IZMIR - KARŞIYAKA TUNNELS AND THE PROBLEMS ENCOUNTERED DURING THE EXCAVATION M Emre Ertem, Eregli Iron & Steel Works Inc., Kdz. Eregli, Turkey Celik Tatar, Halil Köse Dokuz Eylül University, Department of Mining Engineering, Izmir, Turkey

Abstract The tunnels, which are base structures of transportation, irrigation and mining facilities, have been finding new application areas in recent years as a result of advancing technology. The Karşıyaka tunnels which are contained in the Izmir Beltway project to connect Bornova to Karşıyaka are being built under the Yamanlar Mountain in the North of the city of Izmir as 4 tubes, totally 2 km. The New Austrian Tunneling Method (NATM) has been utilized in the Karşıyaka Tunnels for excavation and support work. In this paper, the excavation studies carried out in the Karşıyaka Tunnels are presented. The problems encountered during the tunneling work and the measures to overcome these problems are also summarized. Keywords: NATM, Supporting Application, Tunneling, Timing, Blasting 1. Introduction 1.1. General The reason of new structures construction by increasingly in mining, energy, watering, transportation, etc. is huge world population with acceleration and the covering wishes of people’s necessity in this huge population. Tunneling which has an important place in mining, also is taking more important place in constructions of transportation, watering. Generally, all underground structures can be summarized with one word that’s “tunnel”. The general problem of new engineering structures is finding suitable place in crowded cities. That’s mean, densities of present structures of cities don’t allow to us to construct new engineering structures in the middle of the cities. On the other hand, using simpler and shorter connection for the traffic system of cities is preferred instead of using complex and longer by engineers. Tunnels, which were constructed for motorways or ring roads, became more useful solutions for traffic arrangement. Therefore, a tunnel has an important place in road or railway construction. In the past, tunnel construction was thought for just passing big mountains, but nowadays-free land nonexistent forced to engineers for preferring tunnels construction instead of big excavations. 2.1. Karşıyaka Tunnels Karşıyaka Tunnels was located under the Yamanlar Mountain in the northern part of Izmir (Figure 1). These tunnels have been constructing as a part of Izmir ring road and Aydın - Izmir motorway. Two tunnels with four tubes were excavated less than 2 years.

K2 Tunnels K2 Tunnels

İzmir Bay

Figure 1. General position of Karşıyaka Tunnels 2. Investigation of Karşıyaka Tunnels Excavation 2.1. General Geology Tunnels lie into Yamanlar andesite complex, which occurred altereted-highly altereted andesite, andesite - dacite compound lava flowing, agglomerate, tuff and dyke flowed structured Miocene aged a volcanic structure. 2.2. Comparison of Uniaxial Compressive Strength between done by company and in Laboratory of Dokuz Eylül University The uniaxial compressive strength experiments done by company in investigation stage was given below, 2

Point Loading Experiments: 0.787-1.73 MN/m Uniaxial Compressive Strength Experiments: 187.2-415.2 kg/cm2 for K2P1 zone, Uniaxial Compressive Strength Experiments: 115.2-552.0 kg/cm2 for K2P2 zone. These results show similar values with the value of experiments done at Dokuz Eylül University, Department of Mining Engineering, which was given below. Table 1. Uniaxial Compressive Strength Experiments 1 2 3 4 5 Uniaxial Compressive 449,0 658,2 480,9 557,3 308,8 Strength (Kg/cm2) (K2P1SB) Uniaxial Compressive 121,3 266,9 389,0 375,6 205,5 Strength (Kg/cm2) (K2P1NB)

Av. 490,8 271,7

3. NATM for the Karşıyaka Tunnels New Austrian Tunneling Method was applied for excavating and supporting Karşıyaka Tunnels. The face was excavated as three parts; Top heading, bench and invert. Drilling (with jumbos) and blasting was used for excavating. Four rock classes were determined at investigation stage, but threerock classes system (4-5-P) was used in applying. Excavation and supporting was applied according to these rock classes, also excavation areas, amount of shotcrete, amount and numbers of rock bolts etc. were determined.

Figure 2. New Austrian Tunneling Method (Köse et. al, 1992) 3.1 Excavation and Supporting for the Tunnels (4-5-P) Karşıyaka Tunnels was excavated and supported by %3 of total length rock class P, %34 of total length rock class 5 and %63 of total length rock class 4. Standard support system was given for rock classes P-4-5 at Table 2-3. Table 2. Standard Support System for Top Heading Rock Class Unit 4 5 P Length of Round M 1.5-2.5 1.0-1.5 1.2 Shotcrete Cm 15 25 25 Thickness Steel Set Type GI 110 GI 140 GI 140 Wire Mesh Layers 1 2 2 4m 5/6 Number 6/7 Per Rock Bolts 6 m 4 8/9 4 Round 9m 4 Face Sealing Cm 3-5 5 3-5 Forepoling Number 10 25 37 Grouted Type 2″ Pipe Inj 2″ Pipe Inj 2″ Pipe Inj Pipe Spiling Length 4m 3m 2.5 m Lagging Table 3. Standard Support System for Bench Rock Class Unit 4 5 Length of Round m 3-5 2-3 Shotcrete cm 15 25 Thickness Steel Set Type GI 110 GI 140 Wire Mesh Layers 1 2 4m Number 4 Rock Per 6m 6 Bolts Round 9m Face Sealing cm 3 Yes/No Yes Yes Invert

