To detect anomalies in diaphragm walls

Abstract

Diaphragm walls are potentially ideal retaining walls for deep excavations in densely built-up areas, as they cause no vibrations during their construction and provide structural elements with high strength and stiffness. In the recent past, however, several projects using diaphragm walls as soil and water retaining elements have encountered severe problems. The problems primarily arise around the joints between panels. After excavation of the building pit, the joints slowly or suddenly start to leak. If a leak coincides with a permeable soil layer outside the building pit, the soil can erode, causing settlements adjacent to the retaining wall. An average 16% chance of leakage per project has been estimated from previous projects, making the chance of a calamity due to a leaking joint unacceptably high for current litigious society. Detection techniques have traditionally focused on groundwater flow, as groundwater flow through the wall is an important link in the calamity chain: no groundwater flow: no transportation of soil: no settlements. The flaw in such a detection system is the nature of the anomalies in diaphragm walls. Due to the production procedure of diaphragm walls, anomalies in most cases consist of bentonite (clay) pockets in the joint. These clayey anomalies have a high hydraulic resistivity, making them almost impossible to detect based upon the groundwater flow detection principle. After excavation and thus exposing the anomaly, the clayey material is too weak to retain the groundwater pressure, causing a leak which can quickly erode the remaining material in the anomaly. In contrast with the above mentioned detection principle, this research has primarily focused on the quality of concrete around the joints between the diaphragm wall panels. It is assumed that when persistent high quality concrete in the joint area is present, no leakage or soil transport through the wall can take place. Based on the physical characteristics of concrete, soil and bentonite slurry, several measurement techniques have been chosen for examination in the laboratory and in the field. Using the test results from pilot projects, three techniques have been chosen for further validation and, if possible, cross correlation with the other techniques and the actual shape of the anomalies. The validated techniques are (in order of effectiveness in a project setting): Crosshole Sonic Logging (CSL), Distributed Temperature Sensing (DTS) and Electrical Resistivity (ER). The effectiveness of all three methods has been based upon the cost of the measurements, the accuracy of the interpretation, the ease of interpretation and the interference with the production process. CSL is commonly used in large diameter bored pile integrity testing and is based upon the sound velocity in a medium. The velocity is determined by the stiffness and density of the material. For concrete these parameters are relatively high compared to the characteristics of the material that is expected to be present in an anomaly (soil or bentonite). An increase in the observed travel time of the ultrasonic signal indicates an anomaly. The simultaneously observed attenuation of the signal offers additional information about the properties of the anomaly. In this study the CSL technique has been verified for the novel application investigating the joints between diaphragm walls. This research has shown that the ultrasonic signal of current CSL devices can pass the joint between diaphragm wall panels while remaining interpretable. With the reference measurements of this study showing linear correlation between delay in arrival time and anomaly width, the size and material of an anomaly can be estimated, making preemptive repair decisions possible. DTS is generally accepted in diverse monitoring applications such as monitoring power lines, hydrological flow patterns, concrete curing temperature distribution and down-hole oil production parameters. The technique uses optical fiber sensors that provide a continuous temperature profile along the length of the fiber when read out by a DTS device. In this study the DTS technique has been validated for application during diaphragm wall production. The spatial resolution for tracing a progressing temperature front has been determined. This resolution is an order of magnitude better than suggested by the specifications of the measurement equipment. With the appropriate processing of the recorded temperature profiles in the time domain, the bentonite refreshing and concrete casting processes can be monitored meticulously. During bentonite refreshing, the temperature of the freshly mixed slurry should ideally show up at all depths of each recorded profile (positioned at critical locations in the trench e.g. in the joints). If locally the arrival of fresh slurry is not observed, the refreshing process can be repeated after additional clean-up of the trench. This ensures consistent slurry characteristics before concrete casting takes place. During concrete casting, the interpreted DTS recordings will show the casting progress in time for each DTS profile position. The observed temperatures also reveal valuable information about the purity of the concrete, making an estimate of the local concrete quality possible. Electrical resistivity methods are often mentioned as a possibility for detecting leaks. The method is based upon differences in electrical resistivity of soil and concrete. It is assumed that a continuous (fully cured) concrete wall will have a relatively high resistivity compared to a wall with clayey anomalies. In this study the method has been tested for detecting anomalies in diaphragm walls. Detection limits for several electrode configurations have been determined. From the test results, requirements for field tests have been derived. To obtain adequate measurements, at least a four electrode setup must be used with the potential electrodes placed no further than 0.2 m from the diaphragm wall. The research comprised laboratory and site testing in several projects. The project experiences are an important component of this research, as they illustrate the practical implications of the measurement techniques. As a result, it was possible to derive a manual for the execution of the measurements, containing practical tips for the interpretation of the measurement results. CSL is the primary recommended method because of the relatively low cost, low impact on the building process and reliable and fast interpretation. DTS shows great potential for a step forward in quality control during diaphragm wall production. Currently, the method will be beneficial for verifying concrete flow in pilot panels with rebar spacing beyond the design code requirements. The still relatively high cost of data acquisition and interpretation (about 100% of one panel building cost) limit large scale application. Electrical resistivity has been least successful in determining anomalies in diaphragm walls. In specific circumstances, the method could provide useful information, especially if other methods have not been applied and preemptive repair of the diaphragm wall with jetgrout poses a risk to the surroundings. Due to the data acquisition time and space requirement, this method is more costly than the other methods.