What appears to be "clipping"

[drhurd]

Can someone please tell me why I appear to get "clipping" sometimes as shown by the red arrow in the photo?


Thanks
Doug

[steve]

Oh that's weird. And that's immediately after recording?
What are you recording, and how are you recording it?

Could you post a close-up (zoom in) of the clipped waveform.

[kozikowski]

I bet I know.

You have extreme DC offset somewhere in the analog audio chain before digitization. It's negative which means the negative going parts of the waveform will get damaged (clipped) way before the positive portions do.

Note that the damage isn't constant over the example. I bet there is an electrical component breaking down and allowing a DC voltage to leak into the sound. I personally would be dragging out my oscilloscope and voltmeters, but I suspect you need to replace parts of your analog audio system one at a time until the problem goes away. Nobody would be shocked to find out that your US-$13 sound card is failing.

Koz

[drhurd]

Thank you for the offset clue. Unfortunately my signal chain is a little longer than a "$13 sound card"




Thanks again
Doug

[kozikowski]

<<<Unfortunately my signal chain is a little longer than a "$13 sound card">>>

But somewhere, the sound had to go from analog to digital and that's where I would be looking. It's also possible in a very complex system, for a buss to be leaking or overloading sending distortion the rest of the way down the system.

Are you trying to use the digital out of the board? I would digitize a bit of nothing for a minute or so (sound channels open as normal, but nobody singing) and then select a short chunk in the middle and remove the DC offset. I think it's under Normalize > Remove DC Offset. If there is suddenly a really big baseline change in the selected piece compared to the two ends, that's your problem.

Koz

[drhurd]

I am using the analog matrixed outputs from the board into a seperate mixer so that I can mix the 8 channels down to two before hitting the input to the laptop. I will try the old "if in doubt, rip it out" process to try to identify the part of the chain where this is happening.

Thanks again for your help
Doug

[kozikowski]

Do you know how to drive a voltmeter? Find a voltmeter that will measure down to .01vdc and cram it across the sound lines just before they enter the laptop. I expect the needle (needle, they still use needles?) or the reading not to budge from 000.

Since you're using a Windows laptop, how sure are you that you have a "Line-In" connection? A great many portable PCs do not. Just turning down a "Mic-In" connection to accommodate the increased volume of a Line level signal will not do it. There's 60dB (1000 to 1) difference between them and there's an excellent chance of distortion even if it seems to be working.

You may have just found your problem.

Koz

[steve]

Thank you for the offset clue. Unfortunately my signal chain is a little longer than a "$13 sound card"




Thanks again
Doug

Are you taking the Mickey?
You're implying that you are recording through a professional mixing desk, a rack full of other gear, then you show us an image of clipped audio and ask "What's wrong?"
The problem is you need a sound engineer that knows their job.