Investigation of effects of soil conditioning on EPB TBM performance

The vast majority of tunneling projects are now being carried out by mechanical excavation methods due to their superiority over traditional drilling and blasting methods. Although mechanical excavation methods are superior to other excavation methods, problems are encountered especially during excavation of urban tunnels. Most of the tunnels in urban areas are built on soft ground, shallow depths, and under water table. For this reason, the problems caused by tunneling on soft grounds and studies on the solution of these problems have attracted attention in recent years. With the developing technology, Tunnel Boring Machines (TBMs) are widely used in difficult and mixed ground conditions in urban tunnels where the risks are high. Under these conditions, it has become very important that the efficiency and safety of underground excavations. Engineers and contractors have to do tunnel excavation quickly, at low cost and with minimal damage to the environment. Although the developing technology makes it possible, especially in urban tunnels, mixed ground structure is encountered with problems due to low tunnel progress performance, increased cost and dents due to the size of the risks, thus causing damage to the surrounding structures. In order to improve the existing technology and reduce the problems mentioned above, it is necessary to combine new researches and existing experiences with the experiences gained from the realized projects. Soil conditioning has a great prospect for increasing the tunneling performance both in the Earth Pressure Tunneling Machines (EPB TBMs) and reducing the damage to the environment by reducing surface settlements and creating an effective counter face pressure if necessary. This preliminary response to soil conditioning is often carried out by field engineers on experience and assumptions, particularly with regard to tunnel progressive performance, or on the basis of soil conditioning in mixed and difficult ground conditions and understanding of machine-soil-conditioning interaction. Due to limited work in this area, the issue of soil conditioning in the EPB TBM tunneling continues to be a "black box" in our country and in the world. The main objective of the thesis is effect of soil conditioning on the EPB TBM excavation performance parameters (Torque, thrust, net penetration) and to develop an empirical model for the EPB TBM perormormance prediction. The effect of soil conditioning on the EPB TBM excavation performance parameters as well as the interaction between the soil, the foam and the cutting head will be understood and optimum foam utilization rates for different types of soils will be recommended. This information will be used by planners, contractors and consultants during the planning phase as well as providing solutions to the problems encountered during excavation. The study will enhance existing knowledge on soil conditioning and will provide new literate methods on the effects of soil conditioning. Foam bubble sizes foam density and half-life were measured inorder to determine the properties of the foam and anticlay chemicals used in the experiments. Particle size distribution and atterberg limit test were carried out to determine the characteristics of the soil samples. Optimum foam designs defined by making soil conditioning experiments with slump, flow table, cone penetrometer, power measuring mixer, pocket type vane shear and pocket type penetrometer tests to determine interactions foam types with the soil samples. Experiments were done firstly by only with water, followed by ground + foam mixture. Foam optimization with such a "multiple" test system has not been found in the literature. Studies in this project have also shown that there is a margin of error in the optimizations made with the single test system in the literature. The foam characterization test results were compared in order to investigate the effect of different types of (surfactant-polymered) foams of different manufacturers on the soil conditioning performance. Foam density has been found to decrease with increasing FER. In general, polymer-containing foams were found to be less dense than polymer-free foams. Foams were produced at different CF and FER values and the amount of drainage and half-life were measured. The most important factors determining the amount of foam drainage and half-life are CF and FER. Foam half-life time was found to increase with increasing CF value and increased FER values. The highest half-life time was measured as 723 seconds at 3% CF and 16 FER, and the lowest half-life was measured at 251 seconds at 5% CF and 8.5 FER. It has been found that polymer based foams have a longer half-life at low FER ratios than polymer-free foams. Three tunneling sites in Istanbul are visited for obtaining information on geology, construction method, TBM; obtaining soil samples; applying in-situ experiments using pocket penetrometer, static cone penetrometer, pocket and field types vane shear devices; and collecting excavation performance data of EPB TBMs. In the laboratory, physical properties of the soil samples are determined (specific gravity, natural water content, size distribution, consistency limits). Foam density