Listen, Inc.

HelpSearchMembersCalendar

What is the best way to use Rub & Buzz Distortion analysis for tweeters?

It is important to look near the resonance of the driver. This frequency is where the maximum driver displacement occurs and is the most likely region for Rub & Buzz defects to be exposed.

Please download the following PDF for a detailed on the procedure.



...

Am I correct in assessing that there is no way to select a time window around the data when using Spectrum analysis?

My understanding is that the windowing avaiable only determines the size of the "slices" that the algorithm looks at. For instance, if I have a waveform that is 100mS long and choose a 20mS window, the algorithm will take 5 slices and run FFTs on each slice and...

I'm using the Multitone Stimulus and Analysis for just one test in my sequence. I never have a problem with the Frequency Response result but randomly I get really odd 'ringing' looking results in the NCD result. If I re-run the test it goes away.

My first thought is that it could be related to the stimulus randomization that occurs with each run. I attached the worse case...

In the actual version of SC, you have a choice of window for almost every algorithm (except Broadband RMS and LogTSR). This windows are used as weighting of the time data, prior to the FFT used in the algorithm.
In SC6.11, the choice is: None, Hanning, 4 term B-Harris, 7 term B-Harris, Flat-top.
Each of them is more precise than the previous in term of amplitude, but also has an...

The correction out compensates the output transducer, in addition to the stimulus equalization.
The correction in compensates the input transducer.
Both curves are specific to the channel in & out used and are part of the calibration curves in the memory list.
There are both applied by multiplication (gain dB & phase addition) on the Frequency Responses measured.