Spatial positioning of sidewall stations in a narrow tunnel environment: a safe alternative to traditional mine survey practice

3.8. Tunnel surveying

Various methods of tunnelling surveying exists. In most cases the surveying method adopted depends largely on the scope of work and the specific site requirements. According to the Tunnel Engineering Handbook the preferred method of establishing a “working line” in a tunnel remains the conventional traverse method familiar to South African Mine Surveyors. Bickel & Kuesel noted that the “working line is usually established by traverse survey methods….” [47] . They describe the installation of survey control or” working points” in the roof of the tunnel excavation as follows:

“…if the tunnel is in stable ground, pins are driven into the crown of the tunnel to serve as work points. The theodolite is centered under a plumb bob suspended from the pin. To establish a work point in soft ground where tunnel movement is anticipated, a plumb line and target are attached to moveable slides bolted to ring flanges or tunnel ribs in the crown of the tunnel. A survey platform is mounted below the slide and plumb bob positions…” [47].

1. The “Zigzag” method

In some cases the tunnel traverse can be carried from specially designed brackets attached to the side of the tunnel and traverses are extended diagonally across the tunnel from one side of the sidewall to the opposite side of the tunnel in a manner that would form a network of interlinked triangles. In a paper detailing the surveying work done in the Eurotunnel linking the United Kingdom and France it was stated that: “The limitations on usable space for surveying led to the design of brackets which are situated very close to the wall of the tunnel. A set of three” V” grooves at 120 degree spacing are incorporated into the brackets to provide forced centring for all normal instruments using a standard WILD pillar plate.” [81] Korritke described using the interlinking of lines from alternate sides of the tunnel as follows: “The alternate position of the brackets at the right-hand and left-hand side of the tunnel allowed a strong zig-zag route for traversing and it was expected that horizontal refraction could be avoided.” [81].

In his presentation at the FIG conference in 2007 Lee described the method used to transfer survey control points in the tunnel as follows: “At first, the horizontal control station in the form of a bracket was installed at the lower part of tunnel. The station was then plumbed up to the roof of tunnel. The bracket was removed to make way for the construction work. The station was transferred onto the walkway from the roof station and it would last beyond the end of the project.” [123]. He described this method, called a double zig-zag method of traversing, as a method of surveying where “Each traverse leg by itself became a quadrilateral. A double zigzag traverse linked up the quadrilateral” and explained the method of establishing and extending the traverse as follows

“The method consisted of beginning with setting a first set of stations in pair about 10m apart on one side of the tunnel and they were connected to the second set of stations in pair about 250m ahead on the other side of the tunnel. The second set of stations was then connected to the third set of survey stations in pair 250m in front on the opposite side of the tunnel. The fourth set and the rest of the control stations were set up in the similar zigzag manners to reach the break through end of the tunnel.” [123].

Lee recommended the double zig-zag method of surveying stating that the method allows the detection of gross error and the orientation of the setup using at least four control stations. [123] The zig-zag method of maintaining primary survey control in a tunnelling environment was also discussed by Fowler, “typically traverses will consist of zig-zag observations, alternating between brackets on either side of the tunnel. The zig-zag method also helps to alleviate the effects of lateral refraction, the error source in tunnelling surveys.” [45]

Figure . The Double "zig-zag" method

Bannister discusses the Channel tunnel survey in his textbook and comments as follows on the survey network: “Because of the difficulties of access, the survey team’s initial idea was to setup a traverse along one side of the tunnel only…The traverse was changed to a zig-zag arrangement, and this allowed the two ends of drives from France and England to meet within 0.5m, a satisfactory achievement when it is considered that a 1” error corresponds to a deviation of 200mm in horizontal alignment over the 37km length of the tunnel.” [28]. The application of the double sided zig-zg traverse was implemented as a result of the fact that the azimuth of a single line traverse indicated a curve to on side. It was determined that this curve was as a result of the lateral refraction caused by observations being taken in the confined spaces close to the side of the tunnel. It was found that if a zig-zag traverse layout was followed the observed angles were “alternately too large and too small by similar amounts” leading to a self-compensating overall azimuth. The rigidity of the zig-zag method is achieved by overlapping observations and a symmetric crossing of two lines of traverse. The network is further strengthened by regular gyroscope azimuth determinations. [124]

3.9. Typical Tunnelling Network layouts

The establishment of survey stations using the resection method with modern instruments seems to have become the acceptable means of providing primary survey network control in some tunnel networks. In narrow tunnels requiring extreme precision the preferred method of establishing survey control is the “zig-zag” method. [63] In the case of larger diameter tunnels, it is argued that the geometry needed for well-balanced surveys should be achieved easily. The question remains whether any of these techniques can be safely applied in a narrow environment such as found on typical South African mines. In order better to understand the techniques of establishing primary surveying networks in tunnels an analysis of some of the more recent tunnelling projects will be made in the following section.

