The curve of best fit was drawn through the data points. The best 

 fit description relies on the statistical measure of the extent of the 

 relationship between the two variables called the coefficient of correla- 

 tion (R) . A value of R = 1 indicates a perfect association or complete 

 dependence of one variable upon another. The value of R is shown along 

 with the curve chosen for each pile. 



The curve of best fit will yield accurate estimates if the variation 

 around it is small. The variation around the curve is called the standard 

 error of estimate. Standard error of estimate was calculated and is 

 also shown in Figure 16 with the respective curves for each test pile. 



Because of the natural variability in materials such as wood, an 

 absolute standard for all timber pilings and standard deviation of the 

 received ultrasonic signal cannot be determined. Each pile tested has 

 its own signature that defines the properties of that pile. During an 

 inspection, an average standard deviation would have to be recalibrated 

 on a solid section for each pile. 



NCEL Ultrasonic Field Testing 



In June 1982, NCEL installed seven timber pilings with various 

 degrees of biological damage along the corrosion test pier located in 

 Port Hueneme harbor in California. This pier is situated where marine 

 borer populations and their activity are known to be great. These timber 

 pilings were allowed to deteriorate further from biological attack for 

 approximately 1 year. In July 1983, three of the seven installed pilings 

 and one of the support piles for the corrosion pier were inspected by 

 Navy divers. Two inspection techniques, ultrasonic and impact testing,* 

 were employed. 



For successful evaluation of any inspection technique, a range of 

 biological damage was desired. First, the divers visually inspected the 

 installed piles to locate the timber pile with the least, and the other 

 with the most, obvious amount of damage. As expected, a new pile had 

 the least amount of external damage. The ultrasonic tests were first 

 conducted on the new pile (no. 8). Following the ultrasonic tests on 

 the new pile, the Navy divers inspected three additional piles with 

 various degrees of external damage. 



The same equipment configuration as that in the ultrasonic labora- 

 tory tests was used (Figure 7); initial preparations were also similar. 

 Variations were as follows: the length of the pile was divided and 

 marked with a line into 1-foot increments instead of 3- to 6-inch incre- 

 ments; the external damage, internal damage, and any other pertinent 

 information were recorded on video as well as in drawings; the separa- 

 tion distance between the transducers in the transducer holder was set 

 at 22 inches instead of 19 inches; and the ultrasonic signal from the 

 receiver was averaged 100 times to reduce the noise level. 



Two divers and an engineer at the surface were employed during the 

 actual testing. One diver used video equipment to examine the pile for 

 damage in the area being tested and documented the actual test procedure. 



*See IMPACT TESTING section for discussion of this technique. 



31 



