Interference Images: Noise-Emission of a Car

View into the motor room

Let's observe now the noise emission of a car. It is hard to decide, which parts of aggregates excite large quantities of noise. The front-compartment of the car is opened. We expected to see the cylinder-head and the motor. Instead of, we see something different: maximum quantities of noise come as ground reflexions from the left bottom side of the car. We remember to find cars beeing very quiet in fresh snow! Note also the week reflexion at the front-compartment.

Measuring object and integrated noise emission with HIT (blue: max., yellow: min.), PSI-Tools algorithm EXP, contrast-factor 0.8 (see PSI-Help). Data: motor ca. 2000 rpm, PSI-PORTKONFIG=2189742931, hardware-filter (-2 dB): 2x TP 5000 Hz, 2x HP 3000 Hz, 16 channels, 20 kSps, calculation integral 1.2 kS = 2 revolutions, files: 2300_1k2.chl, 2300_lg1.ini, 2300_lg1.bmp, comments: 13_02_97.txt

a) Pseudo-3D-view and b) motorbloc in detail

a) A pseudo-3D-image gives an impression of fields with week interferences. Note the quiet field within motor-room and compartment. b) Surprising for us is the absence of any noise comming from the cylinder head themselves. The picture only partially shows the elbow in front of the cylinder head. It is possible, that this motor works incorrect. We find a hard, noisy knocking comming from left.



Measuring arrangement. The microphone array is 4x4 orthogonal. Microphon distance is 30 cm each; microphon's bottom line in 1,08 m hight, distance to mapping layer 1.75 m

---

Horizontal Views

For the following scenes the microphone array was standing on the bottom, planar to the car. The task again is, to find out locations of excitement. May be, this question can be important to construct more noiseless cars.



Frontal view. We find excitement locations as directed noise and as ground reflexion. Note, that the possibility to match photos against noise images in the minute is not more precise then +/-5 deg or some centimeters in a distance of some meters. Data: vorn.ini, vorn.bmp, 17_02_97.txt, 50kSps, 1000 Samples used, 2xHP 3kHz, distance 3 meter.



Side view. We find strongest noise emissions comming from exhaust pipe or the exhaust silencer. Data: links.ini, links.bmp, 17_02_97.txt, 50kSps, 1000 Samples used, HP 1kHz/-40dB, distance 5 meter.



... with another color map. Aliasing artefact (star glow effect) or reality?

Slow-motion movies



Movie with 1000 pictures per second (p/sec), integrated over 10 msec each, channel sample rate
50kSps, 16 channels, measuring distance 5 meters, 1276 rpm, idle motion/neutral gear,
color sequence to heigher emissions: gray-yellow-green-red-blue.
Find from left to right the pre-silencer, the catalyser, the final silencer
and the exhaust pipe as maximum emission regions.

1000 p/sec, integration over 10 msec, 47 images, file size 690k/1.6MB (*.avi)
1000 p/sec, integration over 10 msec; 171 images, file size 2.6/10 MB (*.avi)

Data: Channel Data File = ...\FIESTA\17_02_97\LINKS.CHL, Palette DEZENT.PAL, X1, Y1, Z1 = -1.5000, -0.5000, -5.0000, X2, Y2, Z2 = 3.0000, 1.5000, -5.0000, Medial Speed = 340.00 m/s, Algorithmus = 'Maximum Peak', Start sample = 0, End sample = 9000, Sample rate 50kSps, li50_500.ini



View under the car. We find the exciting place between exhaust pipe and the pre-silencer. May be it can be the leackage between the pre-silencers cover and the silencer themselve.

Page counting since january 21, 1997