U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Bridges & Structures

<< Previous Contents Next >>

Technical Manual for Design and Construction of Road Tunnels - Civil Elements

Chapter 15 - Geotechnical and Structural Instrumentation

15.1 Introduction

In the context of this manual, the primary purpose of geotechnical and structural instrumentation is to monitor the performance of the underground construction process in order to avoid or mitigate problems. If such monitoring also serves a scientific function, or leads to advancement in design procedures, that is a bonus rather than a primary reason for its implementation. A few decades ago monitoring was not a particularly easy task because the tools were few and some not so well developed. Monitoring was generally performed manually, and the refining of data to a state of usability from the raw readings often required long hours of "number crunching" with relatively crude calculators and more long hours of plotting charts and graphs by hand.

The world of the early 21st century is very different for those who pursue the art of determining what ongoing construction is doing to its surroundings, or even to itself. Advanced and refined types of instrumentation abound, and electronics coupled with computers has made remote monitoring, even from half a world away, practically an everyday affair. It is common for even medium sized projects to run a computerized database that reduces raw readings to usable data and can report on any combination of instruments and data plots within minutes. It can also inform interested parties any time of the day or night if movements or stresses have reached pre-set trigger levels that demand some kind of mitigative action. The possibilities have not gone unnoticed by project Owners, and comprehensive instrumentation and monitoring programs are becoming the norm rather than the exception. This is perhaps especially true in the world of tunneling where even small mis-steps can result in damage that may lead to lawsuits or the shutting down of operations.

Readers should be aware that much of the instrumentation described herein may not lend itself particularly well to rural highway tunnels, especially those located in hilly or mountainous terrain that may limit the need for instrumentation if great tunnel depth minimizes ground settlement at the surface, and if lack of surface development minimizes the number of third-party abutters who could be affected by construction. Also, even if a tunnel does require monitoring for whatever reason, great depth may minimize possibilities for damage to surface installations and push designers and constructors toward more in-tunnel installations.

The amazingly large number of instrument types available to tunnelers means that this chapter can do little more than "broad brush" the subject. The most common and/or most promising types of instrument will be covered, but readers will have to turn to the references to see what else is available. A few types will be covered to some degree in other chapters; for example, earth pressure cells that are commonly used by those who specialize in Sequential Excavation Method ( SEM) tunneling (Chapter 9), but are not so much used by those who work in other types of underground construction. Although vibration monitoring will be covered herein, the monitoring of noise will not be covered because it is normally considered an environmental rather than a structural or geotechnical concern. Some instruments, such as those used to determine in-situ ground stresses prior to tunneling, will not be covered because they more rightly belong in the category of site investigation instrumentation. And finally, there will not be space to delve deeply into the theory of operation of the various instruments discussed, so readers will again have to turn to the referenced publications for more details.

The first few sections of this chapter will discuss the types of measurements typically made:

  • Ground Movement away from the tunnel
  • Building Movement for structures within the zone of influence
  • Tunnel movement of the tunnel being constructed or adjacent tubes
  • Dynamic Ground Movement from Drill & Blast
  • Groundwater Movement and Pressure due to changes in the water percolation pattern

The first three items comprise quasi-static changes in position, and the last is also concerned with long-term effects. In contrast, Dynamic Ground Movement covers response due to vibration caused by the shock waves generated by explosive charges used to excavate rock.

All of the monitoring needs to be coordinated to fit with the tunnel construction schedule, and to establish the actions that must be taken in response to the instrumentation findings. These topics are discussed in the final section of this chapter.

15.2 Ground Movements - Vertical & Lateral Deformations

15.2.1 Purpose of Monitoring

The primary purpose for monitoring ground movements is to detect them while they are still small and to modify construction procedures before the movements grow large enough to constitute a real problem by affecting either the advancing excavation or some contiguous existing facility. For the advancing excavation, ground support has to be based on conditions encountered; monitoring either confirms the adequacy of the support or indicates whether more or different support may be required. Existing facilities may be at the ground surface - roads, railroads, buildings and the like - or they may be below ground in the form of utilities or other transportation tunnels such as subways. The first line of defense against potentially damaging movements is to detect them at depth in the ground immediately surrounding the advancing tunnel and take mitigative action before those movements can "percolate" upward toward the surface. This kind of monitoring can provide an indication of whether ground treatment such as grouting is effectively limiting movements that might otherwise result in troublesome settlements. Ground can, of course, move upward as well as downward, in the form of heave from unloading that can destabilize the invert of the tunnel under construction, and as a side effect lead to lateral, possibly damaging deformations as the ground moves toward the excavation to take up the slack. In addition to helping control the ground, the data developed can be used (and this may be said of all monitoring discussed in succeeding paragraphs) to verify design assumptions and to evaluate claims by construction contractors and third-party abutters.