P 2.4 25 GI 140 2 4 3-5 Yes

Figure 3. Arc Shape Formed by Bolting (Tamrock, 1983) 3.2. Supporting Applications When tunneling excavation is carried out by drilling and blasting, damage to the remaining rock mass can occasionally cause over-breakage. Causes for this problem are related to the properties of the rock mass and the presence of discontinuities along with the utilization of incorrect blasting patterns and/or with excessive explosive charges. In most cases over-breakage may be prevented by appropriate site investigations and rock mass analysis prior to the excavation. But in some cases when the problem occurs, it may be dealt with the following method of support system. Extra Bolts

Air Pipe

Concrete Pipe

Shape made by wire mesh Non-Scale

Figure 4. Supporting Applications for Overcome A frame is formed between two GI Profile with a steel mesh overlay. Before spraying this mesh with concrete, two pipes, one for grouting and the other for venting of the trapped air during grouting process, are placed behind the frame. Following this process, the gap between the tunnel wall and the frame is filled with concrete using a concrete pumping machine. The pumping stops when the vent tube is completely filled with concrete. Installing rock bolts that are 1-2 m longer than the length of the over excavated area to enable the rock to support itself follows this. 4. Conclusions Karşıyaka Tunnels excavation was started on September of 1997 and was finished on August 1999. 2000 meters long, 16 meters width, two motorway tunnels with four tubes was excavated less than 2 years.

Rock classification was determined correctly, that can be understood from the data of deformation through the excavation line, allowed us to have high excavation speed in construction and complete excavation less than 2 years. The uniaxial compressive strength experiments before constructing of 280 meters K1 and 1825 meters K2 tunnels was found between 115.2 kg/cm2 and 552.0 kg/cm2, which were found in the limit of uniaxial compressive strength results that were done at laboratory of Dokuz Eylül University, Department of Mining Engineering. (271.66 kg/cm2-490.84 kg/cm2) Rock classifications of Karşıyaka Tunnels were designed according to New Austrian Tunneling Method as three parts. (4-5-6-P) But rock class 6 wasn’t used in construction with a deal between contractor and engineer. It shows that New Austria Tunneling Method has a dynamic structure. This dynamic structure must be decided in the construction time with a suitable deal between engineer and constructor. This dynamic structure was affected from two things; economy and safety. This was also a proof for chancing of New Austrian Tunneling Method in construction with some measurements. Applied excavation systems and determined rock classes were successful at Karşıyaka Tunnels site. Engineers planned three-shift planning and it reduced excavation period for per round to 10 hours. Progress in a day could become 8 meters. Work product has been hold generally high because of easy rock mass and low losing time in construction as determined time search on the construction. Period time was observed for different formation and different length of round of a period. According to that observation, 2h: 15min: 10sec-5h: 52min: 10sec was measured (%17-33 according to total work time) for drilling (face, rock bolt fore polling) and it takes first line in the total work time. 1h: 30min: 00sec-2h: 55min: 50sec was measured (%9-19) for assembling steel set and welding and it takes second line in total work time. After that, shotcrete, blasting, excavation transportation takes an important place in total work time in turn in order. There are several reasons for long drilling time with jumbos, which are positions of machinery, formations and operators. Using high productive operators is easier than other problems to shorten drilling time. There should be enough operators for each booms of jumbo. According to formations, period time shows increase values in andesite and andesite with blocks. 14 h: 59 min: 44 sec period time was measured at the face with %80 very altered andesite at K2SB-600+803.60, 15 h: 38 min: 20 sec also was measured at the another face with %90 altered andesite at K2SB-600+048.20. When looked at the average of a months time measurements for a period, 11 h: 40 min was calculated for 1.5 m advance at rock class 4, 15 h: 29 min was calculated for 2 m at rock class 4 again. In addition, 45-435 min/month-loosed times because of human reason and 605-625 min/month-loosed time because of machine reason were determined. This loosed time, especially comes form people can take under control with suitable organization, planning and good management. The blast pattern was taken under control from the initiative of workers and new blasting pattern 2 3 3 was applied. 64.307 m *2 m = 128.614 m excavation, 171.7/128.614 = 1.34 kg/m explosive was suggested for top heading of the rock class 4, 40.18 kg explosive was determined for bench of the rock class 4.Total explosive used for all face of rock class 4 is 227.8 kg. 64.307 m2*1.5 m = 96.461 m3 excavation, 132.25/96.461 = 1.37 kg/m3 explosive was suggested for top heading of the rock class 5, 36.18 kg explosive was determined for bench of the rock class 5.Total explosive used for all face of rock class 5 is 184.35 kg. References 1. Ertem M.E. (2000) “Excavation Investigation of Karşıyaka Tunnels” MSc. Thesis, Dokuz Eylül University, Department of Mining Engineering, June 2000, Izmir 2. Erol A., (1999) Personal Interview) 3. Karaoglan H., (1999) Personal Interview) 4. Murat H., (1999) Personal Interview) 5. İçöz O., Aslan M.Z. (1998) Work-Time Analysis in Izmir - Karşıyaka Tunnels (Site Studies) 6. Jodl, H.G. (1995) Construction Method NATM, New Austrian Tunneling Method Summer Course, July 2 - July 8 1995, IACES, Bureau of Vienna 7. Köse H., Gürgen S., Onargan T., (1992) Tünel ve Kuyu Açma, Dokuz Eylul University, Izmir 8. Kutlutaş Dillingham Joint Venture (1999/2000) Karşıyaka Tunnels, Izmir 9. Tamrock. (1986). Handbook of Underground Drilling. Tempere, Finland 10. Tanrıver T., (1999) Personal Interview 11. Turhan R., (1999) Personal Interview