Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
24 CHAPTER THREE TESTING AND RESEARCH LABORATORY TESTING At the request of the New England Transportation Consor- tium, the U.S. Army Cold Regions Research and Engineer- ing Laboratory conducted laboratory studies to develop standardized procedures for the evaluation of bridge deck membranes (24). They reported that although there are ASTM tests to evaluate various engineering properties of tape, rubber, roofing, plastics, and geomembranes, there is no group of standards or ways to interpret them that all manufacturers follow when reporting performance data for their products. The intent of the work was to recom- mend tests to compare membranes. Six sheet products were tested to measure adhesion, tensile strength and elon- gation, puncture resistance, and water vapor permeability. Liquid membranes were not included in the scope of the study. Conclusions based on testing and analyses included the following: â¢ A membrane does not have to be perfectly adhered to the deck to avoid blistering. â¢ High bond strength matters less than continuity of bond. â¢ The smallest void size that can originate a blister is about the size of a quarter. â¢ Elongation rather than strength is a more appropriate property to judge a membraneâs ability to span a crack. â¢ Puncture resistance is an important property of a good sheet membrane. â¢ ASTM E96 Procedure B (Water Method) is an accept- able method to measure water vapor permeability. The authors stated that even though laboratory tests can help rank membranes according to individual properties, exposure to the complex combination of natural forces is essential for proving a materialâs durability. The European Organisation for Technical Approvals has a report that describes a method for determining the resis- tance of liquid-applied bridge deck waterproofing member to chloride ion penetration following the indentation of the membrane by simulated hot asphalt (25). In this method, three heated concrete blocks with the membrane applied are indented at four locations using a heated 8-mm (0.3-in.) diameter truncated cone applied at a specified rate until a maximum force is applied. The surface of the membrane is then exposed to a saturated sodium chloride solution for 28 days. A sample of the concrete directly below the membrane is then obtained from each block and chloride ion concentra- tion determined. The measured chloride ion concentration is then compared with the background chloride ion concentra- tion of the reference concrete block. EVALUATING FIELD INSTALLATIONS Manning (5) described various methods to evaluate water- proofing systems in the field, including visual inspection, electrical methods, embedded devices, physical sampling, ultrasonic methods, and air permeability methods. These same techniques still exist today, though many have been improved through the use of electronics and automation to make them more practical to use on large areas of bridge decks. One of the challenges of detecting defects is that the defect has to be large enough to be detected using the selected method. If the defect is small, it is like looking for a needle in a haystack. If the defect is large, it may be detected by visual observation of surface defects such as delaminations or water leakage through the deck. Seven agencies responding to the survey reported that they had used the following nondestructive test methods to assess the condition of the in-place waterproofing systems: â¢ Visual inspection, â¢ Electrical conductivity or electrical resistance, â¢ Ground-penetrating radar (GPR), â¢ Chain drag or hammer sounding, and â¢ Leak testing. Visual Inspection Visual inspection requires observation of the top and bottom surfaces of the bridge deck from a relatively close position, such as walking on the deck surface. With this method, the condition of the membrane cannot be directly observed. The most direct method would be observation of the deck under- side after a period of rain to check for wet spots or efflores- cence. Rust stains or spalled concrete may also be evident, but by the time these are visible, active corrosion has been ongoing for some time.
