Negative values with impedance tube

[jmrplens]

Originally posted by @mwp4444 in #1 (comment)

I am a Ph.D. student in mechanical engineering. We are going to measure noise reduction based on aerogel developed by us during nanocomposite material research. I feel a lot of gratitude for A-Lab. Thanks to this, it seems that noise reduction research can be conducted. I have an inquiry during the measurement.

Based on the impedance tube, sound absorption was measured, and the measured value was negative. The sample file you attached also has a negative number.
It is known that a value between 0 and 1 should be obtained for the sound options coefficient value. (Refer to the paper)
Matlab determines that modification is necessary mathematically, and I am also looking at the code from the beginning.
Please tell me if I'm going in the wrong direction.
Measurement_of_the_Acoustic_Absorption_Coefficient.pdf

[jmrplens]

Hello,
Sorry that it took so long to respond, I have also seen the email you sent me.

The impedance tube method greatly affects small imperfections in the tube or positions. I don't know exactly what could be happening, the possible problems could be:

1. Air leaks in tube. Once the sample has been placed and the tube closed, the microphones are placed and the entire interior of the tube must be completely sealed. Any leak, no matter how small, can cause errors. Note: If the tube is hermetically sealed properly, the microphones should be placed last, otherwise the change in tube pressure can damage the microphone membranes.
2. The source is emitting at a higher level than supported, generating distortion in the emitted sound.
3. The microphones have a very high gain and clipping on the sound card and / or the computer.
4. The positions of the sample or the microphones is not quite correct.
5. The sample must be parallel to the rigid end of the tube and perpendicular to the acoustic field.
6. The tube is not rigid enough and vibrates when the measurement is made.

I have reviewed the measurements that you have sent me to the email, you can see in the impulse response that there is a lot of noise, it is quite likely that the system is not receiving the signal correctly. You have to check the input and output channels, confirm that the signal is received and also that it is received without clipping.

Your impulse response:

Usual impulse response (It does not have the latency correction applied, it does not matter, what you should compare are the waveforms):