I'm seeking help with best practice methods for installing secondary and primary survey control to a narrow vein underground metals mine.
Preferably seeking procedures or worked examples explaining best geometry to minimise positional and orientation errors.
I'm from an U/G Coal background which is completely different to U/G metals. Planning to carry out a mine of life survey at the current mine I am at and use this as my PAP as I'm seeking an "ALL" mining endorsement.
Any help or suggestions would be much appreciated.
Static GPS vectors from state survey control marks to mine site surface baselines, incorporating sets of angle total stations obs will easily satisfy Class B accuracy. Follow 2nd (preferred) or 3rd Order principles as the methodology.
Carrying control underground: I'm assuming that the survey is being carried down a portal so the key is maintaining a strong bearing, especially if a tunnel drives straight for a considerable distance. Forced centered setups and observing the previous setup, the next setup and all wall stations visible. Some wall stations between both previous setup and forward setup, some beyond the previous setup and forward setup. Try and shoot each wall station at least twice but more is better. Keep the baseline distances (between setups) to no more than 200-300m. Some people say don't survey further than 100m, this is where it gets personal based on experience. To me, it's all about the bracing. Follow 2nd order procedures and run a minimum of 4 sets of obs. Along the traverse, set up under back station pins at regular intervals so as "baselines" can be referred to for gyro checks and as a redundant means of advancing control if the wall stations fail. Run the obs through a processing package like Compnet to test the quality of the obs, run a network adjustment and calculate absolute/relative error ellipses followed by a Class test. Mines Class D isn't too hard provided a close can be performed and/or a gyro is observed to independently verify the bearing.
Working control is typically done using the resection function on the total station to wall stations. What I've found is that if you set up and survey inside a minimum of 3 wall stations with a good resection SD, you'll carry good bearing. Again, more stations observed, more confidence can be established. Principle errors with two station wall stations are 1) EDM inaccuracies (2mm+/- 2ppm can throw a bearing) and 2) Wall stations not inserted into the wall correctly or the wall has shifted since the station was first surveyed in. These need to be mitigated before advancing control. We mined an 800m incline with 100-200m long straights and broke through with less than 20mm error using the above mentioned working control method.
a final comment on the control survey activities of the mine. Keep records and write reports on all primary control surveys. Back everything up in a clear/detailed standardized format. It is a lot easier to read a report detailing how a survey was done (including problems or issues - how they were handled) than having to re-process it yourself to know how good or bad it is. Control Stations in a database are meaningless if you have no idea on how they were established.
Survey Class & Order has now been superceded by the transition to Uncertainty but has Class & Order been misunderstood in the past and is there going to be a legacy from this?
One sees many references to mine surveys requiring to be done to a specific Class "accuracy" and the Queensland Standard 10 is a classic example of this (Section 3.1 Mine Coordinate System). However Class was/is much more to do with field methods, equipment, network design, observations, technique, methodogy, reduction approach etc and these principles really have not changed all that much with Positional Uncertainty. Class does have an inherint "accuracy" (for want of a better term) but only in an unconstrained network. Class alone does not define the site "coordinate acucracy" as in connection to datum in a fully constrained network adjustment, which is where Order comes into things.
A GNSS survey may have been done to Class B "standards" hence the best Order that could be achieved is 2, however there is absolutely nothing preventing a Class B survey from being Order 3 or Order 4 or Order 5 or even NO order at all so the site's control station accuracy is really dependent on a survey's Order and it is the Order that should be specified when referring to the control "accuracy" required, not the Class.
Hopefully the right terminology will be carried into the Uncertainty era?