled signal plin(t) 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Time [s] next Audio System Evaluation with Music Signals, 17 Time Frequency Analysis of the Residuum residuum in time domain Defective car 0.02 Normal car (without tamborine) 0.01 0.00 defective car (with tambourine) -0.01-0.02 Normal car 60 62 64 66 68 70 Time [ms] dbfs dbfs -38 k -38 k Defect -38 < 1k -71-71 -74-74 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Time [s] 0 0.5 Time [s] 1.0 1.5-38 < 1k 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Time [s] 0 0.5 Time [s] 1.0 1.5 next Audio System Evaluation with Music Signals, 18
Freq. [Hz] 4 3 2 Sound Pressure [db] Freq. [Hz] 4 3 2 Standard EOL test of defect Using log sweep 1 K LIP PE L Normal car (without tamborine) 90 80 70 Defect Fundamental defective car (with tambourine) 60 50 40 Pass/Fail Limit 30 Rub&Buzz 1 2 3 4 dbfs -38 k dbfs -38 k 1k 1k -71-74 -71-74 -77-80 -83-86 -77-80 -83-86 -89-92 -89-92 -95 0 Time 0.5[s] 1.0 0.1 0.2 0.3 0.4 0.6 0.7 0.8 0.9 Time [s] -95 0 Time 0.5[s] 0.8 0.9 Time [s] 1.0 0.1 0.2 0.3 0.4 0.6 0.7 Audio System Evaluation with Music Signals, 19 Acoustic In-Situ Measurements Auralization p Adaptive Identification provides: DUT Auralization Separated linear response Purified from time variant linear distortion Residual to be scaled up and down Benefit: Objective Criteria Rate importance Is this defect acceptable? Action required? Audio Source Time- Variant Adaptive Modelling Residual Physical Physical Characteristics Characteristics - d r S dis p lin p A Adaptive Identifcation of slowly changing linear system Perceptional Modeling Listening Test Perceptual Characteristics Audio System Evaluation with Music Signals, 20
db rel. Finding most critical section ratio Challenge: Find highest distortion in a long recording Non-Stationary excitation 5 0-5 - peak distortion RMS Solution: Check Crest factor of Residual: High Peak and Low Mean distortion indicate impulsive distortion (defect symptom) -15-20 -25-30 mean distortion 25 50 75 125 150 Time s Select Best Section Audio System Evaluation with Music Signals, 21 Summary Simple Setup Wave file / Web based solutions Comprehensive System Evaluation Response + Residual distortion Using any stimulus Customer specific audio material Fast Identification of defects Automatic selection of critical section Time Domain Analysis Auralization Audio System Evaluation with Music Signals, 22