 We live in the age of the automobile, an age where despite energy shortages, rising operating costs, and mass transport options, over 124 million passenger vehicles alone are registered in America, each averaging over 9,000 miles of travel annually. This ever-increasing volume of traffic has stimulated various methods of traffic management to minimize congestion, driving costs, and driver frustration. Today, through the use of modern vehicle detection devices combined with computer technologies, it is possible to monitor vehicle speeds, to record traffic volumes, to create elaborate signal systems, to display various warning signs, and even to re-root traffic away from problem areas. Vehicle detectors originally evolved for traffic signal control in the late 1920s, as increasing traffic volumes made it apparent that fixed-time signals had limitations. Fixed-time signal controllers are mechanical devices with a fixed sequence of right-of-way assignments. Although these were once the standard for signal control, their inability to respond to traffic flow and direction soon proved wasteful and unsatisfactory. In 1928, the first vehicle detector was installed at an intersection in Baltimore. A small microphone mounted on a lamp post activated a signal switch when a car horn sounded. Although this may have been convenient for the waiting driver, it wasn't too popular with the neighbors. Various other detector systems have evolved since that crude beginning, such as pressure-sensitive, radar, magnetic, magnetometer, sonic, and self-powered systems. Currently, the most widely used type is the inductive loop detector. This consists of one or more loops of wire wound into slots cut into the pavement. These loops are connected by a lead-in cable to a detector unit that serves as an AC power source. The combined lengths of the loop wires and lead-in cable dictate a certain natural frequency of vibration for that system. The power source is adjusted to match this resonant frequency, creating a tuned circuit. Current flowing through the wire creates an electromagnetic field around the loop. This field can be illustrated as lines of force or flux surrounding the wire, which contain energy. When a vehicle enters this field, eddy currents are induced into the metal of the body and frame, reducing the lines of force, a kind of absorption of energy from the loop. This absorption causes a decrease in the self-inductance of the loop, which in turn raises the frequency at which the loop will resonate. Some detector units respond to this by means of feedback circuitry that increases their frequency of oscillation. Others respond to the phase shift between the loop oscillation and the reference oscillation. Either of these response designs activates an output relay in the detector unit, which sends a signal to a controller indicating that a vehicle is over the loop. Because of its flexibility in application, relatively low cost, and comparatively safe design, the inductive loop is the most common detector now in use. Recently, much research has been done by various agencies toward extending the lifespan of these detectors. In 1981, a project was undertaken by the Engineering Research and Development Bureau of the New York State Department of Transportation to study inductive loop detector failures. The study's purpose was to identify how and why these detectors fail, and how to prevent or reduce these failures in order to improve their life expectancy. The first phase of the project included a field survey of over 300 failed loops. Failure types were documented in the field and then analyzed by a statistically based computer program. In most instances, failures could be attributed to one or more of the following factors, design and installation methods, sealer reliability, and wire durability. The formulation of solutions to these problems became the major areas of study and evaluation in the second phase of the project. The question of loop placement was examined first. Because this system includes wire loops only inches below the road surface and lead-in cables sometimes extending hundreds of feet, it is vulnerable to construction and repair of the roadway, walkways, curbing, and various utilities which often damage or destroy a loop system. Before the actual installation, the work site should be inspected to determine the location of utility lines which should be avoided. A check for scheduled construction in the work site area could also prevent these types of conflicts. After the exact installation site is chosen and proper safety measures are taken, detector installations should begin by marking the loops outline on the pavement surface with chalk line, spray paint, or any other marking material. This outline is then used as a guide for sawing the slots. Formerly, loop patterns in New York State were cut according to this standard. Diagonal saw cuts at the corners were intended to limit the sharpness of bends in the installed wire. But pavement sections at these corners were subject to breaking and popping out, thus leading to exposed wire and loop failure. In March of 1983, a new loop design was approved in New York State, in which the corners are cut square and then rounded. This new pattern decreases corner-related failures and at the same time requires less sawing. Masonry saws for loop installations are available in a range of sizes and sawing capabilities. This is an 18 horsepower saw with a diamond blade in a wet cutting operation. This was found to be a most efficient method. An 18 horsepower saw is preferable to the widely used 9.5 horsepower model. The smaller model is slower and breaks down frequently due to excessive workload. A variety of saw blades can also be used in loop installation, but the diamond blade is considerably faster, neater and perhaps safer than traditional dry cutting abrasive blades. There are two acceptable methods to prepare corners in the new loop design. One is to remove a core to the full depth of the sawed slot from each corner and then round all rough and sharp edges. The other more popular method is to use a small air hammer to chip back the sharp edge, making a smooth curved surface for the signal wire to bend around. When the corners are completed, the saw slot must be cleaned in preparation for sealant