Friday, September 01, 2006

Shearing Box

Box of Sand with a line of colored sand as a tracer.
Shaking Frequency: 5 [Hz]

Tracking of the line over a shaking-cycle.
Shear rate (\dot{\gamma} = \frac{du}{dy})
assuming all the shearing happens in layers. Valid just for the first cycles.

Tuesday, August 22, 2006

Sound Recordings

Right Boom
booming time: 30 [s]




Left Boom
booming time 10 [s]

Friday, July 14, 2006

Sound Recordings

Sample of the signal at the begining of the booming, total time 30 [s].




Recorded with TASCAM, Audio Pro.

Wednesday, November 02, 2005

TunningFork C, 128 [Hz]


Hearing test uses this kind of forks, and they are the closer to a booming sand frequency. Performing the FFT of the sound (using MatLab at 8000 [samples/sec]). This might be useful to test the resonance responses in the loose sand layers with a wet sand layer (like a damp) underneath

Wednesday, September 14, 2005

Clipped signal


With this simple plot, it's possible to see that even having clipped signal, it is possible to get the right characteristic frequency peak.

Microphone calibration

The low frequency microphone that we tried before, didn't work due to the fact that changing the SamplingRate of the mic function, it would give a different result for the characteristic frequency peak. Today, I tried a regular microphone (RadioShack unidirectional dynamic microphone, IMP 450 [Ohms]) and a tuning fork, A 440 [Hz], the results are as follows:
  • 8000 [samples/s] ------ 440.1855 [Hz]
  • 11000 [samples/s] ------ 439.2548 [Hz]
  • 22000 [samples/s] ------ 439.2548 [Hz]
  • 44100 [samples/s] ------ 440.0848 [Hz]

In fact, I just found in Matlab "...standard audio
rates are 8000, 11025, 22050, and 44100 Hz...", 101205
The rest of the SamplingRates in between give a wrong answer. I ran an experiment were a frequency generator was playing for a constant period of time, at the same time I’m recording the sound for 0.5 [s] at a given SamplingRate, doing an FFT, and finding the peak, this process will repeat over until the SamplingRate in MatLab is covered (8000 - 44100 [samples/s]). I'll non-dimensionalized the frequency peaks dividing them by the frequency from the frequency generator, so that all the points which are one are correct, and the rest are not.

Friday, September 09, 2005

Resistivity

Run6

Run6: recordings were taken 55 [ft] down the crest, and about 100 [ft] left from the Run1.Two geophones: I/O 1 Yellow, I/O2 Yellow. HandHeld settings; SamplingRate: 20800 [samples/s], 10000[samples], therefore 2.08 [s] of recording time.

Run5

Run5: recordings were taken 55 [ft] down the crest, and about 100 [ft] left from the Run1. Two geophones: I/O 1 Yellow, 55 [ft] down the crest, I/O2 Yellow, 65 [ft] down the crest. HandHeld settings; SamplingRate: 2000 [samples/s], 10000[samples], therefore 5 [s] of recording time. Although the 2 geophones were 10 [ft] apart, the characteristic frequency was the same.

Run3

FAILURE, HandHeld had a full memory, and the laptop didn't record what was expected.

Run4


Run4: recordings were taken 55 [ft] down the crest, and about 100 [ft] left from the Run1.Two geophones: I/O 1 Yellow, I/O2 Yellow. HandHeld settings; SamplingRate: 2000 [samples/s], 10000[samples], therefore 5 [s] of recording time.
Problems: One of the geophones wasn't amplifying the signal properly, but the peak on the frequency is the same for both of them.

Run2

Run2: recordings were taken 100 [ft] down the crest, with two geophones: I/O 1 Yellow, I/O2 Blue. HandHeld settings; SamplingRate: 2000 [samples/s], 10000[samples], therefore 5 [s] of recording time.

09095

  • Place: Dumont Dunes
  • Objectives: geophones recordings, sand samples, and resistance probe.
  • Difficulties: recording signal with the computer: Changing the sampling frequency modifies the corresponding characteristic frequency. Using the computer to get the booming signal through the mic-jack port is not RELIABLE .
  • Experiments and settings:
  1. Run1: recordings were taken 50 [ft] down the crest, with two geophones: I/O 1 Yellow, I/O2 Blue. HandHeld settings; SamplingRate: 2000 [samples/s], 10000[samples], therefore 5 [s] of recording time.

Tuesday, September 06, 2005

More Places

So far, we used to go to Dumont Dunes, which is close to Baker, CA. In fact there is a trip planned for this coming Friday, September 9th. I found a couple of places which are close to California, in fact in Baja California Norte; Los Algodones, San Carlos, Cerro Juan, Cerro Kino (close to San Felipe), El Pinacate, and El Gran Desierto de Altar.
I hope being able to provide more information about those places.