> The spectrum analysis in Audacity says it's all below 9kHz, with a
> peak below 500Hz which might be noise from outside that I couldn't
> block. Very interested to hear opinions about the nature of the
> noise.
>
> I haven't had time to take the recorder anywhere truly quiet to
> record a better sample, so this was done at home with doors and
> windows closed, sandwiched between two cushions with a blanket over
> the top. I'm sure there's plenty of sound from outside in the
> recording, but at least I've reduced it.
Peter,
This is good stuff. I'll have a better listen later, but for now I
selected the constant noise section, Normalised it and took a power
spectrum. The higher frequency external noise rolls off nicely under
the cushions, leaving a steady hiss. I used my 4K high pass filter and
again normalised and took another spectrum. I could have used the HPF
again, but the spectrun shows what I expected.
Mid range constant noise falls in the higher KHz. This is typical of
amplifier noise, either in the mic or in the input stage. "Thermal
noise" which is a more fundamental noise caused by moving air
molecules and electrons in very quiet amplifiers rises with frequency
at 3 dB per octave and this did't show up and couldn't be heard, so
the noise levels are higher than the "ideal" system, so no surprise
there. All par for the course for built-in mics.
Try to memorise this hiss and aim to get recordings with this noise
level nicely lower than what you want to hear.
I listened to the other tracks and the "noise removal" has killed the
atmosphere and yet not removed the helicopter. I'll try to do a clenup
tonight on the original to see what can be done but it is a good
recording apart from a not very good stereo image, again par for the
course. Experiment with a piece of cardboard between the mics with a
slot for the recorder. It may not work but it won't cost anything. :-)
If that helps, try two sheets widening out to get as near to the mics
as possible. Test the stereo images with something like my shaking
peanuts test, walking around the mics.
I have a hunch that putting the recorder on the ground may reduce the
LF noise. It's always worth trying.
> I also did a similar recording with the level turned right down.
> There was similar noise, but of much, much lower volume. I assume
> this proves it's not preamp noise?
No. The preamp most likely comes before the volume control which may
not go down to zero.
David
David Brinicombe
North Devon, UK
Cogito cogito ergo cogito sum - Ambrose Bierce
|