15.2.2 Equipment, Applications, Limitations

Several types of instrumentation are used to monitor ground movement:

  • Deep Benchmarks
  • Survey Points
  • Borros Points
  • Probe Extensometers
  • Fixed Borehole Extensometers, either measured from the surface or during advance of the tunnel
  • Telltales or Roof Monitors
  • Heave Gages
  • Conventional Inclinometers
  • In-place Inclinometers
  • Convergence Gages

15.2.2.1 Deep Benchmarks

Deep Benchmarks (Figure 15-1) are steel pipes/casings drilled into stable strata - preferably sound bedrock - outside the advancing tunnel's zone of influence. They are used when existing benchmarks, such as those installed by the USGS, are not available and it is important to know actual elevation changes of other instruments meant to detect movements. If installed close to the construction, deep benchmarks need to be carried below invert. They must be absolutely stable in spite of any ground movements that are occurring because it is the surface level collars of these devices that become the unmoving points from which locations and elevations of other instruments can be determined by surveying. A major complication in the installation of benchmarks can be the difficulty of installing them in a location and/or to a depth that absolutely guarantees no movement as tunneling proceeds. In this regard the lowering of groundwater in a soft ground environment can contribute to ground settlements well outside the immediate projected footprint of the advancing tunnel, so the instrument has to be well placed to guard against this eventuality. In cases of very large projects or overlapping projects that cause the water table to be drawn down across a large area, benchmarks have been known to settle even when founded in bedrock because some rock types can be dependent to a degree on pore water pressure for their ability to carry load.

Deep Benchmark

Figure 15-1 Deep Benchmark

15.2.2.2 Survey Points

Survey Points are used to detect ground movements at the surface or a few feet below the surface. They may be as simple as wooden stakes driven into the ground and their elevations surveyed through backsighting to a deep benchmark (Figure 15-2). Penetration needs to be at least a foot or so to guard against dislodgment, and the tops should not extend high enough to interfere with mowing machines if they are in a grassy area that requires routine maintenance. A survey point may also be somewhat more sophisticated and take the form of a steel rod with a rounded reference head driven several feet into the ground for better avoidance of possible dislodgment and surface effects such as frost heave (See Figure 15-3). This type of point needs to be protected at the surface by a small utility type roadbox with a secured cover so there is no disturbance to the rounded head. A rounded head is considered best because a surveyor can then always find the high point that has been surveyed in the past for good continuity in the readings. Because there is no hard connection between the rod and the roadbox - the one sort of "floats" inside the other - the survey point is also protected from being pushed down in case of the passage of a heavy vehicle. The major concerns with any type of survey point is the need to keep it out of the way of other users of the area and also protected against damage that may require replacement and lead to loss of continuity between the latest reading and the string of readings taken in the past.

Survey Point

Figure 15-2 Survey Point

Survey Point in Rigid Pavement Surface

Figure 15-3 Survey Point in Rigid Pavement Surface

15.2.2.3 Borros Points

A Borros Point is basically an anchor at the lower end of a driven pipe (See Figure 15-4). The anchor consists of three steel prongs housed within a short length of steel pipe with points emerging from slots in a conical drive point. Installation is achieved by advancing a borehole in soft ground to a few feet above the planned anchor depth and the anchor inserted by attaching extension lengths of riser pipe and outer pipe. When the point reaches the bottom of the hole, it is driven deeper by driving on the top of the outer pipe. The prongs are then ejected by driving on the riser while the prongs are released and the outer pipe bumped back a short distance to achieve a positive anchorage. Such installations are useful for determining the amount of settlement at one precise depth with more certainty than the simple driven steel rod described above, and they are relatively simple and economical. The amount of anchor movement is determined by surveying or otherwise measuring the movement of the inner riser pipe at the ground surface. One disadvantage with such movement detection (and this can be said of most instruments whose data depends on movements measured in a surface mounted reference head) is that, if settlement is great enough to have affected the surface at reading time, then the whole instrument may be moving downward by a certain amount while the anchor is moving downward by a greater amount. Absolute anchor movement may then be difficult to judge unless ground elevation surveys are undertaken at that time and the changes added to the apparent anchor movement.

Schematic of Borros Point

Figure 15-4 Schematic of Borros Point (After Dunnicliff, 1988, 1993)

15.2.2.4 Probe Extensometers

Probe Extensometers are used to measure the change in distance between two or more points within a drilled hole in soft ground, through use of a portable probe containing an electrical transducer. As shown in Figure 15-5, the probe, which contains a reed switch, is inserted into a casing in the drill hole in which the reference points, each of which contains an array of bar magnets, have been fixed in a way to surround the casing on the outside. In the most common type of installation, the reference points are held in place by spring loaded anchors - leaf springs - that "bite" into the ground. The points are free to move with the ground because the outer support casing will have been removed and replaced by grout. The probe detects the depth of the reference points for an indication of whether the soil at those depths is settling due to disturbance from construction. A probe extensometer can thus measure the settlements at a much larger number of depths than can a Borros Point. Probe extensometers are generally drilled to a depth below any potential zone of influence near a cut-and-cover or mined tunnel. The bottom reference point then becomes the unmoving reference from which the movements of the shallower points are judged. In a typical situation near a mined tunnel, it is likely that the lowest moving point will exhibit the most settlement, and that settlements will prove to be less as the probe moves up the casing to where the settlement trough is widening. One problem with probe extensometers is that collection of data can be operator sensitive as the instrument reader strains to detect the exact location of the probe at each reference point depth by listening for the electronic "beep" to ensure readings at precisely the same spot time after time. Another concern may be the time required for monitoring, especially if a large number of reference points have been installed, because the probe does have to be lowered to the bottom of the casing and then readings collected as it is slowly winched back to the surface.

