Thanks for looking at it, Vicki. I assume because the bias is really a supe=
rimposed wave with no overall bias, it should show up in a spectrogram as a=
signal at 0.25Hz. If I set Audacity's spectrogram window size to 32768 (th=
e maximum) and the max spectrogram frequency to 100Hz (the minimum), I see =
a bit of a lump in the 0-10Hz area at each spike in the waveform, two big o=
nes and a little one.
I don't think it's really that useful to look at the spectrogram though, th=
e problem is plain to see in the waveform. I played with fixing it by dupli=
cating and inverting the track, then using eq to filter out anything from t=
he duplicated track above about 20Hz, then mixing it with the original. The=
bias disappeared from the waveform, and it sounds ok to me.
I'm still not absolutely convinced it's caused by the flat battery because =
I've had the mic on for a couple of hours now with that same battery, and i=
t's still not doing it. I think it was about 10 degrees C when it was happe=
ning, and it's nearly 20 here in the house, so that might be boosting the b=
attery output a bit. I might try putting it in the fridge for a while.
Peter Shute
From: O=
n Behalf Of vickipowys
Sent: Sunday, 27 October 2013 9:52 AM
To:
Subject: Re: [Nature Recordists] ME66/K6 with signal bias
Peter,
How curious! I downloaded the sound clip and looked at it in iZotope
RX. No anomalies show up in the spectrogram but they do in the
waveform, and the levels jump upwards according to the levels meter
(but not audibly), in a cyclic pattern unrelated to the birdsong.
I have no explanation for this phenomenon. But it does seem to be a
lesson for people to keep their batteries fresh. A chap I know
changes his K6 gun mic battery daily 'whether he needs to or not', on
recording trips.
Vicki
On 27/10/2013, at 9:18 AM, Peter Shute wrote:
> I recently left my ME66 switched on overnight, and when I
> discovered this in the morning and switched it off and on again,
> the battery confirmation light was very dim.
>
> The first recording I made after that seems ok, but a couple after
> that show a bias in the signal, showing in the waveform as a funny
> shaped wave, repeating every 4 seconds or so. Later in the day this
> decreased to about 1 second, although the recordings sound ok to me.
>
> I've uploaded a sample to http://snd.sc/1a7KEuG
>
> Is this bias the normal result of a dying battery? When I tested it
> again just now before changing the battery, it seemed ok again,
> perhaps because the battery had recovered a bit. I've left it on to
> see if I can replicate the problem.
>
> I'm not at all concerned about saving the recording, only about
> whether the mic is faulty or just needed a fresh battery.
>
> Peter Shute
>
|