Laboratory Test Results . Low frequency ultrasonic theory indicates 

 that the strength of the signal picked up by the receiver will be stronger 

 in solid wood than in damaged wood. The transmitter sends sonic waves 

 into and down the grain of the wood. The receiver picks up the signal 

 and relays it to the oscilloscope, which measures the strength of the 

 signal and its time of passage for the fixed distance in the wood. The 

 amplitude of the digitized ultrasonic waveform received indicates the 

 strength of the signal. As a result, the decrease in the RMS amplitude 

 of the ultrasonic signal received is the parameter that holds the greatest 

 potential for correlation with the amount of cross-sectional wood loss. 

 The RMS is the square root of the mean of the squared amplitudes. The 

 formula for calculating the RMS amplitude is as follows: 



RMS 



1/2 



After looking at the RMS amplitude of the ultrasonic signal for the 

 various test piles, a large discrepancy was apparent between the RMS 

 values for different timber piles. For example, the average RMS values 

 for the standard pile were 1.4 to 2.2 mV, while the average RMS values 

 for pile 5x were 14.5 to 15.0 mV. The DC level of the ultrasonic signal 

 was shifting, seriously affecting the calculated RMS. To compare RMS 

 values of the ultrasonic signals, a synchronized horizontal time base is 

 required for identifying the wave amplitude components for attenuation 

 measurement. However, to compensate for the irregular DC shift exhibited 

 in all the ultrasonic signals, the standard deviation of the ultrasonic 

 signal over a specific time period was determined. The formula used to 

 calculate the standard deviation is shown below: 



Standard Deviation = 



Ex. 



N 



- ;» 



1/2 



The standard deviation creates a synchronized horizontal time base by 

 subtracting the mean amplitude squared from the sum of squared ultrasonic 

 signal amplitudes as shown. This eliminated the erroneous readings 

 obtained from comparison of only the magnitude of the amplitudes (RMS) . 

 Standard deviation, mean and largest peak to peak value were cal- 

 culated over four different time domains of the ultrasonic signal. This 

 signal was separated into discrete time frames or "bins" (100, 200, 300, 

 and 400 usee) to evaluate the difference in sound velocity between wood 

 and water. Compressional sound waves travel in solid wood with a 

 velocity of approximately 5,000 m/sec along the grain, whereas waves 

 travel through seawater with a velocity of 1,500 m/sec. Figure 11 shows 

 two ultrasonic received signals. A typical acoustic sound wave received 

 after traveling only through water is shown in Figure 11a. In comparison, 

 an ultrasonic signal received after traveling through a solid timber 

 pile for the same path length or transducer separation of 22 inches is 

 shown in Figure lib. The smaller amplitude wave indicated by a "1" in 

 the figure is the part of the acoustic signal that traveled through the 



23 