Schematic of Probe Extensometer with Magnet/Reed Switch Tansducer, Installed in a Borehole

Figure 15-5 Schematic of Probe Extensometer with Magnet/Reed Switch Tansducer, Installed in a Borehole (After Dunnicliff, 1988, 1993)

15.2.2.5 Fixed Borehole Extensometers Installed from Ground Surface

Fixed Borehole Extensometers installed from ground surface may be used in soft ground or rock and may be Single Position (SPBX) for settlement measurements at one specific elevation or Multiple Position (MPBX) for measurements at several elevations. Figure 15-6 illustrates a schematic of an MPBX. The anchors of a borehole extensometer are grouted into the ground, commonly at various distances above the crown of an advancing tunnel, and connected to surface mounted reference heads by small diameter rods of steel or fiberglass. By detecting movement of the tops of the rods at the surface, one can tell how much each anchor - and hence its increment of soil or rock - is moving in response to excavation and so take steps to mitigate developing problems. Manual readings can be taken in a matter of minutes, assuming there is no problem with access to the instrument collar. However, automatic readings with an electrical transducer and datalogger - which can be salvaged/moved for use on other instruments - are relatively inexpensive and can provide real time data that feeds directly and quickly into a computer for fast analysis and databasing. Although extensometers oriented vertically over mined tunnel crowns are the most common installations, two others may prove useful in particular situations: (a) instruments angled in toward tunnel crowns or haunches from sidewalks where vertical installations are precluded by heavily travelled roads; and (b) instruments installed along the sidewalls of mined tunnels or cut-and-cover excavations where a knowledge of the vertical component of overall ground movement may be advantageous. A common problem with manually read instruments is the one of operator sensitivity, and if more than one reader is employed, they need to practice together to make certain they can monitor with good consistency. Remote monitoring leads to the concern that data collectors and analyzers may, without themselves personally having an eye on the construction operation, be unaware of the type and scheduling of activities that are affecting the data. Hence it may be necessary to make arrangements for construction progress reports to be delivered on a tighter schedule than otherwise might be necessary.

Multiple Position Borehole Extensometer Installed from Ground Surface

Figure 15-6 Multiple Position Borehole Extensometer Installed from Ground Surface

15.2.2.6 Fixed Borehole Extensometers Installed from Advancing Excavations

Fixed Borehole Extensometers installed from advancing excavations are a fairly obvious need if sidewall movements are required for a cut-and-cover excavation. Such horizontal installations are common and the drilling/installing operation has to mesh with the construction so that the larger operation is not overly impacted by what may appear to be a peripheral activity. (Note: "Horizontal" installations are seldom truly horizontal because angling downward by 10 or 15 degrees makes it much easier to manage the grouting of the anchors.) The installation of extensometers oriented from the vertical to the horizontal - including all angles in between - from inside advancing mined rock tunnels may be mandated by the lack of access from the ground surface (Figure 15-7). If possible, they are normally installed just behind a tunnel working face or the tail shield of a TBM. In this position they can provide data on incipient fallouts or more subtle rock movements toward the opening. If installed where a small tunnel is to be enlarged to greater size at a later time, the instrument heads can be recessed beyond the initial excavation outline and saved for use in monitoring the larger excavation. In this way they provide an almost complete history of rock movements from the earliest to the latest point in time. Another way to use these instruments is to install them from a first driven tunnel toward the location of a following twin tunnel. Readings then indicate whether the pillar between the two tunnels is loosening so that steps can be taken to mitigate the problem.

Horizontal Borehole Extensometer Installed from Advancing Excavation

Figure 15-7 Horizontal Borehole Extensometer Installed from Advancing Excavation

Complications for these in-tunnel instruments are more numerous than for those installed from somewhere outside the excavation. As noted, the installation has to be meshed with the construction operation, a particularly tricky proposition in the confines of a small mined tunnel, where constructor complaints of interference are extremely common. Even the collection of data, if it is performed manually, may be obtrusive, especially if cessation of tunneling, use of ladders, or help from constructor personnel are involved. Remote monitoring is also possible, but then there is electrical wiring to be run and the need to find a place for the datalogger(s) to be out of the way. By whatever method the in-tunnel instruments are monitored, the reference heads need to be protected, often by countersinking them in the tunnel wall and perhaps through installation of protective covers. This is especially true where there is going to be more blasting in the vicinity, but also true even where blasting is not involved. Miners tend to have little reverence for objects whose importance is not obvious to them, so vandalism and theft of instrument accoutrements has to be guarded against. Finally, there is the fact that an in-tunnel instrument is almost always installed after the tunneled ground has started to relax, so the initial readings are seldom true zero points from which to compute follow-on movements. The instrumentation specialist's only recourse is to continually press the constructor for access to install instruments at the earliest possible opportunity.

15.2.2.7 Telltales or Roof Monitors

Telltales or Roof Monitors (Figure 15-8) are other devices that can be installed from inside an advancing rock tunnel. They are designed to be installed with anchors in stable rock beyond the tips of rock bolts in tunnel roofs to provide fast feedback on stability. The immediate safety of the miners/tunnelers is the primary reason for the instrument's use. The devices were pioneered in French coal mines in the 1970s and further refined by the British and others in succeeding years. The first ones were steel rods with a single anchor and visual movement indicators in the tunnel roof that could be seen by miners as they worked. Simple and installable by rock bolting crews, they proved vulnerable to shearing due to movement of rock blocks and were eventually replaced by more flexible steel wires that are less prone to failure. Modern versions have as many as three anchors and can be wired for remote reading by a trained person watching the data on a laptop computer. Roof monitors are widely used around much of the world and are gaining acceptance in the U.S., where they deserve to join the ranks of commonly used instruments. They are now used in civil as well as mine construction and also in rock other than flat lying sedimentaries commonly associated with coal seams. As of this writing, the primary factor in considering use of roof monitors in the U.S. may be the need to educate tunnel designers and constructors in their efficacy and ease of use.

