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

8.6. Phase 3. Observation protocol

The procedure for surveying a sidewall survey station is similar to the forward observation of a new hangingwall survey point. It is recommended that the minimum observations from an unknown point should be at least 3 points in the case of a resection in order to ensure redundancy for observations. Middelton et al remarked that “…as a system, however, observations to three points are unsatisfactory and lazy, the method is not susceptible of minute accuracy, and there is no check, unless four points are observed...” [116].

1. Risk factors in observation

The research project has highlighted the following risk factors in the observation protocol used for sidewall survey stations. It was found that the accuracy of the survey network was greatly affected by the short steep sightlines in the vertical components of the network. It became apparent that unless very stringent control measures are in place a large error in direction can be propagated over a very short distance. In order to minimize the effect of direction error propagation it is suggested that the establishment of a gyro baseline is essential to ensuring the accuracy of an underground network. During the evaluation phase of the research it became evident that short sighting distances made accurate alignment on the reference points difficult, specifically when standard, large prisms are used. The findings underline the fact that advanced survey equipment and software cannot replace solid basic survey experience and observation procedures. The following hazards where identified and listed in the following risk assessment of the observation protocol:

Table . Risk analysis for the observation phase of installing sidewall stations.

Risk	Hazard	Contributing factors	Mitigating Actions
Potential injury to survey crew	Ground conditions	Hangingwall Sidewall Geological conditions	Obtain permission to enter area Request inspection and barring by appointed person
Survey reference survey stations damaged or destroyed	Off-line mining Mine layout must be changed Unplanned breakthrough Mis-holing Damage to structural design (pillars) Insufficient control for redundancy	Blast damage Theft or intentional vandalism Only two reference objects used to fix position	Use non-metal survey stations Ensure survey stations are placed in area where probability of damage is minimized Freestation software should be able to identify error in position fix Use more than two survey stations to orientate
Incorrect identification of survey stations	Off-line mining Mine layout must be changed Incorrect survey observations Inaccurate measuring and profiling	Incorrect numbering Survey stations identified incorrectly Incorrect marking of survey stations Incorrect reference survey stations observed Incorrect survey stations information on instrument database	Communication between crew members Surveyor must check all installations personally Freestation software should be able to identify error in position fix Database verified and checked Additional lighting of targets by survey crew
Survey control inaccurate	Insufficient control for redundancy Off-line mining Mine layout must be changed Unplanned breakthrough Mis-holing of excavations Damage to structural design (pillars)	Weak reference survey stations geometry (angles exceeding tolerance) Survey stations too close to observe (instrument focal length) Survey stations identified incorrectly Measurement error due to incorrect instrument or prism settings Poor observation protocol Instrument off-level Incorrect prism constant used Incorrect EDM setting used Prisms damaged Incorrect prisms used Damaged or incorrect prism adaptors used Incorrect reference object identification Incorrectly numbered survey stations used Incorrect reference bearing input in instrument Incorrect target alignment	Training in application of method Direct supervision of installation process Freestation software should be able to identify error in position fix Software should prevent final position fix if not within determined standards Instrument calibration and settings verified Surveyor to verify instrument settings as part of observation protocol Take at least two observations and measurements to each reference survey station Prisms and attachments in good condition Correct prism and prism attachment used Prism attached correctly
New control survey stations surveyed inaccurately	Control stations in database have incorrect co-ordinates Mine layout is affected by inaccurate co-ordinates Off-line mining may result from incorrect survey information	Instrument off-level Incorrect prism constant used Incorrect EDM setting used Poor observation protocol
Incomplete or no records of observation	Instrument software does not provide a clearly defined template of essential survey information	Software template insufficient or not available	Develop template to show required information Print hardcopies Backup electronic records

[159]

2. Geometry of observations

In the underground environment it is not always possible to install survey stations at the optimal position for ideal geometry. The references found on the geometry of survey stations to be observed for an intersection seem to differ substantially in the approach to the ideal geometry. Shewmon noted that “the 2½º, 2½º, 175º triangle offers greater accuracy than the 45-45-90º triangle…” [15]. Arthur states that “we have found that the best geometry use when surveying a new station is within 2-3m of the 2nd reference station and < or > than 90 degree angle from that station.” [43]. Briggs gives the impression that he is contradicting the findings made by Shewmon, noting that “the ideal (strongest) triangle is one having angles close to 60 degrees (equilateral), although angles as small as 20 degrees may be acceptable.” Briggs concludes that “Providing that the measured angle lies between 165º and 180º, and the known sides are roughly equal in length, the effect of angular error is inappreciable as compared with that of linear errors...” [154] It is not certain whether the current view held by certain surveyors and software manufactures of maintaining angles larger than 20 degrees and smaller than 160 degrees are as a result of these observations made by Shewmon and Briggs. It is obvious that observations to two points where one angle is 160 degrees and the other is 20 degrees will add up to 180 degrees therefore the points will not form a triangle. Jaroz and Shepard notes that “ data suggests that to achieve optimal directional accuracy in “wall station traverses” configuration of acute triangle geometry is necessary” [96] A similar constraint on angles appears in the Institute of Mine Surveyors of South Africa. [41]