25 In Denmark, where more than 85% of the bridge deck area has a bitumen overlay and waterproofing membrane, invasive inspections are sometimes performed on bridge decks. An area of wearing course and membrane approxi- mately 0.8 x 0.8 m (30 x 30 in.) is removed so the condi- tion of the structural concrete deck can be inspected (15). A similar procedure is followed in Sweden when deterioration is observed at the deck surface. Visual inspection of the asphalt surface may offer some indications of the condition of the membrane. Wide cracks, radial crack patterns, wet spots, and gaps at curbs or barriers may be signs of potential problems. Electrical Methods Virginiaâs standard specifications require that the water- proofing effectiveness of the membrane system be deter- mined in accordance with Virginia Test Method T 39. In this test method, the electrical resistance between the top surface of the asphalt and the top mat of reinforcement is determined using an ohmmeter. The specification requires a minimum resistance of 500,000 ohms. Areas having a lower resistance are to be repaired if determined by the engineer to be detri- mental to the effectiveness of the system. If more than 30% of the deck area is determined to be detrimental to the effec- tiveness of the system, the membrane is to be replaced. Washington State has a similar procedure, Test Method T 413. The scope of the method indicates that it may be used for either membrane alone or membrane-pavement combina- tion. The use of the method has been discontinued because of difficulty in training staff to use it and because membranes rarely failed the test. Interestingly, McKeel (26) commented on Virginiaâs T 39 method that a great deal of judgment is necessary to perform the test and it is advisable to use the same crew as much as possible. Manning (5) also points out that low resistivity readings are not necessarily associated with defects in the membrane but may be the result of mois- ture in the surface layers. ASTM D3633, Standard Test Method for Electrical Resis- tivity of Membrane-Pavement Systems, is similar to Virgin- iaâs T 39 method and Washington Stateâs T 413 method and may be used to measure the electrical resistance between the saturated top surface of the system and the reinforcing steel embedded in the concrete. Ground-Penetrating Radar GPR consists of transmitting pulses of radio frequency energy into the deck and recording the reflected signal. Reflections occur from each interface where there is a change in the dielectric constant, such as at voids, cracks, or steel reinforcement. The use of GPR for evaluating sub- surface conditions was the subject of NCHRP Synthesis 255 (27). That synthesis reported that GPR is a noninvasive and nondestructive tool that has been used successfully in trans- portation structures for applications such as profiling asphalt thickness, detecting air-filled voids, and determining rein- forcement spacing and depths in concrete. However, no pub- lished papers about the use of GPR to evaluate waterproofing membranes were identified for this synthesis. Kansas reported on the use of GPR on a bridge with a waterproofing membrane. Based on the results, Kansas decided to rehabilitate the bridge deck. The deterioration levels found in the concrete during the rehabilitation work were much higher than expected, and near full-depth patch- ing was needed throughout most of the deck. The final reha- bilitation cost was almost as high as the estimated cost for complete deck replacement. Chain Drag and Hammer Soundings Chain drag and hammer soundings are simple techniques to detect delaminations in bridge decks. In both methods, the change in sound from dragging chains across a deck or striking a local area with a hammer is used to identify areas of delaminations. The method is labor-intensive and is not foolproof. Leak Testing Leak testing involves ponding the deck top surface with water and checking underneath for leaks. This method may not be feasible on some bridge decks owing to longitudinal or transverse slopes. Oregon requires leak testing as soon as the deck is ready for traffic. No water leakage is allowed. Missouri reported that it has recently started to do leak testing on newly constructed adjacent box beam bridges. In some instances, it has flooded the deck before waterproof- ing to establish which joints leak or after the membrane and asphalt overlay have been placed and prior to bridge open- ing. Missouri reported that the best way to perform the test is during a rainstorm. This approach does not delay the project or impact traffic but is dependent on the weather. Bond Testing The New Hampshire special provisions for liquid-spray bar- rier membranes requires that the prepared substrate and the completed membrane be tested for adequate tensile bond strength in accordance with ASTM D4541, Standard Test Method for Pull-Off Strength of Coatings Using Portable Adhesion Testers. At least one test is specified for every 55 yd2 of deck area, with a minimum of three tests per structure or deck construction phase. When the bond strength of the substrate is less than 100 psi, the engineer may request addi- tional surface preparation. Illinois has a similar specifica- tion and specifies a minimum tensile adhesion value of 100