Triple Height Telltale or Roof Monitor

Figure 15-8 Triple Height Telltale or Roof Monitor

15.2.2.8 Heave Gages

Heave Gages are most commonly used when excavating for open cut or cut-and-cover in soft clay where there is potential for the bottom to fail by heaving as overburden load is removed. There are several instruments with which heave can be detected and measured, but almost all either suffer from lack of accuracy or are prone to damage or malfunction. Interestingly, the magnet/reed switch gage packaged as for a probe extensometer is probably the best alternative (Figure 15-9). In this type of installation the user measures increasing rather than decreasing distances between spider magnets and a fixed bottom anchor. With care taken to make certain the bottom anchor is well below any expected zone of movement, the installation is made inside the cofferdam prior to start of excavation. After initial readings are taken the access pipe is sealed 5 to 10 feet below the ground surface through use of an expanding plug set with an insertion tool, and the pipe is cut with an internal cutting tool just above the plug. A good fix is made on the plan location of the instrument and, just before the excavation reaches the plug, the pipe is located, a reading made, and the pipe again sealed and cut. The procedure is repeated until excavation is complete. The concern with such installations - a concern not overcome with alternative installation types - is that any large excavation is made by means of heavy equipment, and operators are not prone to watching and caring for things as small as a heave gage pipe. It is common for the gages to be damaged beyond use, and their protection can be assured only through some forceful construction management and sometimes the levying of penalties for instruments damaged as a result of contractor carelessness.

Heave Cage

Figure 15-9 Heave Cage

15.2.2.9 Conventional Inclinometers

As shown in Figure 15-10, conventional Inclinometers are aluminum or plastic casings drilled vertically to below the level of construction into a stable stratum and used to determine whether the surrounding ground, either rock or unconsolidated material, is moving laterally toward the excavation. Each casing has tracking grooves to guide the sensing probe for orientation both parallel to and at right angles to the axis of the excavation. The probe, which contains tilt sensors, is lowered on a graduated cable to the bottom of the hole and winched upward, with stops at 2-foot intervals for collection of inclination data by means of a readout unit at the ground surface. An iterative process of tilt calculations from the unmoving bottom of the casing permits plotting of a profile that fixes each measured increment of casing in space in relation to the excavation. An initial set of inclination readings is taken before excavation begins and each set of readings thereafter during construction provides data on how the ground is moving when the user plots the newer movement curves against the initial pre-construction curves. The inclinometers are normally situated a few feet from the excavation periphery of open cut or cut-and-cover excavations, but may also be installed just outside a mined tunnel where lateral movement data may be combined with vertical movement data from the extensometers discussed above. The term "conventional inclinometer" is used herein to distinguish the manually read instrument from the "in-place" instruments described below. The major concern with a conventional inclinometer is the time consumed in the monitoring process. Readings are performed twice in each monitoring visit, once with the probe inserted in the "A" direction tracking grooves, then again with the probe in the "B" direction. A "check sum" procedure is carried out by examining the sum of the two readings at the same depth, 180 degrees apart, in order to remove any long term drift of the transducers from the calculations. It commonly requires 45 or so minutes for a reader to collect data from a 100-foot deep instrument, and that is assuming no indication of excessive movements, which, if discovered, may require another set of readings for confirmation that the movements are real and not due to a reading error or instrument malfunction.

Principal of Conventional Inclinometer Operation

Figure 15-10 Principal of Conventional Inclinometer Operation (After Dunnicliff, 1988, 1993)

15.2.2.10 In-Place Inclinometers

In-Place Inclinometers are typically used for monitoring subsurface deformations around excavations when rapid monitoring is required or when instrumented locations are difficult to access for continued manual readings. The sensors are computer driven, gravity-sensing transducers joined in a string by articulated rods, and they can be installed equidistantly in the casing or concentrated in zones of expected movement (Figure 15-11). With the in-place instrument, as many as ten or twelve sensors are mounted in the casing and left semi-permanently in place. A larger number of sensors would be difficult to install in a standard size drill hole because each sensor has its own set of signal wires that take up space, and a very large number of sensors could result in the need for an uneconomically large diameter drill hole. Signals are fed to a datalogger at the surface and can be collected as often as required, or even fed by telephone line to the database computer for something close to real time monitoring. Compared with conventional instruments, the in-place inclinometer hardware is expensive and complex. This can sometimes be compensated to a degree by removing sensors from a bypassed instrument and installing them in a new location as the excavation progresses. A not-so-easily-overcome disadvantage of the in-place instrument lies in the fact that, if there is any long term drift in any of the sensors, it cannot be overcome through the check sums procedure described above. It is also true that the somewhat limited number of sensors in a standard in-place installation leads to a less smooth plot of movements compared with what can be achieved with the conventional inclinometer.

