GPR: To Calibrate, or Not to Calibrate?
By Jami Harmon
When the need for concrete inspection arises, ground penetrating radar (GPR) is an ideal nondestructive technology for the job. GPR systems have continued to gain popularity in concrete imaging since the 1980s. Non-destructive testing methods are greatly preferred for locating buried metal targets or electrical conduits in existing structures. For many decades, X-ray technology was the predominant non-destructive technique used to locate and identify embedded targets. While concrete X-ray technology is still useful in some rare cases, scanning concrete using GPR has somewhat overtaken X-ray technology. It is now seen as the most efficient process to employ when locating rebar, post-tension cables, conduits or other targets in concrete prior to concrete cutting, coring or any other destructive procedures.
GPR technology allows personnel to work quickly, efficiently and accurately in real-time. Unlike X-ray technology, concrete GPR imaging equipment can locate targets and their depths safely, without the need to vacate the jobsite or handle hazardous materials. Yet, the technology itself still requires a significant amount of education—both with operators and contractor clients. A primary question that continues to come up in industry presentations, association meetings and on the jobsite is, “do I need to calibrate my RADAR system?”
The term ‘calibrate’ is defined as the correlation of readings of an instrument with those of a standard in order to check the instrument’s accuracy, or to adjust readings to take external factors into account or to allow comparison with other data. Relative to GPR, we calibrate systems to ensure accuracy and precision. For this type of equipment, calibration comes in two forms; distance and depth.
Distance & Position: The Importance of the Distance Measuring Instrument
A GPR system is comprised of three major components; the control unit, the antenna and the distance measuring instrument (DMI). The DMI, also commonly known as the survey wheel or encoder, is typically some type of optical encoder mechanically coupled to a wheel which rotates as the GPR antenna is moved. The encoder generates pulses which, when calibrated, has two important functions. The first is that it measures the distance traveled, and therefore allows the user to determine the position of targets. The second is that it sends a fixed number of pulses per revolution to the control unit used to trigger the antenna at equal distance intervals (scan spacing). As such, when one is speaking of doing a ‘calibration’ on the jobsite, they are typically referring to calibrating the DMI to ensure distance and positon accuracy. According to Dan Welch, training manager at CSDA member GSSI, “Under optimal conditions, which include a smooth surface, proper calibration and no slippage—a survey wheel distance error will not exceed more than 2 percent. That is +2 feet over a 100 foot distance.”
Depending on the manufacturer, survey wheel calibration may be configurable within the system settings. Although, some manufacturers require the GPR unit to be calibrated at the factory. Check with your equipment manufacturer to determine the appropriate calibration requirements.
Depth: Drilling Down on the Depth Scale
Equally as important as the accuracy of distance and position, is the precision of the depth. There are several ways in which one can determine depth. Each method includes setting up the GPR system with accurate depth measurements by identifying the correct dielectric value of the concrete. The term dielectric is also directly related to as ‘velocity’ or ‘concrete type’ depending on manufacturer terminology. The user must set the appropriate parameters for each job location and objective. Doing so may require operational training on the equipment prior to performing the survey. The setting of location-specific parameters is critical to the success of the job.
Accurate surface identification is the first step to a correct depth scale. Most GPR systems do not automatically locate the surface. The time (and depth) scale starts at an arbitrary time zero, usually above the true surface to ensure the top of the surface is captured in the data. This has to be taken into consideration and corrected. In many systems, this process can be done with a quick adjustment. This is the simplest, but least accurate method of depth calibration. By selecting a predetermined value (concrete type), a depth scale is established that is accurate within 20 percent. Sometimes, if the objective is to locate an area in the concrete to safely core drill and opening, a very accurate depth value is not required. However; if the 20 percent error percentage is unacceptable, there are more accurate methods available, such as Ground Truth and Migration.
A known target depth, known as Ground Truth, is the most accurate method to calibrate the depth scale. Any feature identified in the data can be used if its depth is known from an independent source, such as exposed rebar. If the concrete bottom (bottom of slab) is visible in the data, a known concrete thickness can be used if measured at the edge, in a core or known from good documentation. Core drilling is, in most cases, the only way to measure the exact depth to a particular rebar or other structural element.
A spot calibration can be applied to the entire structure or a section built with the same concrete. It is up to the analyst to decide how representative the spot value is.
If it is not possible to obtain a known target depth, migrating the data by means of hyperbola matching is an alternative for depth measurement calibration.
Migration calculates the signal velocity in the medium from the shape of hyperbolic reflections. By matching an animated hyperbolic shape to the GPR data, the system is able to measure the velocity of the signal and recalculate the depth scale. Migration is only accurate if you scan perpendicularly across the target, as the shape of the hyperbola will change if the target is not scanned perpendicularly.
Migration, commonly referred to as Focus Mode, is becoming a standard in depth calibration. An added benefit is that this process eliminates hyperbolas by collapsing them into dots representing the actual targets. This can be helpful to make target identification more intuitive. Migration can reduce clutter in the image and make it easier to interpret.
To Calibrate, or Not to Calibrate?
The accuracy of all measuring devices, including GPR, change over time. This is typically caused by normal wear and tear, damage, etc. However, changes in accuracy can also be caused by factors related to the survey environment, such as concrete surface or age. Depending on these factors, there is no specific schedule related to calibration needs, as frequency may change.
Understanding your GPR system will help identify when and how often you should consider calibrating it. Each equipment manufacturer will outline calibration techniques in their user manuals.