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

Error propagation

It is important to study the effect of error propagation in the proposed method of surveying. McCormack stated that “Any error in measuring distance would be carried throughout the entire traverse, but that error would not compound. Errors in turning an angle would be magnified by the further traverse from that point.” [110]. In their paper Jaroz and Shepard noted that the results obtained in their work indicated an increase of traverse errors related to the positioning of the instrument relative to the wall stations, stating that “It is known that the shape of the resection triangle does influence the accuracy of bearing transfer (as in the two wire Weisbach method of direction transfer through a vertical shaft)” [96]. Furthermore they suggested that an acute triangle geometry is required for optimal directional accuracy, observing that “…stations located normal or near-normal to “wall-stations” impact negatively on bearing transfer.” [96]. Simpson observed that in surface surveying trilateration is liable to error propagation in direction. [119]. Wolf and Ghilani remarked that “Whenever an instrument’s circles are read, a small error is introduced into the final observed angle...” [36]

1. Sources in error in running a traverse

Wolf and Ghilani listed the following errors in running a surface traverse, similar errors to which can be found in the conventional mine surveying technique of an open traverse, including:

1. “poor selection of stations, resulting in bad sighting conditions, …line of sight passing too close to the ground, lines that are too short…”

2. “errors in observations of angles and distances”

3. Failure to observe angles an equal number of times direct and reversed.” [36]

Furthermore mistakes made during the process of traversing can include:

1. “occupying or sighting on the wrong station”

2. “Incorrect orientation”

3. “Confusing angles to the right and left”

4. “Mistakes in note taking”

5. “Mis-identification of the sighted station.” [36]

Random error can be reduced with correct observation procedures. In addition to the prevention of error through proper observation techniques and procedures the “check survey” as defined in the MHSA remains the only method of ensuring the accuracy of a mine survey. Chrzanowski described the problems encountered in mine surveying as follows: “the elongated shape and comparatively small diameter of tunnels do not provide much choice and flexibility in the design of the underground control network.” [30], and defined the orders of survey that will normally be carried in the excavations of a mine, with the first and lowest order of traverse described as the survey carried with the advancing tunnel, providing directional control to the production personnel. The second order of traverse is what is commonly understood to be the “check survey” in the South African mining industry. Chrzanowski described a check survey as “the higher order traverse of a higher accuracy is established which checks the location of the last point of the construction traverse”. [78]. Such a survey may in some cases be followed by a third traverse of the longest possible distances in order to ensure the correct position of the final survey control. [78]. Cawood advances that surveyors seem to think that a check survey checks the accuracy of their work when in fact such an “error of closure primarily checks the precision, not the accuracy.” [120]. It is imperative that the surveyor not only ensures that correct procedures are followed but also that check surveys are regularly done to ensure the accuracy and precision of a survey.

6. Alternative surveying techniques

Although the open traverse from pegs in the hanging wall of an excavation is by far the most common method for surveying mine tunnels in South Africa, a number of alternatives in tunnel surveying have been investigated and implemented with great success. None of these methods seem to have been adopted in the South African mining context. Stengele et al proposed that all survey polygon networks aim at “minimizing unfavourable error propagation in elongated traverses and the systematic effects of dangerous horizontal refraction” [59].

1. The “Three Spad” Method

In the text: “A study of Mine Surveying Methods And Their Applications to Mining Engineering”, Young described the following method of surveying in poor ground conditions:

“When the roof is so poor that' a station cannot be established in it, three spads are driven in the wall near the roof for the support of the following device… In the center of a sheet of galvanized iron, cut to form an equilateral triangle, the base is punched a small hole through which is hung a plumb bob. To each corner of the triangle is attached a chain by which the entire device may be hung from the three spads in the walls." [6]

2. The “Random Transit” Method

In an article by Shewmon, a Mining instructor from the West Virginia Institute of Technology, reprinted from the Engineering and Mining Journal, Chicago USA, the random setup method is described. This method was tested before the introduction of Electronic Distance Measuring devices and focuses purely on an angular solution for the co-ordinates of the unknown point. “The random transit setup described here does not require taping to known stations. …where it is very difficult or impossible to tape to known observable stations, he may determine the transit location by shooting both known stations to determine the interior angle and recording vertical angles to both stations.” [106]. The method incorporates the installation of surveying pegs in the sidewall of an excavation and making angular observations to these points from a randomly selected position. No documentation exists that this method was ever evaluated in South Africa. The method did not incorporate the measuring of distances to known points. Shewmon stated that the advantage of this method was that “there will be no taping errors, and with good plug locations and a transit in good adjustment, accuracy as good or better than by taping can be obtained.” Shewmon concluded that “the bearing error was one minute and the horizontal error 0.02 ft. This is well within the accuracy desired for stope work.” [106], furthermore “this random setup method can also be applied in shooting two nearby triangulation stations or other located objects, and will give location to a transit with accuracy matching the accuracy of the original known points …” [106]. He concluded that “… it should be noted that transit men are occasionally prohibited from using random setups because a superior believes them to be too inaccurate. Actually, a random setup can be made with almost any accuracy desired up to 1ft in 400ft, through the use of strong triangles with four of five significant figures in the sides.” [106].

It must be kept in mind that no significant technological advances in the mine surveying field had been made by 1951, the only possible advantages being that offered by the “facet” type calculator. Observations were still made using a theodolite and distances measured by steel tape. Calculations were done using logarithmic tables. As a result of these limitations to the observation and calculation of sidewall plugs the article by Shewmon was critically analysed by Thomas from the London School of Mines in an article published in the Journal of the Institute of Mine Surveyors, in which he cautioned that when the calculation is done, certain solutions could become “unstable” leading to errors in the calculation, which errors would be created when the logarithmic tables of the day were used: “…if five figure tables are used, the fifth figure in the calculated value of Sin “n” will be doubtful… “ [121]. Moreover, for the calculation to be correct “…it will be necessary to work to eight significant figures in order to obtain reliable results.” [121] Thomas resolved that “The random transit setup can be used in certain cases, but care must be exercised when selecting a suitable position.” [121] With the introduction of electronic calculation methods the concerns raised by Thomas have been addressed as the conversion to logarithmic values for calculations are highly accurate.