1. Diamond Light Source tunnel, Rutherford Appleton Laboratory, Oxfordshire

Wilson briefly describes the method of aligning a tunnel using floor reference points to provide reference points for key installations for machine installation. According to Wilson “Each Monument is mounted on a fixed wall bracket and centred over its respective floor reference pin using a Leica ZNL optical plummet” [125]

2. Euro “Channel” Tunnel

Bannister described the initial establishment of the survey on the “English side” of the tunnel in the following manner: “Stations were marked by brackets at about 75m intervals high on the tunnel walls, but this meant that the theodolite sights had to be made close to the cold concrete lining of the tunnel, while the general body of air in the tunnel was warm from plant and machinery.” [94]. Checks made by gyrotheodolite indicated that refraction was causing larger closure errors than anticipated after which the method of surveying was changed to a zigzag arrangement. Fowler supported this information on the survey system in his paper stating that: “The English side of the Marine Service tunnel consisted of a scheme of two zig-zag traverses that overlapped and occasionally linked up “ [45]. He continued to describe the method employed by the French side as follows: “…consisted of a much denser network, called a “module topographique”. This network has two main advantages. Firstly, it was able to identify refraction effects by analysing angles within the many polygons that could be formed. Secondly, the density of the network meant that brackets further back could be dismantled to allow for construction and the control would still be able to be re-established.” [45]. These arrangements ensured a satisfactory arrangement of the tunnels from France and England meeting within 0.5 metres over the 37 kilometre length of the tunnel. [45]

3. Fermilab Main Injector ring

Bocean et al described the survey and alignment process followed at the Fermilab tunneling site as follows: “Considering the dynamic nature of the tunnel control system due to deformation, thermal conditions, other environmental factors and especially the ability to detect these small movements with today’s instrumentation, a strategy was adopted using inexpensive yet durable monuments…The final control network will utilize a system of braced quadrilaterals enveloping the volume of the tunnel between the constraint points and incorporating all floor monuments and bench marks.” [126].

4. Hong Kong tunnel Construction

In a paper presented at the FIG conference in 2010, Lam describes the surveying methods used in Hong Kong tunnel construction:

“…underground tunnels by zig-zag or braced traverses through access portals…the braced quadrilateral method is often applied with least squares computational software…After the coordinates of horizontal coordinates of horizontal control points at the intersection of the tunnel and the shaft have been fixed, gyro-theodolite is used to establish and maintain direction in driving the tunnel.” [127] .

Lam recommended that “The merit of the method is that number of the backsight of a station has been increased to four, offering more chance of getting an unobstructed line of sight from one station to the other. Also a number of known stations available on site make the task of fixing a station by the Method of Satellite easier and quicker.” [127].

5. Lane Cove Tunnel, Sydney, Australia

In his dissertation Fowler remarked that the Lane Cove tunneling project in Australia used a combination of survey methods to establish the survey network, noting that “the stations within the network are a combination of brackets, free stations and spigots. Whilst there is no clear pattern to the network, it contains a lot of redundancy. The idea behind such networks is that each bracket is connected to at least three stations outbound and at least three stations inbound” [45].

6. Japanese tunnel construction

The construction tunnels in Japan is described by Oshima used standard underground traverses, checked by a check survey and gyroscope observations. Unfortunately not a lot of detail is available for this project. Oshima notes that: “The two points dropped in tunnel are extended by means of traversing survey with total station system... traversing points set on the upper half of tunnel section and lower part were compared to the coordinates and confirmed within +30mm and direction errors by Gyroscope were also checked within 5 second.” [128].

7. Project Hallandsas, Sweden

Fowler describes a method that /*-*to be similar to the zig-zag method of surveying, but it notes that freestation setups were made between these points: “the control network incorporates brackets every 150m, on each side of the tunnel. In addition a free-station setup is made between each pair of points, alternating from the tunnel centerline by 1 to 1.5m every 150m (Lewin, 2006)” [45].

8. 
   Superconducting Super Collider , Waxahachie, Dallas , USA
   

A paper describing the survey method followed at the supercollider site in Texas, USA, states that survey control was established using wall-stations, using zig-zag traverses, checked by gyro-theodolite bases at regular intervals. The survey method and network was described by Greening et al. as follows: “This network consists of 22 points and is augmented by a number of similar monuments which serve as calibration lines for gyro-theodolites and EDM instruments…A length of 150m was eventually selected for the primary traverse legs. This provides a reasonable compromise between the accuracy requirements and conditions such as station intervisibility.” [129] The network is described as dual zigzag traverses through pairs of points with gyro-theodolite measurements to ensure that the accuracy was met. [129] Chrzanowski described the method as a combination of “underground control was run with double zig-zag traversing (to reduce refraction errors) using GYROMAT 2000 precision gyro-theodolites and TC2002 total stations placed over specifically designed wall brackets spaced at 150m intervals” [30].

9. Synchotron Radiation Source

Prasad et al described the survey method of surveying as primary control points being placed in the floor of the excavation, with the secondary survey system as wall brackets. The primary network consisting of concrete pillars equipped with forced centring sockets for the mounting of instruments and prisms. The secondary survey network is described as consisting of brackets distributed along the inner and outer walls of the tunnel. [130] These brackets were equipped with bases to accommodate the centring of instruments and targets. The locations of these brackets were selected “…after considering the parameters like focusing (magnification) distances of theodolites, visibility of components, uniform positional strength (error ellipses) in all directions ….” [130].

10. Uetliberg tunnel, Zurich Western Bypass

The Uetliberg tunnel is the major part of the Zurich Western Bypass. In a paper published by Amberg, the authors described the use of their tunnel monitoring system in the construction process of the transport tunnel. From the description it is clear that sidewall stations consisting of brackets mounted in the sidewall of the tunnel were used for the primary survey control of the tunnelling project. “However on the Uetliberg tunnel on each side of the tunnel a Leica TCRA1105 is mounted on a permanent bracket out of the way of the construction equipment. Each of these instruments contains TMS SETout (plus) on-board software with all project data loaded via PCMCIA card from TMS OFFICE on an office based PC. This is the common platform for handling all the design data, such as alignment, excavated profile, position of arches, thickness of shotcrete required etc..” [131].