and half-life tests, along with measuring foam bubble size under microscope, are performed to determine properties of the foams and anticlays used in the tests. Soil samples and foam generated with a laboratory type generator are mixed and their interaction is determined using slump, flow table, cone penetrometer, mixing with power measurement and sticking amount) and pocket type vane shear and pocket penetrometer tests; and optimum foam designs are determined by each test. The findings of these studies and the effects of various foam designs and formation properties on the performance of different EPB TBMs were examined and the most important parameters were determined. The optimized soil conditioning parameters and conditioning methods on field were analyzed in order to validate the experimental results. During the excavation of Zeytinburnu Ayvali Tunnel it is observed that soil has a high potential of adhesion. During laboratory conditioning studies it is observed that with optimum soil conditioning parameters (FER 16, CF 3% and FIR 250% and 12% more water) sticking problem would be solved and also performance will be increased. Optimum ground conditioning parameters could be partially applied in the field and it was seen that the cutter head torque decreased by 13% and thrust force by 10%. On the other hand, the net excavation speed increased by 32%. During the Mahmutbey-Mecidiyeköy Metro Tunnel excavation, two different soil types, silted sand (Mahmutbey-M.köy1) and sandy silt (Mahmutbey-M.köy 2), were studied. The Mahmutbey-M.köy 1 sample is optimized with SLF 30 at CF 3%, FER 16 and FIR 20%, in natural water content (17%). In Mahmutbey-M.köy 2 sample, the sample shows aggregation property and it is anticipated that the increasing amount of water will increase the stickiness property and it is optimized with CF 3%, FER 16 and FIR% 250 with ACP 143 in 35% water content. According to the performance measurements during tunnel excavation, there is no relation between EPB TBM thrust and cutter head torque and FIR. On the other hand, there is a relationship between FIR and EPB TBM penetration. At decreasing penetration values (increasing excavation difficulty) FIR appears to be increased. It is also seen that the FIR value is increased with increasing thrust force. Tuzla Akfirat Tunnel contains clayey sand with gravel which has a potential of adhesion. The optimum soil conditioning results obtained with 12,5 % more water and ACP 143 polymeric foam ( FER 8.5, CF 3% and 120% FIR). Tuzla Akfırat Wastewater Tunnel has encountered different grades of clay content. As the clay content in the soil increased, the stickiness increased and the EPB TBM performance decreased. The relationship between FIR, FER and Foam Quantity values and some performance parameters of EPB TBM (EPB TBM penetration, thrust, cutter head torque, specific energy) were investigated. The FER and foam amount did not correlate significantly with the performance parameters. There is a general trend between FIR and penetration. When considering only clayey soils (when clay-free, non-stick soils are kept out), the applied FIR limit valuesgenerally range from 150-300% in very sticky zones, 50-200% in medium sticky zones and 50-100% in less stickey zones. There is a strong correlation between EPB TBM penetration and FIR, as the excavation rate decreases (excavation becomes more difficult) FIR increases. Mini EPBM is developed within the scope of project, which is a first in the world, in order to investigate the effect of soil conditioning on EPB TBM performance parameters under real working conditions. Unfortunately any test was performed by using this machine. The "mini-slump" test tool developed in the scope of this thesis has been reduced by ½ of the standard slump test instrument, reducing the need for ground sample by about 90% and providing a very strong relationship with the flow table in homogeneous soils without coarse material. The findings of the studies and the effects of different foam designs and soil properties on the performance of EPB TBMs were examined and important parameters were determined. In the scope of the thesis, it is demonstrated that the consistency of soils and soil + foam mixes can be predicted by using pocket type vane shear. This test apparatus has been shown to be used in conditioning experiments. Within the scope of the thesis, studies were carried out in Zeytinburnu Ayvalı 2 tunnel to estimate the excavation performances of EPB TBMs used in fine grained soils. Observations were made during the excavation of the samples taken from 3 different formations along the tunnel route. Particle size analysis, Atterberg consistency limits, cone penetration and vane shear tests were performed on the soil samples. TBM performance analysis was performed on the 3 different geological formations and the effects of soil properties on excavation performance were investigated. In order to develop EPB TBM performance prediction models, soil properties were correlated by simple and multiple regression analyzes with excavation performance parameters, and statistical excavation performance models were developed. The results of these studies can be used by engineering companies, project owners and contractors in feasibility and planning stages.