3. Grade pegs²⁸ in concrete Piers

Rapoport and Heller discussed the difficulty in placing sidewall pegs for grade elevation control of a shaft excavation. In a paper published in the Journal of the Institute of Mine Surveyors, they observed that “Owing to depth, the side-walls scale, and it has been found necessary to construct concrete piers where grade pegs will be required. These piers are approximately three feet wide, and extend from the footwall to about two feet above the grade-line. When the grade pegs are no longer in use, the piers are removed.” [54]. A similar method of installing grade pegs in a development end is now commonly used on most totalstations with on-board mining software. The method however is not used for the establishment of survey control networks, but only for the control of the excavation gradient.

4. Central European Survey methods

Chrzanowski describes a method used in some mines in Central Europe where it was considered common practice to use “reversed hanging theodolites which fit on supporting bars screwed into the wall timbering [30], noting that the method was considered to be very convenient for the establishment of third order surveys in cramped conditions.

5. Wall Markers

The most common form of surveying in tunnelling mentioned in literature seems to be a method of building special monumentation or alternatively “wall brackets” to which a special bar could be attached and on which a theodolite could be setup. Chrzanowski described this method as using wall markers into which a portable bar was inserted for the mounting of the survey instrument, alternatively “the markers are used as eccentric stations to which the position of the survey instruments is referenced by measurements of short distances and/or angles.” If the markers were used as eccentric stations it was argued by Chrzanowski that the instrument could be set up at any convenient point “without the time-consuming task of centring under the marker.” [30].

6. Australian Wall Stations

Wall station surveying implies the use of resection to replace conventional underground traversing. “if resection is performed instead of traversing then surveyors could set up anywhere they wanted, as long as they were in sight of two known points.” [110] McCormack stated that the method relies on the measurement of angles and distances in contrast with conventional underground traversing that relies mostly on the accuracy of the angles observed. [110]. In the paper by Jaroz and Shepard stated that “Recently this technique has gained widespread acceptance in underground metalliferous mines in Western Australia.” [96]. This method of surveying has not been tested in the South African mining context. The authors observed from the results obtained that “the calculated results show a significant increase of traverse errors and a review of survey geometry suggests that such errors are related to positions of theodolite stations in relation to wall stations.” It seems that the study was based on a survey established from a 30m fixed base and characterised by five survey legs of thirty metres long and that wall stations were only installed on one side of the excavation. A significant improvement to the survey would have been possible if two more stations on the opposite sidewall could be used. Fowler stated that “personal experience has shown the author that the accuracy of such free station set ups is typically better than 0.002 for easting, northing and elevation.” [45] and confirmed that “the above techniques are now widely being used in underground mining operations across Australia.” [45].

Ludvigsen based in Norway, made the following observations on the method: “Benchmarks in the sidewall have been the normal for years in our part of the world…. Starting the survey by using free stationing, we can avoid errors by using wrong instrument heights and reflector heights. They are now always zero.” [97] He stated that “I have not documented and ”tested” the way we are working, but it works.”, and continued by stating that “… When total stations appeared, measuring distance along the line of sight, I started to use another arrangement, inspired by a paper given by Andrew Jarosz, at the ISM congress in China.” [97] in reply to a similar enquiry, Wetherelt, a Senior Lecturer in Mining Engineering at the Camborne School of Mines Of Exeter stated that “We use the in-built resection software on the Leica TPS1200s that we have”
In one of only two South African papers to mention the possibility of utilizing a similar surveying technique Venter and Suttie comments on the possible application of fixing the position of a survey station from two known points. “It is intended to introduce a system whereby the total station may be set up anywhere in the road and the position fixed by measuring distances to two pegs and the angle at the set up.” [117]

7. The “Grade-peg” method

Two South African papers discussing a similar type of survey were published in the Journal of the Institute of Mine Surveyors of South Africa. Hutchinson discussed the installation of grade pegs in a developing inclined mine shaft in Botswana, commenting on a possible solution for ensuring the accuracy of a survey when short distances are surveyed “...it has been standard practice to overshoot the day to day surveys with, what is termed, longshot surveys. At this mine, lines and grades are expected, and provided every 20 to 30 metres, even in areas where lasers have been installed.” [122]. He proposed that these “longshot” surveys could be 400 to 500 metres long and would be checked by gyro surveys. [122] Hutchinson affirmed that this method had been invaluable in giving less qualified surveyors the ability to perform repetitive tasks in arduous conditions that it had increased the productivity of a surveyor [122].

8. The JCI method of co-ordinated sidewall pegs

A brief mention of a system making use of a standard opto-mechanical theodolite and grade poles in the survey guidelines of the Johannesburg Consolidated Investments Company, Technical Services Division for Gold and Trackless Mechanised mining methods:

“In wide or high drifts co-ordinated sidewall pegs may be installed at a convenient height, from which, at a fixed calculated distance, the line of sight is taken with standard ranging rods. These pegs can be used to re-establish a survey by resection when, by using a programmed calculator the bearings, co-ordinates and elevation of a station can be established.” [107].

The method was only allowed to be used to carry a survey through a difficult area and was then re-established using conventional hangingwall survey stations. It can be attested that this method has rarely been used at the Randfontein Estates Gold Mine due to the complexity of the survey method. At that time no electronic distance measuring equipment or sophisticated calculation software was available.