In-Place Inclinometer

Figure 15-11 In-Place Inclinometer

15.2.2.11 Convergence Gages

Convergence Gages may be used for monitoring closure of the ground across either open excavations or mined tunnels. In the first instance they perform a function similar to an inclinometer, although with many fewer data points to give a full picture of movements. In the second function, they detect the load redistribution during and after excavation and the extent to which resulting structure/ground interaction affects the tunnel shape and the lining. Until now the typical gage has been a Tape Extensometer, which includes a steel tape with holes punched at 50 mm intervals (see Figure 15-12). Anchors that define monitoring points consist of eyelets on the ends of grouted rebar sections that extend into the ground for a foot or so (Figure 15-13). The tension in the tape is controlled by a compression spring, and standardization of tension is achieved by rotating the collar until scribed lines are in alignment. After attachment of the extensometer to the anchors and standardization of the tension, readings of distances are made by adding the dial indicator reading to the tape reading. In a typical mined tunnel the pattern of anchors includes one in each sidewall at springline level and one as close as possible to the center of tunnel crown. Three readings are taken in a tent shaped pattern and the results indicate whether the tunnel support is behaving in a predictable way. For very large tunnels, the patterns may be more like trapezoids or overlapping triangles, which requires the installation of additional anchors. Such readings are only relative readings, and if absolute elevation changes are needed, this is usually accomplished by surveying the anchor that is in the crown. (Installation directly in the high point of the crown is seldom possible because of the presence of the ventilation and other lines.)

Tape Extensometer Typical Detail

Figure 15-12 Tape Extensometer Typical Detail

Typical Convergence Bolt Installation Arrangement

Figure 15-13 Typical Convergence Bolt Installation Arrangement

Whether the tunnel is conventionally mined or excavated by TBM, it is important to install anchors and begin readings at the earliest practicable time before the ground has begun to "work." Unfortunately, this cannot always be accomplished, especially in a TBM tunnel because, even if the anchors can be installed in a timely manner, there are scores or even hundreds of feet of trailing gear that make the stretching of a tape extensometer essentially impossible. This means that measurements may not begin until the machine is a long way past the monitoring point and knowledge of total from-the-beginning movements cannot be obtained. For this reason it seems likely that an alternative to the tape extensometer is going to be the best choice for future monitoring of tunnel convergence, and it will be in the form of a distometer. The device is small, hand held, and can be used to very accurately determine distances to a target by emitting a laser or infrared beam that is reflected from the target and detected by the same device. By installing brackets or bolts that also include targets at the places where tape extensometer eyelets would normally be placed, monitoring personnel can detect the changing shape of a tunnel without having to stretch a physical connection between points. There remains the problem that a physical object - such as TBM trailing gear - between targets will interfere with the distometer lines of sight and still not permit measurements in the standard tent shape. By judicious placement of additional brackets and targets at monitoring sections, it should be possible to gather data by working around the trailing gear in a TBM tunnel with patterns of measurements more like the afore mentioned trapezoids or overlapping triangles.

15.3 monitoring of existing structures

15.3.1 Purpose of Monitoring

If the different parts of a structure should move uniformly by even large amounts, damage could be minimal, maybe non-existent, except perhaps for penetrating utilities such as water pipes that might not be able to accommodate themselves to such movements. However, most structures affected by construction react by exhibiting more movement of the parts closest to the excavation than of the parts that are further away. This differential movement is the principal cause of construction related damages because the affected structure may be subjected to forces it was not designed for. A building, for example, whose footings are settling on one side while the other side settles less or not at all will suffer tilting of some walls, and the racking that ensues may cause cracking or spalling of some architectural features, freezing of doors and windows, or, in the worst case, failure of one or more of the structural members. A bridge whose footings are subjected to differential movements may undergo extensions that literally tear it apart. In general, the detection of settlements is the first line of defense in the protection of existing facilities, whether they be surface (roadways, buildings, bridges) or subsurface (utilities, transit tunnels, other highway tunnels). The detection of tilting can also be useful and has become more common as the development of monitoring devices has proceeded in the direction of increased automation. The simplest kind of monitoring involves the detection and the tracking of joint separations and crack propagation in structural concrete or architectural finishes. The ideal is to detect and mitigate some or all of these movements before they have become severe enough to cause serious damage or perhaps constitute a hazard.

15.3.2 Equipment, Applications, Limitations

As with ground movement instrumentation, there are a number of choices of instrumentation:

  • Deformation Monitoring Points
  • Structural Monitoring Points
  • Robotic Total Stations
  • Tiltmeters
  • Utility Monitoring Points
  • Horizontal Inclinometers
  • Liquid Level Gages
  • Tilt Sensors on Beams
  • Crack Gages

15.3.2.1 Deformation Monitoring Points

Deformation monitoring points on roads, streets or sidewalks can be as simple as paint marks that get surveyed on a routine basis. However, paint has the disadvantage that it can be visually obtrusive, may wear off with time, and may not display a single spot that surveyors can return to time after time for good data continuity. A better alternative is a small bolt-like devise set in an expansion sleeve that can be installed in a small hole drilled in concrete or asphalt as shown in Figure 5-14. The point should have a slightly protruding rounded head with a consistent high point that is always findable by a surveyor as he or she searches for the same unchanging spot on which to set the stadia rod. It is important that the point not protrude too much because it might then become a tripping hazard or be vulnerable to damage from equipment such as snow plows. Although they are inexpensive to purchase and install, the ultimate cost of deformation monitoring points can grow to become relatively high if data collection becomes intensive because it depends upon the mobilization of survey crews. Also, such monitoring is not always foolproof because surveyors are not necessarily attuned to the need for that high degree of accuracy that is sought by instrumentation specialists. It is very common for data thus generated to exhibit a fair amount of "flutter," i.e., apparent up-down movements that are not real, but are only the result of inconsistencies in the survey process. Such inconsistencies may result from the too-often changing of personnel in survey crews, changes that happen commonly due to the nature of the business. Luckily, extreme accuracy is not required in much of this paved surface monitoring, so if the surveyors can reliably detect changes of one-quarter inch or so, that is often good enough.

