Surface penetrating radar for condition assessment of concrete masonry structures

Surface penetrating radar (SPR), also known as Ground penetrating radar (GPR), is a non-destructive technique that is well suited for use in condition assessment, repair, and rehabilitation of concrete masonry structures.

Structural details of interest to engineers, architects, and contractors may be determined quickly using SPR, such as masonry unit thickness, location of grouted and ungrouted cells, placement of horizontal and vertical reinforcement, splice lengths, and other internal features.This information can be used for quality assurance purposes in new construction, for use in rehabilitation or seismic upgrade of structures, or to determine as-built conditions when detailed drawings are not available.

SPR has several advantages compared with other types of non-destructive test (NDT) methods, especially when testing must be performed within existing structures. Advantages include the ability to test surfaces covered by carpet, tile, or plaster, or where access is limited to only one side of the object. In addition, radar equipment is portable, poses no radiation hazard, and the output can be viewed on a computer screen in real-time. These features permit rapid testing with little or no disruption to the surrounding area.

SPR background Radar is an acronym for radio detecting and ranging. Electromagnetic energy, such as radio waves, travel at the speed of light through air. When the radar waves encounter other materials, the speed of propagation is reduced, and some of the energy is reflected from the boundary between the substances.

In a concrete mason ry wall, for example, these reflections may occur at the surface of embedded rebar or conduits, voided cells or head joints, or at the interface with the air behind the wall.

The speed of the electromagnetic wave through a given material is related to a property known as the dielectric constant.

However, radar waves can propagate only through non-conductive (i.e. dielectric) materials. Steel and metallic objects act as perfect reflectors of radar waves, while salt water and other conductive materials tend to absorb mu ch of the incident energy.

Interpretation of SPR data—SPR systems commonly used in structural investigations consist of four main elements: a signal processor, a display screen, an antenna/receiver, and a survey wheel device.

The system operates by detecting reflections of the input wave pulses that are transmitted by the antenna. As these time-domain waveforms return to the surface, they are received by the antenna and digitally processed by a computer. The resulting data is sent to an LCD (liquid crystal display) screen where it is displayed in real-time. The picture produced by a series of radar signals, or scans, represents a two-dimensional cross-section of the material, with depth measured on the vertical axis and distance traveled by the antenna indicated on the horizontal axis, which is accurately measured by the survey wheel. Identifying the objects or interfaces that cause these reflect ions is done by visually interpreting the radar signals based upon the amplitude of the reflected signal, phase of the reflected signal, and patterns produced by a series of scans.

SPR output often is confusing to an inexperienced observer because the resulting display image does not look like a normal picture. This is because SPR output typically is displayed in a line scan format. In a line scan display, each reflected radar signal, or scan, is represented by a vertical line on the screen. The amplitude and phase of the signal at each point on the line are represented by a shade of gray (although color transforms also can be used). For the gray scale color transform used in Figures 1 and 2, high-amplitude, negative phase reflections are shown in black, while high-amplitude, positive phase waves are shown in white; low-amplitude signals are shown in various shades of gray.

Proper interpretation of SPR data is based on visually identifying various shapes and patterns that form recognizable signatures in the output. Discrete features, such as rebar, conduit, and pipes, show up as hyperbolas, or arch shapes, on the data record. This pattern occurs as the result of the spread of the radar waves in front of and behind the centerline of the transmitting antenna. This distortion does not occur with a planar feature, and they appear as a linear band in the data record.

Objects with a geometry that is neither cylindri cal nor planar may generate a pattern that is not easy to recognize. For this reason, irregular voids, such as honeycombing, can be difficult to detect accurately using visual interpretation alone.

Evaluation of concrete masonry structures

Proper interpretation of features identified using SPR is highly dependent on the knowledge of the operator, who must be familiar with not only the SPR equipment but also typical concrete masonry construction details.

For example, antennas typically used for structural investigations are designed to detect cylindrical objects and are most effective when operated in a direction perpendicular to the long axis of the desired target.

Therefore, for a typical concrete masonry wall with vertical main reinforcement, scans are performed horizontally along each course to identify the location of the rebar. Horizontal joint reinforcement also can be detected by scanning in a vertical direction. Production rates on the order of 150 square feet (13.9 square meters) per hour can be achieved for typical concrete masonry wall surveys—this includes scanning each course and identifying all hollow, grouted, and reinforced cells.

Figures 1 and 2 were created from data collected in the field and show how various phys i cal features can be identified from the SPR output. All scans were performed using a 1,500-megahertz (MHz) antenna, and the data is displayed in a line scan format with a linear gray scale color transform. The solid, vertical white lines at the top of the scans are tick marks produced by a survey wheel at 6-inch (150 millimeter) intervals. Dashed lines are reference marks created by the radar operator at head joint locations.

Figure 1 shows a typical scan performed on a section of wall constructed from nominal 8-inch (200 millimeter) concrete masonry units (CMU). This data shows that the CMU with grouted cells appear monolithic because the grout and masonry have similar dielectric properties. The characteristic black-white- black banding can identify the back edge of the wall associated with the transition from a higher dielectric to a lower one (in this case, masonry to air). The thickness of the wall is measured from the center of the first gray band of the input pulse (surface) to the center of the first black band of the reflected wave (back of wall). A similar pattern is generated by the reflection from a hollow cell. Unlike the planar geometry of the wall surface, how ever, the defined edges of the hollow cell produce a hyperbolic reflection pattern. The reflection at the steel rebar consists of a white black-white hyperbolic pattern because the phase is inverted upon contact with the metal target. The depth to the surface of the bar is 3.5 inches (90 millimeters), which places it at the center of the cell.

