Topic: Impact of marginal batteries?

[loughney]

Back in November, I taped a show and when I opened the file at home, I noticed that the wave form started to become unequal, see attached image, with the red box indicating the area that I am talking about.  On Jan 1st, I taped the same band, same venue, same recording location and didn't have the issue.  I did change my battery box battery, and now just checked the old battery and the battery meter shows that the power level is marginal.  I'm assuming that is the source of the problem.  Anyone else have this type of experience or know if my hunch is the right one?

[vanark]

Yep, definitely could be due to a low battery issue.

[nulldogmas]

I have absolutely had that happen to me — sometimes one channel gradually drops, then the other.

[fireonshakedwnstreet]

Looks right. My MixPre would have a lower signal when the batteries were giving out.

[morst]

Sure thing. Used the Beyer MV100 for many years on two alkaline 9-volt cells, and after about 5 hours powering the KM140 pair, this was the risk.
I never made it past approximately 5.5 hours with useful output, unless I changed the batteries.
As the voltage drops, the gain gets lower and lower until you can't hear the signal from the noise!
Note also on your graphic that the positive excursions of peaks are less affected by the loss than the negative excursions.
The MV100 did this as voltage dropped.

[DSatz]

Whenever anything affects the positive-going part of a waveform differently from the negative-going part, that's non-linear distortion (in addition to the change in overall gain / signal levels). So even if you process this recording to even out the levels, the character of the sound may become somewhat different in the part where the gain is fading.

But maybe it won't be much of a difference, since any given amount of distortion on changing musical waveforms is generally much harder to discern than distortion on pure sine waves, depending on a whole host of factors. It's an audiophile myth that musical signals are a more sensitive test for distortion; experimentally it can be shown that that belief is way, way wrong. And personally I only hear distortion on pure sine waves when the THD is greater than one or two per cent. (A power amplifier with a burnt-out output transistor can be a useful learning tool sometimes.)