Re: Reply to the article of M. Sansalone and W. B. Street I will begin by quoting from the first page of Chapter 18 in the book, 'Impact-Echo: Nondestructive Evaluation of Concrete and Masonry' by Sansalone and Streett. The title of this chapter is, 'Bonded Post-Tensioning Tendons'.
_____________________________________ 'The impact-echo method can be used to detect voids in grouted tendon ducts in many, but not all, situations. The methods applicability depends on the geometry of a structure and the locations and arrangement of tendon ducts. Just as is the case for other types of flaws, small voids in tendon ducts can be located too deeply within a structure to be detected. In some cases, complicated arrangement of multiple ducts, such as often occur in the flanges of concrete I-beams, can preclude detection of voids in some or all of the ducts. In other cases, portions of structures can be successfully tested and information can be gained that permits an engineer to draw conclusions about the condition of the grouting along the length of the duct. In all cases, the impact-echo method is restricted to situations where the walls of the ducts are metal rather than plastic.' 'Effective use of the impact-echo method for detecting voids in grouted tendon ducts requires knowledge of the location of the ducts within the structure. This information is typically obtained from plans and/or the use of magnetic or eddy-current cover meters to locate the centerlines of the metal ducts. Once the duct locations are known, impact-echo tests can be performed to search for voids.' _____________________________________________
Mr Krause describes a structure that is 600 x 2000 x 1500 mm. For impact-echo this is a bounded structure that will respond with many modes of cross-sectional vibration, and the resulting waveform and spectrum will be complex and difficult to interpret. No one who understands the capabilities and limitations of impact-echo would attempt to locate flaws in a structure of this shape without first performing extensive 3-dimensional, finite-element, computer simulations to obtain detailed information about the response of the solid structure. Mr. Krause states that computer simulations were performed for his specimen, but no details are given. The only computer simulations we know to be successful for a case such as this are those described by Sansalone et al. in references [39, 40, 41] in http://www.ndt.net/article/0298/streett/refer.htm Even with information from computer simulations it is unlikely that impact-echo signals from flaws at the mid-point of the 600mm dimension in a 600 x 1500 x 2000mm structures can be separated from the cross-sectional mode signals. It has been made very clear in published papers on the impact-echo method that structures bounded in 2 dimensions, such as beams and columns, respond differently from plate structures, which are defined as structures in which the lateral dimensions are at lease five times greater than the thickness. The impact-echo response of beams and columns is discussed at length in Chapters 21 and 22 in the book by Sansalone and Streett: http://www.impact-echo.com/Impact-Echo/bullbrie.htm
A 600mm thick slab that has lateral dimensions greater than 3 or 4 meters will respond to impact-echo tests as a plate, and it will be a straightforward matter to locate many types of flaws. However, a post-tensioning duct located at a depth of 300mm (one-half of the thickness) presents a special case. Multiple P-wave reflections from the steel in a fully grouted duct will have the same frequency the P-wave reflections from the bottom of the plate, and it will not be possible to identify the response from the fully grouted duct. It will, however, be possible to identify a void in the duct. For ducts that are located well above the middle point of the thickness, both grouted and ungrouted ducts can be detected. This is explained in Chapter 18 of the book by Sansalone and Streett. If the ducts are located very close to the bottom of the plate, they may be too deep to be detected. In Chapter 9 of Sansalone and Streett it is pointed out that flaws with lateral dimensions less than about one-fourth of their depth cannot be detected. A 100mm wide flaw at a depth of 450mm, in a 600mm thick plate, cannot be detected. If the opposite side of the plate is accessible, where the depth is 50mm to 100mm, the flaw can easily be detected.
If the post-tensioned structure to be tested is a portion of a beam or column, such as the web of an I-beam, then locating voids in grouted tendon ducts will be straightforward only if the dimensions of the web are such that it behaves as a plate for impact-echo purposes. This was exactly the case in the testing the beams of a large highway bridge in the northeastern United States in 1995. The webs of these beams were 200mm thick and the lateral dimensions were 2 meters or more. The post-tensioning ducts were 70 to 80mm beneath the surface. After preliminary impact-echo tests, which identified both ungrouted and fully grouted regions in the ducts, several ducts were opened using a jackhammer, and the impact-echo predictions were found to be 100% accurate. Photographs of the bridge and of the opened sections of the post-tensioning ducts can be found in the published papers on this project, and in Chapter 18 of the book by Sansalone and Streett. The published papers are references [11,12] in http://www.ndt.net/article/0298/streett/refer.htm
In summary, I wish to point out that in the article on impact-echo in this months on-line journal, Sansalone and Streett have said that impact-echo can be used to locate voids in grouted tendon ducts in MANY TYPES of post-tensioned structures. It has never been claimed that it will be successful for all post-tensioned structures. The inability of impact-echo to locate voids in the ducts in the specimen tested by Mr. Krause and others is very predictable, and it is regrettable that others may have made the claim, in advance, that impact-echo would be successful in this case. However, it would be wrong to generalize from Mr. Krauses experience, and conclude that impact-echo cannot locate voids in the tendon ducts of any post-tensioned structure. Those who wish to understand the capabilities and limitations of impact-echo in detecting voids in the grouted tendon ducts of post-tensioned structures should read the book and the published papers cited above.
In answer to Mr. Krauses question about the effects of steel reinforcing bars on impact-echo tests, I refer those who are interested to Chapter 17 in the book by Sansalone and Streett. The short answer is that impact-echo is not a suitable method for LOCATING reinforcing bars. By choosing the proper diameter of the sphere used to make an impact, it is usually possible to minimize the impact-echo response of the bars, as long as they are bonded to the concrete, and this makes it possible to focus on other problems. Cracking caused by corrosion of steel reinforcing bars, and the resulting deterioration of the surrounding concrete is easily detected. A project which saved more than $5 million in repair costs in a California sea wall was concerned with exactly this problem. A report on this project by the U.S. Federal Highway Administration can be found on the web at http://www.ota.fhwa.dot.gov/roadsvr/CS011.htm. Additional details, including photographs, can be found in Chapter 17 of Sansalone and Streett.
These discussions reinforce the point made by Dr. Grosse, that impact-echo is not in a 'plug and play' state of development. Inflated claims have been made for its performance, and some scientists and engineers who have used it have spent little time studying the physics and mathematics of the method, and examining reports of its successful use. Those who have taken the time to learn the capabilities and limitations of the method have found it to be useful.
Prof./Dr. William B. Streett Cornell University Ithaca, New York 12 February 1998