Tunneling Factors

Compared to the Model T, today’s cars are complicated.  But they are more reliable.  Excavators, loaders, rock drills, and TBMs are complicated.  Jet airplanes are even more complicated, and yet, more reliable.  Complexity has not and does not necessarily bring unreliability.  We must not be over impressed by the old tunneler’s aphorism, KISS – “Keep It Simple, Stupid”.  It’s worth keeping in mind, but it’s not a General Rule.  In fact, there may not be any general rules other than: “Pragmatism Rules”.

However, there may be general principles that can inform our approach.  The first general principle is that it’s going to take energy to remove the in situ ground and take it outside.   In the attached “Derivation A” we come up with this equation:

Tunnel Advance Rate = Power Intensity x Effectiveness x Continuity x Intelligence

It is important to note that this equation is merely a reflection of the definitions that went into it. No empirical information is expressed.  It is opinion.  But, hopefully, it seems pretty intuitive. 


The last term, “Intelligence” (volume of rock excavated per amount of energy applied) provides an interesting departure point.  In 1955 the U.S. Army Snow, Ice, and Permafrost Research Station undertook research to build military installations under the ice in Greenland as part of the DEW Line.  They measured the specific energy for ice excavation by many methods including hand labor, many kinds of mining machines, and even melting. 

The lowest specific energy, by far, was a man with a pickaxe.  What they saw was that a man with a pickaxe would attack the ice and once having started a crack would locate subsequent blows to enlarge the crack.  Each blow was predicated on the results of the previous blow.  He was using his intelligence to maximize the amount of excavated ice.  Anyone having done hand digging has done the same, seizing every opportunity to pry out large chunks.

A relevant aside at this point is to note that rock is generally already cracked.  Often there are three intersecting planes of cracking producing an in situ rock body of tightly assembled chunks.  We might usefully view rock excavation in such cases as “rock disassembly”.

Rock is much weaker in tension than in compression.  An interesting development that might not be familiar to working tunnelers is the CERAC Breaker.  Developed in 1978 by Atlas Copco, this breaker is designed to capitalize on the weakness of rock in tension.  The device is inserted into a short drilled hole.  It mechanically expands, gripping the inside periphery of the hole and, pushing off of the bottom of the hole, pushes the gripped rock outward breaking it in tension.

In the 1980’s the CERAC breaker was extensively studied by the U.S. Bureau of Mines as an enabling technology for what they termed the “Drill/Break System” of mining.  The Bureau investigator found the system very promising, but it has not been implemented commercially, perhaps because mechanizing the process was perceived as too difficult.

The Rapidex system, developed and tested in the 1970’s was a semi-continuous drill and blast system designed to advance into the rock utilizing small blasts in a spiral pattern resembling a lighthouse stairway.  The government funded development through many testing stages, but the mechanical difficulties encountered in finding, loading, and initiating previously drilled holes were too much for the technology of the time.

Many other systems have been proposed for excavating rock more “intelligently”.  Some involve drilling small holes and microblasting on a semi-continuous basis.  Others involve injecting a highly pressurized combustible gas and igniting it, or pumping in water and creating hydraulic shocks.  They all require drilling holes into the rock face and introducing some means to push the rock outward towards the free face.  Although most hold promise, none of the many novel methods proposed have been developed to the point of commercial success.  Consistently, the difficulty seems to be in the mechanization.

All these concepts were developed and prototyped years ago before the recent proliferation of computer controlled industrial manipulators.  Now, with position sensors and control programs worked out and generally available, an array of mining duty manipulators could be developed and used to test and refine any of these systems.

Or, turn it around; a general purpose manipulator array not wed to any specific method could interchangeably and alternately test and utilize any and all promising methods.  The idea is that the manipulators could grab drills and other tools of many kinds and through programming changes turn them into an Adaptable Tunnel Excavation System.

Robots are generally recommended for harsh, dangerous environments.   A rapidly advancing tunnel heading is a crowded, loud, wet, and dusty place.  Good for robots - not so good for humans.  But, to take maximum advantage of intelligence the manipulator array would need to be under human supervisory control. The operator would not be, as today, working the valves. Rather based on the results, he or she is seeing, they would be adapting the supervisory program.

Normal excavation with a given complement of tools would see the details of the attack vary as the lay of the rock changes. Major discontinuities such as faults or water inflows would call for a rapid switchover to different tools in a new configuration with a different control program.

Beyond adapting continuously to differing rock conditions on any one project, an Adaptable Tunnel Excavation System would be easy to adapt between projects. The manipulators, control systems and the many, many tools would be modular. They would only need to be reconfigured to mine different size tunnels in different ground conditions. The basic backbone of equipment and expertise would always carry over.