The sound file was imported into a video editing
suite and the sound waveform was extracted.
Launch Sound
Analysis, showing the different features of
water rocket noise during take-off.
Conclusions / Analysis
We had a close look at the sound produced
by the rocket, as much more accurate
measurements can be made on the timing of
events on the rocket. (refer to diagram)
You can clearly see the click the
rocket makes at is leaves the launch tube.
With that you can accurately measure for how
long the launch tube is effective, and get
the rocket's initial acceleration.
You can also measure exactly how long
the water phase is when comparing your
rocket to simulators. There is a
distinctive transition to air pulse.
You
can also measure the length of the air
pulse to a certain degree but as the
pressure tails off in the rocket it is
hard to determine where the air pulse
finishes.
When we zoomed in on the start of the
air pulse there was a distinctive set of
regular spaced pulses, with fairly high
amplitude. The camera easily picked these
up at a range of perhaps 20 meters. From the waveform we measured
the frequency to be 297 Hz ever so
slightly increasing over time, as their
amplitude decreased before being swamped
with noise. They are clearly visible for
about 20 ms (see diagram).
We do not know what these represent,
whether they are perhaps some sort of
sonic shockwaves, as the air accelerates
close to mach 1, or the resonant frequency
of the rocket or nozzle.
Since all objects have a resonant
frequency, we tried just blowing across
the nozzle of the rocket and recording the
sound. We couldn't quite get a pure note,
but when we removed the nozzle we got
quite a nice pure note. The resonant
frequency of the rocket segment in this
case was: 122Hz which isn't close to the
frequency we observed. The waveform shape
in this case was also very different -
more sinusoidal rather than the sharp
pulses. This is why we are leaning to some
form of shock waves rather than just
resonance.
Thinking that perhaps this was a
unique case we reviewed the audio from a
number of launches and other rockets and
similar pulses were present (in both
stereo channels) although in this example these were the
clearest.
Since the audio was gathered from the
ground there is likely to be a Doppler
shift in the audio frequency also, but
since we don't have a clear measurement
how far away from the rocket we were, it
may be difficult to accurately predict the
expected Doppler shift.
We don't know if these pulses at the
start of the air pulse and their frequency
are significant, but it may turn out to be
important when designing CD nozzles.