Deformation Monitoring Point in Masonry or Concrete Slab

Figure 15-14 Deformation Monitoring Point in Masonry or Concrete Slab

15.3.2.2 Structural Monitoring Points

Structural Monitoring Points are survey points that are placed directly on the structures of concern, most often being installed on a vertical wall of a building or a structural element of a bridge (See Figure 15-15). Except for buildings, most structures can accommodate the monitoring point likely to do the best job and the "points" may take several forms. The simplest is a tiny scratch mark that can be easily found on each monitoring visit by a survey crew. A similar point is a stick-on decal target, which is a bit more obtrusive, but easily removable once it is no longer needed. A problem with such surface treatments is that, for buildings particularly, the monitoring point may be only on a facade that moves independently of the underlying structural elements whose movements it is important to detect. This may be overcome by the installation of a bolt-like device that penetrates to the underlying structure for a truer indication of the movements taking place. The choice of monitoring points will often depend on the wishes of owners or managers of buildings who may object to the visual obtrusiveness or potential for damage from whatever may be installed. Possible damage can extend to the post-construction period when the monitoring point may have to be removed and patched, something that is often insisted on by the party who permitted its installation. Thus, it may be necessary to repair the scars left by the removal, which may entail the use of solvents, infilling, spackling, polishing, painting or replacement for satisfactory restoration.

Structure Monitoring Point in Vertical Masonry or Concrete Surface

Figure 15-15 Structure Monitoring Point in Vertical Masonry or Concrete Surface

A large consideration in the use of structural monitoring points is the need to depend upon surveyors for the collection of data. Compared with roads and sidewalks, most structures have tight specifications on permissible movements (a lower mitigation-triggering level of 1/4 inch being not unusual), so surveying generally needs to be of a somewhat higher order, not necessarily as stringent as Class I, but at least done with additional care. One way to achieve this is to hold briefings in which the importance of great accuracy is instilled in the surveyors who will do the work. Another (if it is possible in the economic climate of the day) is to write and enforce the survey contract so that each group of structures is always monitored by the same crew using exactly the same equipment. In this way, the "flutter" may be reduced so as to minimize the need for instrumentation interpreters to average the peaks and valleys in determining if settlements are real or only apparent.

15.3.2.3 Robotic Total Stations

Robotic Total Stations are used for obtaining almost real time data on movements in three dimensions when it is not feasible to continually mobilize survey crews to collect data. The operation of a Total Station instrument (theodolite) is based on an electronic distance meter (EDM), which uses electromagnetic energy to determine distances and angles with a small computer built directly into the instrument. Accuracy is generally much greater than that achievable with the use of classical optical surveying. Moreover, the equipment based on EDMs is capable of detecting target movements along all three possible plotting axes, the x, the y and the z. Total stations used in geotechnical and structural monitoring are electro-optical and use either lasers or infrared light as the signal generator.

Robotic (also called automated motorized) total stations are configured to sit atop small electric motors and to rotate about their axes. As shown in Figure 15-16, they are mounted semi-permanently and, at pre-determined intervals, automatically "wake up" to aim themselves at arrays of special glass target prisms (Figure 15-17) that can provide good return signals from a variety of angles. The target prisms, which are 2 to 3 inches in diameter, are installed on structures of concern and the total station instruments installed on other structures as much as 300 feet away. It is best to have the total stations installed outside the expected zone of influence for absolute certainty of measuring target movements with accuracy. However, it is standard practice to install some of the prisms definitely outside the influence zone so that they become reference points from which the total station can determine its own position and calculate the positions of the other prisms that may be subject to movement. Clear lines of sight from total station to target prisms are a requirement so that careful planning is required for proper placement. Data is recorded by means of the total station's own computer and may be fed to a centralized database computer by means of telephone lines or radio signal.

A major aspect of robotic total station use is the front end expense incurred. Depending upon the number purchased, the cost of top quality target prisms can range from $80 to $200 each in 2009 dollars. The total stations can cost from 30 to $40 thousand each, and they generally require the services of a specialist for the installation and maintenance. Nevertheless, for many projects where almost real time data on structural movements is necessary, this may be the only monitoring system capable of meeting all requirements.

Robotic Total Station Instrument

Figure 15-16 Robotic Total Station Instrument

Target Prism for Robotic Total Station

Figure 15-17 Target Prism for Robotic Total Station

15.3.2.4 Tiltmeters

Tiltmeters are used to measure the change in inclination of structural members such as floors, walls, support columns, abutments, and the like, which may tilt when the ground beneath is being lost into an advancing excavation. Manual tiltmeters generally consist of reference points on plates attached to the surface of interest and monitored by means of a portable readout unit, the functioning of which is based on an accelerometer transducer. Because such an arrangement can be operator sensitive and reading is somewhat labor intensive, especially where continued access is not easy, it is becoming more common to collect data remotely by means of electrically powered tiltmeters whose sensing elements may consist of accelerometer or electrolytic level transducers placed in housings that can be attached to the element to be monitored. If only one direction of movement is expected, the chosen instrument may be uni-axial, but if there is a possibility of combinations of movement, the bi-axial instrument would need to be used. Figure 15-18 illustrates a biaxial tiltmeter. Because tiltmeters can inform users only about rotational components of movement, data must be combined with that from other instruments to determine levels of settlement that may be affecting the structure. The most difficult tiltmeter installations are those required for structural elements somewhere inside a building that is occupied. Even the manually read instrument, with a flat 6 to 8-inch diameter plate being the part attached, is somewhat visually obtrusive and may be objected to by a building manager. Remotely read tiltmeters are even more obtrusive because they need to be wired for electric power and connected to a powered datalogger that will probably need to have telephone connections if true real time data is needed. There is some controversy within the monitoring community about the best installation height for these instruments, with some opting for lower floors and some for higher floors where absolute wall movement - though perhaps not tilt per se - will be greater. The argument is often laid to rest by a building manager who will permit such installations only in basement levels to better keep them out of the way.