The findings of this research suggest that acceptable results for a resected point can be found even under conditions that would normally be considered to be poor triangular geometry, and that the implied accuracy of the setup improves as the geometry of the triangle increases in the angle at the observation point, and that in such cases distance does not pose a problem. It was found that the grade peg configuration provides an accurate fix only if more than two points are observed. From these findings it emerges that the accuracy of points is not influenced dramatically by the geometry of observations or the distance of the freestation to the reference objects. However, it is apparent that the points surveyed forward using the freestation fix, do not always give accurate results, unless a strict observation regimen is followed.
The fundamental principle of the “danger circle” is a well-known concept with its origin in plane table surveying and solution of resections by the use of an error figure. The term “danger circle” is used when the unknown point comes close to lying on the same circle as three of the reference points. In such as case the solution of the co-ordinates of the unknown point may be ambiguous. The fundamental principles for the observation of a resection remain valid. A surveyor that understands the underlying principle of the “danger circle” can ensure that precautions are taken to prevent any such occurrence, which preventative measures are listed in the Standard Operating Procedure in

Appendix 8.

3. Maximum and Minimum angles and distances

The research indicated that a control point could be transferred forward from a freestation setup provided that the freestation setup complied with the constraints of a maximum of 42m ahead of the last survey control or alternatively a minimum angle of 2 degrees. As indicated in Figure the constraints will conform to the current MHSA requirements for minimum standards of accuracy. The results listed in the table were obtained using a setup of using two survey reference points were observed in a configuration that approximates a Weisbach triangle. It has been found that a position fix using only two reference points does not provide acceptable accuracy on the position of forward surveyed points. It is therefore argued that the use of three or more reference objects should provide a more accurate fix of position and possibly allow longer distances from the last cluster of reference objects. The geometry of such a setup remains weak and it is therefore recommend that such a configuration only be used for activities that will allow a lower standard of accuracy.

It is indicated by the results of the research that the error in a network is propagated in bearing rather than distance. The increased distance from reference objects in a narrow tunnel environment will have an influence on the size of the angles observed. A theoretical spreadsheet calculation using a configuration of two sidewall stations spaced at a distance of 3.0 metres apart indicates how the subtended angle at the instrument position changes as the setup point is moved outside the strong configuration of sidewall stations around the resection point, refer to Figure .
The design of the model is based on a standard mine tunnel and a regular spacing of survey stations similar to the grade peg layout discussed earlier. The layout assumes that the freestation is placed in the centre of the tunnel and the design used to calculate and BPC as measured at the freestation point. For the purpose of the graph offsets in 0.25m increments were calculated and the angle at point P_i calculated and graphed.
Figure . Idealized station layout.
It can be observed that the subtended angle decreases rapidly beyond 9 metres. For a narrow tunnel, restricting the subtended angle to 20 degrees or more, as recommended by some authors, will reduce the distance from the reference point group to approximately 7.5 metres. A tabulation and plot is attached in Appendix 9.

Figure . A graph of the subtended angles at incremental distances from the base stations

[160]

8.6.4 Number of observations

It is important to note that all observations made for fixing the position of a resection point and those made for the purpose of determining the co-ordinates of new survey stations should be checked by at least one additional observation. It is accepted survey practice on South African mines to observe a set of four “face-left” and then in “face-right” observations. The standard procedure is followed both to prevent fraudulent bookings by the surveyor as well as compensating for any errors in the axis of the instrument. The duplication of observations provides a check on the observer and the correct identification of the reference object. Some modern survey instruments have a “survey forward station” function will allow the surveyor to select single or multiple face observations. A theoretical analysis of the number of redundant observations obtained using a random setup of an instrument at an unknown point observing configurations of two, three and four known sidewall station reference points are listed and discussed below. The theoretical analysis is based on a document discussing the Freestation software used in Leica instruments. According to Zimmerman, the provisional co-ordinates of the point is calculated using “each possible unique observation combination” including triangles, resection(s) and a helmert transformation, from these unique solutions differences between calculated and observed values are calculated. A solution is then selected based on the smallest number of large differences between calculated and observed values. The final adjustment is made by robust least squares adjustment. [112] The freestation method of solution is discussed in section 3.4.7 on page 76 of this thesis and illustrated in Figure . In order to better clarify the “freestation” method of calculation and adjustment, the observations are described in detail in the following sections.