installation. In the past, this was done by blowing compressed air into the slot. This method was found ineffective in slot cleaning because it left fine debris dried on the inside of the slot. This debris prevented proper bonding of the embedding sealer to the inside of the slot and eventually led to detector failure. One effective way to clean the slot is to use this hydroblasting system. This system has a nozzle that, by using the Venturi principle, combines water with compressed air to give a hydroblasting effect sufficient to clear away any debris left in the slot. Once clean, the slot must be dried completely before wire installation. This can be done by using the same instrument with the water supply disconnected. Based on the failure survey, a large percentage of failures could be prevented by using a more durable wire. Most failed loop systems studied were constructed according to a specification calling for 14 gauge stranded copper signal cable with a polyethylene insulation. This wire could not survive the stresses caused by even minor pavement disruptions such as creeping, cracking and thermal expansion and contraction. Nor could it withstand exposure to water and debris from minor sealant failure. It was found that the use of 14 gauge stranded copper wire, loosely encased in a flexible vinyl tube, had been very successful in other state's loop detector installations. With the loop wire loosely encased, it is free to move in the tube and can compensate for pavement disruptions by distributing tension throughout the entire length of the wire. Encasement also provides additional protection from water and abrasive debris. The incorporation of tube encased signal wire was carefully evaluated. It was found that a wider and deeper slot is necessary to accommodate this type of wire, resulting in increased saw cutting and sealant costs. There is also an increase in the cost of the wire itself, but increased material and installation costs are greatly offset by the resulting extended loop life. Once the proper number of wire loops are in the slot, the wire's two ends are wound together for the specified number of twists per foot. They are then threaded through a conduit into a pull box, where an electrical inspection is performed and recorded. This includes measurements for leakage to ground, loop resistance, correct inductance and induced voltage. Wire splicing in a loop system is done basically in two steps. The first, physically connecting the wire ends, can be done by soldering, a screw and nut, or a crimping method shown here. The second step is to environmentally seal the splice by either applying several layers of various re-insulating materials, or by encapsulating the spliced area in a container filled with embedding sealant. Another cause for loop failure is called wire float. When the embedding sealant is soft and plastic, the wire has a tendency to float up to the surface, exposing itself to the abrasive effects of traffic. Since some sealants remain soft, this can also happen after the sealant's curing, depending on the product used. To avoid this, various materials can be used to hold down the loop wire. Hold down should be placed at all corners, wherever the loop wire changes direction, and spaced at two-foot intervals around the entire saw slot. A material well suited for this purpose is an open-celled foam rubber backer rod. This material is readily available, easy to work with, and is unaffected by most sealants or their installation temperatures. When using this one-half-inch diameter backer rod, one-inch-long strips are torn off and pushed into the three-eighth-inch wide slot. Another major problem area is embedding sealant failure, which includes cracking, debonding, shrinking, tracking, and running out of the slot, or poor encapsulation of wire, all of which lead to loop failure. Bituminous-based materials, or hot tires, have been popular in the past because they were relatively inexpensive and easy to obtain. However, several problems exist with use of these materials. Most bituminous sealers must be heated to a pourable temperature, sometimes exceeding the melting point of the wire insulation, which, when damaged, will lose its protective properties. Working with this heated material has proved dangerous to both workers and nearby pedestrians. Hot tires also become soft in warm weather and can be tracked out of the slot or contaminated by dirt and debris. This makes it necessary to clean and reseal the slot regularly. A laboratory testing and evaluation program was developed by the Materials Bureau to produce specifications for acceptable sealants. Sealants were tested for curing time, water absorption, pot life, wire encapsulation, tensile strength, and adhesion. Based on this testing and extensive field evaluations, the best sealants for this purpose were found to be cold-poured epoxy or epoxy resin materials. They have excellent bonding capabilities, are durable, and resist effects of weather and provide good encapsulation of the loop wire. These cold-poured sealants also diminish installation hazards and increase detector cost-effectiveness by reducing periodic maintenance. A V-shaped squeegee can be used to scrape excess sealant into the slot. This reduces sealant waste and leaves the cured sealant flush with the road surface, or even slightly below the surface where it is better protected from traffic. After sealing the slot, a light dusting of Portland cement may be added to prevent tracking out if the surface of the cured sealant is still tacky. This completes the installation. Through use of these installation techniques and similar materials, the dependability of inductive loop detectors has been much improved in other states, such as Illinois and Connecticut. They both formally experienced an annual replacement rate of about 20% as did New York State. After several years of using these improved installation techniques, their current annual rate of replacement has fallen to nearly 1%. If New York State accomplishes a similar improvement, annual replacement costs of over 3400 of these detectors may be saved. This could translate into about $7 million of maintenance funds. The motoring public will also benefit directly from safer, more efficient travel with better inductive loop detectors.