Biaxial Tiltmeter

Figure 15-18 Biaxial Tiltmeter

15.3.2.5 Utility Monitoring Points

Utility Monitoring Points are very simple instruments used to determine whether an existing utility such as a water line is settling in response to an excavation proceeding nearby or underneath. The device consists of a small pipe with a rounded survey point or arrangement for use of a feeler gage at the upper end. This pipe is situated inside a larger piece of casing attached to a road box for surface protection. The lower end of the small pipe is attached to the top of the utility to be monitored and data collected by determining whether the top seems to be moving downward.

Unfortunately, such an instrument works well only if the monitored utility is exposed in a trench, and the inner pipe of the instrument attached before the utility is re-covered with backfill. When such an installation is attempted with a utility that is not exposed, one of two things may happen: (a) because the location of utilities is seldom known with absolute certainty, there is danger that the installing drillers may penetrate the utility, leading to a larger problem than the new tunnel under construction would have created; and (b) in the confines of a small drill hole it is extremely difficult to actually attach the monitoring pipe to the top of the utility, so it is possible for the utility to settle without there being an indication from the instrument of the movement's true severity.

In a case such as this, the best fallback position is to install a Borros Point (Figure 15-4) or an SPBX beside and to invert depth of the utility. If ground movement is observed at that location, itmay be an indication that excavation procedures need to be modified to contain a problem. Depending upon its size and stiffness, a utility may be able to bridge over a zone of disturbance and so be in no immediate danger, but ground settlement of a certain magnitude can be an indication that the movement needs to be arrested before it does become serious.

15.3.2.6 Horizontal Inclinometers

Horizontal Inclinometers are simply inclinometers turned on their sides and the transducers in the probe (conventional instrument) or sensors (in-place instrument) mounted such that the sensitive axes are perpendicular to the length of the pipe (Figure 15-19). In this way, an inclinometer is measuring the vertical rather than the lateral movements of the instrumented structure. One use for a horizontal inclinometer is in the determination of settlement of a utility along a reach that requires continuous data not producible by the utility monitoring points or extensometers described above. Due to difficulty of continuous access for monitoring, such an inclinometer installation is more likely to entail an in-place instrument that can be remotely read, but even here access may pose at least a minor challenge. If the utility is large and the flow of contained liquids can be controlled, then inclinometer casing may be strung and attached to the roof inside the instrumented structure. If the utility is too small for entry or the liquids cannot be controlled, then it would need to be exposed in a trench for instrument attachment to the outside and then backfilled. In either case, arrangements would be made for wiring to be run to a datalogger for essentially real time monitoring. Difficulty of access for installation is an obvious drawback, but when the need for monitoring is over, it should always be possible to salvage the expensive sensors for re-use.

If entry into the utility were possible for installation, then it should also be possible for recovery efforts. If the instrument were installed and then covered over by backfill, a small manhole will have been provided for access to the reference head and the wiring, and it is from here that the sensors and their attached wires can be removed.

Horizontal In-Place Inclinometer

Note: Surface Protection to consist of round casing with lid and be firmly attached to inclinometer casing so that neither can move independently.

Figure 15-19 Horizontal In-Place Inclinometer

15.3.2.7 Liquid Level Gages

Liquid Level Gages are systems of sensors installed in an array that measures the height of a column of water within each gage as shown in Figure 15-20. Sensor gages are connected by small 1/4 to 1-inch diameter tubes or pipes to a reference gage outside the zone of influence. The reference gage is actually a reservoir, with its contained liquid generally kept under pressure to avoid the undesirable effects of barometric changes. The liquid completely fills all of the tubes throughout the array of components, none of the liquid is exposed to outside atmosphere, and so it is referred to as a closed pressurized system. With the liquid always at the same elevation, settlements of the instrumented locations are indicated as the heights of the columns of water within the gages change in relation to the gage housings, which are moving. Signal outputs are most commonly driven by LVDTs (see description under electrical crack gages below) or vibrating wire (see surface mounted strain gages under 15.4.2) force transducers. The closed systems are small and flexible and can be configured to fit into the convoluted layouts of many instrumented structures. Readings are collected remotely through wiring of the system to a datalogger. Such systems are commonly installed in or on a structure where continuous settlement measurement to an accuracy of several millimeters is needed and where continued access for maintenance is not a large problem.