26 stick barrier membranes was discontinued and replaced with welded-by-torch and spray-applied liquid membrane systems because of the latterâs higher adhesion strengths. According to the survey for this synthesis, NHDOT has used heat-applied barrier membranes on essentially every bridge deck since 2000. Although spray-applied barrier membranes are still permitted, contractors provide the heat-applied systems because of their lower initial cost. Various installation methods were also studied, with the following conclusions: â¢ Air blast versus abrasive blast surface preparation showed no effect on bond. â¢ Rolling versus brooming of preformed membranes showed no substantial benefit of either method. Alaska DOT performed field evaluations of selected bridges to determine whether the waterproofing membrane was bonded to the concrete bridge deck and the asphalt over- lay (32). The project was initiated because some of the pre- formed membranes, generally on high-traffic volume roads, had failed to bond adequately to either the asphalt overlay or the concrete bridge deck. Five proprietary products were included in the evaluation. Concrete cores were taken from three bridge decks to inspect for bonding. Bonding between the membrane and the concrete or the asphalt overlay was observed in all but one core, although no measurements of bond strength were reported. Separate pull-out tests using similar procedures to ASTM C900, Standard Test Method for Pullout Strength of Hardened Concrete, were used to determine the tensile bond strength between the asphalt overlay and the membrane. Based on the reported loads, the bond stresses ranged from 22 to 112 psi, with higher asphalt temperatures giving higher bond strengths. One recommendation from the research was to require a 4-in. thickness of pavement over the membrane to allow for future pavement surface rehabilitation without damaging the existing membrane (33). Research about tack coats for use with asphalt is being performed under NCHRP Project 09-40, âOptimization of Tack Coat for HMA Placement.â The objectives of this study are to determine optimum application methods, equip- ment type and calibration procedures, application rates, and asphalt binder materials for the various uses of tack coats and to recommend revisions to relevant AASHTO methods and practices related to tack coats. Bond tests are expected to be recommended. psi with failure in the concrete. Testing is performed using samples of the membrane before installation of the com- plete membrane. Testing of the installed membrane is not specified. Illinois requires and New Hampshire may require holiday testing of the liquid membrane in accordance with ASTM D4787, Standard Practice for Continuity Verification of Liquid or Sheet Linings Applied to Concrete Substrates. The New York State DOT special specification for spray- applied waterproofing membranes also requires testing of the substrate after the primer has been applied and after the mem- brane has been installed. A minimum adhesion of 1 MPa (145 psi) is specified for portland cement concrete decks. New Jersey requires testing of the adhesion between the primer and the substrate in accordance with ASTM D4541, Standard Test Method for Pull-Off Strength of Coatings Using Portable Adhesion Testers, at a frequency of at least three tests for areas less than 5,000 ft2 and one test for every 3,000 ft2 for areas more than 5,000 ft2 and at locations where deficient adhesion is suspected. Infrared Thermography Infrared thermography senses the emission of thermal radia- tion and produces a visual image from the thermal signal. It has the potential to identify defects in waterproofing mem- branes because it permits large areas to be surveyed in a short time. Its disadvantage is the requirement for the appropriate environmental conditions to achieve the heat flow conditions to detect the presence of anomalies. Thermography has the ability to detect blisters in waterproofing membranes (28), delaminations in bridge decks (29), and defects after the installation of waterproofing membranes (30). RECENT RESEARCH In 1996 and 1997, the New Hampshire Department of Trans- portation (NHDOT) evaluated various membrane materials, primers, and application methods to determine the effects of materials and methods on the adhesion strength of com- mercially available membranes (31). Concrete pads simulat- ing dry and wet substrates as typically encountered on New Hampshire bridge decks were constructed at two locations. The test program included 11 preformed membranes, 5 liquid membranes, and 14 primers in various combinations. The pri- mary method of evaluating the systems was adhesion testing. The study findings resulted in a change in NHDOT spec- ifications in 1998. The use of standard preformed peel-and-