Figure 2 shows a series of scans performed on successive courses of a CMU wall. The grouted and reinforced cell c ontains on ly one bar on the lower and upper courses, but two separate bars can be identified in the middle course. The length of the lap can be ve rified by determining the number of courses that contain both bars.

Splices can be identified successfully when the lapped bars are located side by side, as shown here. However, in cases where the lapped bars are aligned perpendicular to the face of the wall, radar may not be able to detect the rearward bar because the shallower bar reflects the electromagnetic energy, effectively shielding the deeper bar.

Abilities and limitations SPR, like all available NDT methods, has various abilities and limitations. It is important for the engineer to be aware of these in order to be sure a particular method can provide the desired results. Based on the theoretical information presented above and previous experience with SPR systems used in field investigations, approximate "rules of thumb" have been developed for use in surveys of concrete masonry construction.

(See "Abilities and limitations of SPR" on page 24.) Be aware of the following concerns relating to penetration depth:

Minimum depth of objects resolvable by SPR is approximately one-half the wavelength from the antenna surface. For typ i cal concrete and masonry materials, this tra n slates to a minimum depth of 1 to 1 1/2 i n ches (25 to 38 millimeters) when using a 1,500-MHz antenna.

Maximum depth varies depending on material and antenna frequency. Maximum penetration is approximately 16 inches (405 millimeters) for plain concrete or solid masonry with a 1,500-MHz antenna. However, this value can be reduced significantly by the presence of closely spaced rebar or large air voids, especially when they are located near the surface.

Depth of penetration below the level of the rebar is approximately one-half the spacing between parallel bars. Congested rebar placed close to the surface can seve re ly limit the ability to resolve deeper targets.

Radar cannot penetrate electrically conductive materials such as metals or salt water. These materials act as reflectors for incident radar waves. Materials that are cove red by metal or saturated with salt water will not be able to be surveyed with SPR.

In addition, consider the following when evaluating SPR’s ability to detect targets:

Minimum layer thickness or minimu m distance between interfaces is approximately one-half the wavelength. For an air void in concrete or masonry, this corresponds to a minimum depth of approximately 1 inch (25 millimeters) for a 1,500-MHz antenna. This means that most cracks and delaminations are not easily detected by radar.

Embedded steel reinforcement as small as 1/4 inch (6 millimeters) diameter can be detected due to the reflective property of metals and other conductive materials.

The nominal size of a rebar cannot be determined from visual interp retation of the SPR output. Once the rebar are located, physical openings must be made to verify the actual bar diameter.

To resolve discrete targets such as individual steel rebar, the minimum spacing between bars must be at least four times the bar diameter.

Radar waves attenuate in air,making it difficult to detect targets within large (deep) air voids. For example, rebar or conduit located in the hollow cell of a concrete masonry unit usually will not be detected.

Because radar cannot penetrate metal, hollow steel objects, such as conduit, generate the same signal as a solid steel rebar (provided the diameters are similar). Plastic conduit can be detected in a grouted cell because of the annular shaped void that it creates,not by the thin plastic walls of the conduit.

Minimum length of survey area is approximately 24 inches (610 millimeters). The length of the antenna is approximately 6 inches (150 millimeters), which leaves a minimum length of 18 inches (460 millimeters) for the data record. Scans which are shorter than 18 inches (460 millimeters) should not be used because the characteristic patterns necessary for data interpretation will not be developed fully.

The dielectric constant of the material must be known in order to accurately calculate its depth or the thickness of an object using S P R. In radar investigations, a unit-less expression known as the relative dielectric constant (i.e. relative to air) is used to simplify the calculations. The relative dielectric constant varies from 1 for air to 81 for water. For concrete masonry, a value of 6.0 typically will provide reasonably accurate results. Accurate measurements can be obtained by calibrating the relative dielectric constant based on a known material thickn ess, such as at a grouted partition wall or bond beam. However, if access to a suitable location is not available, then an approximate value of the relative dielectric constant can be used. Small changes in this value do not significantly affect the thickness measurements because depth is proportional to the square root of the relative dielectric constant. Conclusion SPR is a non-destructive test method that is suitable for assessing the condition of concrete masonry structures. SPR has advantages over other methods when testing within existing structures because it has the ability to test surfaces covered by materials such as carpet, tile, and plaster or where access is limited to only one side of the object. The output is visual and is available in real-time to all ow for rapid testing. Results from field investigations have shown that SPR can provide reliable information of use to engineers, such as the placement of grout and reinforcing steel in concrete masonry walls. Depending on the orientation of the rebar, lap splice lengths also can be determ ined.

However, engineers should also be aware of the limitations of the equipment. For example, congested rebar placed close to the surface can seve rely limit the ability to resolve deeper targets; and, steel conduit and pipes may be mistaken for rebar since they produce similar radar signatures. Because SPR output requires visual interpretation, SPR testing should be performed only by an experienced operator who is knowledgeable about the equipment and about typical concrete masonry construction details. 

This article was developed from a paper that was published in the proceedings of the 9th North American Masonry Conference (NAMC) held in June 2003, sponsored by The Masonry Society. All SPR data and photos used for this article appear with permission from Whitlock Dalrymple Post on & Associates, Inc. (WDP) of Manassas, Va. The author would like to thank Eric Peterson of WDP for his assistance with this article.

Blaise A.Blabac, P.E., is a structural engineer at Modjeski & Masters in Poughkeepsie, N.Y., a leading consulting firm in major bridge design and rehabilitation. He is involved in the evaluation and repair of existing bridges, including seismic rehabilitation projects in the New York metropolitan area and historic bridge repair in the upstate area. He can be reached via e-mail at BABlabac@Modjeski.com.