[B]
In a two point configuration for a resection, the coordinates of two sidewall stations will be known. If measurements are made to the known reference points it would provide two measured distances combined with the bearing and distance obtained from the join between the two known points. The following information will be available to determine the position of the setup point as illustrated in Figure and tabulation in the table below:



Figure . A diagram of a two point configuration
[161]

Bearing	Angle	Distance (m)
Bearing A to B	A-P-B	P to A
	B-P-A	P to B

The combination of two known reference points and two measured distances will make it possible to calculate two unique triangulation solutions for the setup point. The name resection in this case is not a true reflection of the method of position determination. The unique solutions are listed here:

Triangle	Angle	Distance (m)
Triangle APB	A-P-B	P to A
Triangle APB	A-P-B	P to B

In the case where three known sidewall stations can be observed from the random setup position, the following information will be available from observations and previously stored information to determine the position of the setup point. Three known points will provide three co-ordinates and three join calculations, providing three bearing and distance solutions. In addition measurements can be made from the setup point to the three known points and three subtended angles between the known points can be observed. The information available for calculation of the position of the setup point is listed below and illustrated in the diagram Figure below:

Figure . A diagram of a three point configuration
[161]

Bearing available	Angle available	Distance (m) available
Bearing A to B	A-P-B	A to B
Bearing A to C	A-P-C	A to C
	B-P-C	P to A
		P to B
		P to C

The combination of three known sidewall stations and the three measured angles and distances will provide the necessary combinations of redundant measurements in order to calculate seven (7) unique solutions for point P, including one resection:

Resection Solution	Triangle Solution	Angle Used	Distance (m) Used
Resection A-B-C	Triangle A-P-B	Angle A-P-B	A to P
	Triangle A-P-B	Angle A-P-B	B to P
	Triangle A-P-C	Angle A-P-B	P to A
	Triangle A-P-C	Angle A-P-B	P to C
	Triangle B-P-C	Angle C-P-B	P to C
	Triangle B-P-C	Angle C-P-B	P to B

In the case of a configuration making use of four known sidewall stations the following information will be available for fixing the position of the resection point. Four known sidewall stations will provide the co-ordinates to calculate 6 different join and 4 measured distances between the known points and the setup point. From the unknown point [P] four angles to the sidewall stations will be able to be observed.

Bearing available	Angle available	Distance (m) available
Bearing A to B	A-P-B	A to B
Bearing A to C	A-P-C	A to C
Bearing B to D	C-P-D	B to D
Bearing D to C	D-P-B	C to D
Bearing A to D		A to D
Bearing C to B		B to C
		P to A
		P to B
Bearing available	Angle available	Distance (m) available
		P to C
		P to D

Figure below depicts the combination of observations and measurements made in a four point configuration.


Figure . A diagram of a four point configuration

The combination of four sidewall stations with known co-ordinates and the observations and measurements made from the unknown point will provide sufficient information to allow the calculation of sixteen (16) unique solutions for point [P] as listed below:

Resection Solution	Triangle Solution	Angle Used	Distance (m) Used
Resection A-B-C	Triangle A-P-B	Angle A-P-B	A to P
Resection A-B-D	Triangle A-P-B	Angle A-P-B	B to P
Resection A-C-D	Triangle A-P-C	Angle A-P-B	P to A
Resection D-B-A	Triangle A-P-C	Angle A-P-B	P to C
	Triangle B-P-C	Angle C-P-B	P to C
	Triangle B-P-C	Angle C-P-B	P to B
	Triangle CPD	Angle C-P-D	P to C
	Triangle CPD	Angle C-P-D	P to D
	Triangle BPD	Angle B-P-D	P to B
	Triangle BPD	Angle B-P-D	P to D
	Triangle APD	Angle A-P-D	P to A
	Triangle APD	Angle A-P-D	P to D

From this theoretical analysis it can be argued that the use of four or more points will provide sufficient redundant observations to enable a more robust determination of the final position of point P when compared to a sidewall survey method that makes use of only two known sidewall stations for the position determination. Although the method of comparing the amount of calculations may seem trivial, these tabulations illustrate the calculations made in each case by the on-board software during a setup and is essential when explaining the advantages of this method to the conservative mine surveying industry. The method of adjustment by least squares follows after these calculations.

5. Field notes

Modern survey instruments have the function of various user defined printout options and recording formats of field observations. There is concern that the output of resection information is neglected in the current software that is available. In most cases only the final solution for the resection position is provided by the instrument but does not include the observations that were used to calculate the resection solution. It is argued that although the resection point is temporary and therefore only valid for the time of occupation, the control established from the resection point is entered into the survey database and the integrity of the data is therefore dependent on the orientation of the resection point. A hardcopy of the field observations that incorporate relevant data that should be used for the setup calculation is included in the Standard Operating Procedure in Appendix 7.

6.