Hi,
I'm trying to read 12 wave files recorded at 25kHz, perform scaling, FFT (peak hold) and then export the results to Excel (right clicking the chart and selecting is my current method). In doing so I'm having a strange hiccup that I'm having trouble overcoming. My "block size" is 3750 samples so that I get an acceptable frequency resolution (6.67Hz). However, the block size also affects the timing of the display loop. I can't seem to figure out how to separate these.
The attached file is a little needlessly complicated for the behavior I'm trying to illustrate. When you hit run, the user is prompted to select files which are named with the record start time and an identifier. I use search regular string to take the time stamp, convert it to seconds and then add dt*sample number to identify where in time the data being displayed is. This may be a hiccup for those trying to run the code without my big sound files.
The problem is it seems like I am dropping samples in the FFT analysis - the time display is jumping by multiples of the record time. I have figured this out by selecting 1 wave file, performing peak hold for a certain duration and exporting, then doing the same for 12 wave files. There is a noticeable difference in amplitude between the two methods even though the file being compared is the same. The answer may be to go back to single file analysis but I was hoping to save time by doing 12 at once. Thanks in advance for your help!
LabVIEW 2018 SP1, Sound and Vibration Toolkit, Windows 10 64bit.
