|We've got signal, but what the heck is it?
That's your mission. Analyze this bébé.
download the audio file 11.wav.gz
- No need to gunzip this file since baudline can automatically uncompress
Setup baudline to be a Web Browser helper application as described in the
delta measurement bars
to measure pitch change.
Use a Windowing function
such as Kaiser or Gaussian and increase the beta variable to improve the time
Due to deficiencies in the .wav file format, the sample rate is not
1.0000 but it is 0.9766 samples per second. Also note that the base
frequency (0 Hz) is 59.8144 Hz.
What is it?
It is the power grid. 60 Hz AC (alternating current) from the power
lines that have bled into the microphone. The interesting thing is that
the fundamental randomly wanders about 60 Hz with a delta of 0.05 Hz (50
mHz). The frequency is not constant.
Keep in mind that you are looking at almost 3 hours worth of samples with a
bin resolution of 0.00095 Hz/bin. This signal was captured with an ADC
sample rate of 4000 samples/second and a decimation rate of 4096 for an
effective rate of 0.9766 samples per second. The input device's
down mixer was set to a
frequency range of 59.81445 ... 60.30273 Hz. This is an extreme amount of
tuned frequency zoom. Just like an electron microscope, with enough gain
even the smoothest surface will look rough.
The two important questions at this point are "can this be error?" and "why is
60 Hz wandering?"
Is it error or an artifact? Since this signal is microphone-bleed-in
then it is possible that the computer power supply, the ADC clock, or the
microphone itself are creating this wander / drift. So an important test
is to use completely different computer hardware in a different room and see
if the shape of the wandering is identical. This is easy to do with a
local network, X-Windows, and the X11 remote display feature. When you do
this side-by-side the curves are the same. The signal really exists.
Why is the 60 Hz power line frequency changing? Power is created by
huge spinning turbines that turn massive generators. Also the demand for
power is dynamic. I would guess that these two items coupled together
make it rather difficult to maintain a rock steady RPM and AC frequency.
The system control tolerances are probably greater than 0.05 Hz.
We have seen a lot of other strange behavior while monitoring the power
grid. Large phase shifts that show up as a wideband pulse.
Frequency discontinuities, the frequency instantly jumps 0.05 Hz or more.
Both of these are probably caused by quick grid switching of power plants that
have not been synchronized. What we haven't witnessed yet and would
likely be fascinating to see are a brownout and a cascading power outage.