Multipoint Closed Liquid Level System

Figure 15-20 Multipoint Closed Liquid Level System

Maintenance visits are a must with these systems, and so the issue of access has to be taken seriously. During installation, which must be performed with great care, the system has to be charged with de-aired water and then purged to make certain no air bubbles have intruded to remain within it. This is one reason most installations utilize some kind of semi transparent plastic tubing; it permits visual detection of bubbles and makes purging them easier. This is critical because air bubbles will migrate to high points in the tubing or to the sensors themselves and can cause readings to be very inaccurate or can even shut down the system altogether. Then, during operation, it is very common for bubbles to appear in spite of careful installation. This may occur due to leakage from the outside, tiny amounts of air coming out of solution and accumulating, etc. Interestingly, the pressurization of the system can inhibit the emergence of bubbles, but never stop it entirely. No closed system is immune to this problem and maintenance visits may be required for purging and de-airing as often as every 6 to 8 weeks. This is why continued access can be so important to the closed pressurized system's functionality.

The maintenance problem can be largely overcome through the use of an open channel system which consists of sensors connected by pipes that are only half filled with water as shown in Figure 15-21. Open to the atmosphere, neither the liquid nor the sensors are affected by the problem of air bubbles. They can be installed to lengths of several thousand feet, operate for many months with hardly any maintenance, and still detect movements to sub-millimeter accuracy. However, such systems are large, heavy (due to the piping), sometimes difficult to install in structures with complicated layouts, and are much higher in front end costs than the smaller closed systems. At present, only a few open channel systems have been installed in the U.S. and only one or two corporate entities have expertise in their manufacture and installation. It seems likely that they will have a much larger presence in the future if downsizing of the components can lower purchase prices and make installations faster and easier.

Open Channel Liquid Level System

Figure 15-21 Open Channel Liquid Level System

15.3.2.8 Tilt Sensors on Beams

Tilt Sensors on Beams, when packaged to monitor elevation changes rather than tilt per se, consist of sensors attached to metallic rods or beams, with the beams linked together with pivots (Figure 15-22). By monitoring changing tilt of each sensor and knowing the length of each +/- 5-foot long beam, users can calculate elevation changes of each pivot with respect to the datum. The relative tilt of each sensor and beam is set in the field and elevation change data determined by making an initial scan of readings, called the reference set, and mathematically subtracting readings in that scan from each subsequent scan. All elevation change data is referenced to one end of the system defined as the datum. Ideally, the datum is in a stable area not likely to move, and its absolute elevation is generally determined by an initial optical survey. Integrating the data is an iterative process as settlements are computed from sensor to sensor. Readings are collected by having the system connected with a datalogger for almost real time monitoring.

Schematic of Electrolytic Level Tilt Sensor

Figure 15-22 Schematic of Electrolytic Level Tilt Sensor (After Dunnicliff, 1988, 1993)

Such installations can work on bridges, the balustrades of buildings, the walls or safety walks of existing tunnels, or even railroad tracks. However, they do depend upon sensing of the mechanical movements of a string of components, and the components need to be as free from interference as possible. If installed where workers or moving equipment may be present, they have to be protected by installation of metallic housings or half rounds of heavy plastic casing. Another potential problem stems from changing temperatures, especially in the outdoors where there may be exposure to severe or very changeable weather. Although the sensors may fare as well as they would in any other type of installation, such as in a tiltmeter housing, the beams and the pivots are metal and subject to thermal effects with the potential to skew the data in unexpected ways. Users need to be aware that, if even one sensor or sensor/beam combination fails for any reason and requires replacement, the whole string of sensors and beams will need to be reinitialized.

15.3.2.9 Crack Gages

Crack Gages (also sometimes called Jointmeters) as installed on structures are typically used for monitoring cracks in concrete or plaster, or for determining whether movement across joints is exceeding a structure's design limits. The first appearance of cracks can be an indication of structural distress, and their growth, either in width or length, can be an indication that stress is increasing, as can the continued widening of an expansion joint. There are several ways of measuring these movements; only the two most common can be covered herein.

As shown in Figure 15-23, a Grid Crack Gage consists of two overlapping transparent plastic plates, one installed on each side of the discontinuity and held in place with epoxy or mounting screws. Crossed cursor lines on the upper plate overlay a graduated grid on the lower plate. Movement is determined by observing the position of the cross on the upper plate with respect to the grid. Data is kept in notebooks and has to be keypunched into a computer if needed for an electronic database. Such gages are inexpensive to purchase and install, but readings may vary with changes in monitoring personnel and this has to be guarded against. There are three circumstances in which such simple devices may prove inadequate: (a) where cracks are too narrow or are widening too slowly for the human eye to detect their growth; (b) where continued physical access is very difficult and remote monitoring is required; and (c) where something close to real time monitoring is required. Such difficulties may be overcome through the selection and installation of Electrical Crack Gages as shown in Figure 15-24.

Grid Crack Gauge

Figure 15-23 Grid Crack Gauge

Electrical Crack Gauge

Figure 15-24 Electrical Crack Gauge

There are a number of electrical gage types, but most are based on an arrangement of pins attached on opposite sides of a joint or crack, with the pins connected by sliding extension rods whose differential movements are detected by a built-in transducer. The most common transducer is the linear variable displacement transformer (LVDT) that consists of a movable magnetic core passing through one primary and two secondary coils. Data readouts depend upon detection and measurement of differences between voltages generated in the secondary coils, magnitudes of which depend on the proximity of the moving magnetic core to the secondary coils. Users may prefer to pick up the gage signals by using a small low power radio transmitter installed at the instrument location to avoid the transmission of alternating currents through long lead wires that can introduce output-degrading cable effects.

<< Previous Contents Next >>
Updated: 06/19/2013
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000