Ground Penetrating Radar: Dielectrics
GPRS is the nation’s largest private sub surface locating company. GPRS is one of the company that provides concrete scanning, utility locating, leak detection, and video pipe inspection services. Our dedication to safety has helped us achieve an over 99.8% subsurface damage prevention scanning rate on hundreds of thousands of scanning and location projects. GPRS has unparalleled accuracy, and as a company, we want to raise the industry standard. To help push the industry forward, GPRS is releasing articles about our training methodology. This article will review our training documentation and methods concerning dielectrics. Before reading this, we recommend reading our article Basic GPR Theory.
Depth Accuracy – Time Zero
Time zero refers to the point in the data from which the depth measurements will begin. It is impossible to measure the exact moment a pulse leaves the antenna. The GPR signal must travel slightly before reaching the surface to have a surface reflection and a direct wave. The direct wave is the portion of the signal directly from the transmitter to the receiver without ever leaving the antenna.The reflection off the ground or concrete surface happens at nearly the same time as the direct wave, so the two will always be intermingled. Therefore, the direct wave is used as the surface (time zero) to ensure that all the data is being shown and that depth measurements are as accurate as possible.
All of our GPR systems automatically attempt to set the surface (time zero). They typically will set the surface correctly, so the important thing is to be in the habit of looking for times when it is not set correctly. The surface is most often set incorrectly when the antenna is initialized directly over a significant amount of shallow metal (rebar, decking, etc.)
Time Zero Errors
All our GPR systems attempt to automatically set the surface(time zero). They typically will set the surface correctly, so the important thing is to be in the habit of looking for times when it is not set correctly.The surface is most often set incorrectly when the antenna is initialized directly over a significant amount of shallow metal (rebar, decking, etc.). A reflection overpowers the surface reflection and direct wave from a metal object which causes the system to choose the object as the surface.
How to Change Time Zero manually
The example on the left shows what the O-Scope looks like when the surface is set far too high in the data. This view clearly shows the first positive peak. The example on the right shows what the O-Scope should look like after positioning the surface correctly at the center of the first positive peak.
How to Correct Time Zero
This example shows how the top of the data should look for the SIR 3000 and SIR 4000. Since the O-Scope should start at the center of the first positive (white) peak, there should be a thin white band at the top of the screen and a full black band directly below it.
Incorrect Time Zero
In this example, the AutoPosition feature set the surface too low within the data, most likely due to initializing directly on top of the shallow rebar. There is no visible black band at the top of the screen, and some of the hyperbolas are cut off at the top. All of the depths would be too shallow if this data was used, and some items could be missed due to being off the top of the screen.
In these examples below, the Position is too high within the data, most likely due to initializing in the air, not reinitializing after booting up the SIR 3000, or leaving the Position set to “Manual” from a previous job. The full white band is visible here at the top of the screen. All of the depths would be too deep if this data was used.
Depth Accuracy – Where to Mark
Marking Peak Amplitude
The second thing that affects depth is the point at which the target is marked. Of course, it’s evident that the spot chosen in the data as the object’s depth will affect depth accuracy. The technically correct location of an object will be covered here, along with how this could be done in correctly. To be technically accurate, mark the depth at the center of the first (real) reflection. Yes, marking the top of the peak is generally safer by making the object shallower, but it is not technically correct. Depending on the situation, safety can be more important than accuracy and vice versa.
The O-Scope can be used to determine the exact depth, as shown here. The most accurate depth is at the peak of the reflection (the center of the color), not the top of the peak. The O-Scope can also help determine which is the correct reflection. For example, the white halo above the negative (black) reflection is more prominent on the right side of the data. However, the O-Scope helps determine that the negative is still the higher amplitude reflection. Note: The O-Scope cannot be used to determine the accurate reflection to mark in every scenario. Too many factors can affect these amplitudes, but it is nevertheless a helpful clue.
Choosing the Correct Color
In this example, the difference between marking the center of the black vs. the center of the white is exactly 1”. The black should have been marked in this case because it is a plastic conduit with a negative reflection first. Note: a plastic conduit could be metal-filled, causing the positive reflection to be stronger than the negative, but if the negative is the first real reflection, it would still need to be marked as the object’s depth.
The difference between the negative and positive reflections is reasonably evident in this example. For the plastic conduit on the left, the difference between marking the black (negative) reflection and the white(positive) reflection is about ¾”.
Identify the Correction Reaction
In this example, choosing the right color is only a part of the problem. Finding the top of the object itself is difficult, so the correct color must be selected. In this example, the difference between marking the false hyperbola and the true hyperbola (outlined in red) is 3”. Note: cross polarizing would have helped here to have a better view of the conduit(s).
Basic Introduction to Dielectrics
GPR Locating – Determining Max Depths: Attenuation
GPR only measures the time and amplitude of reflection. GPR cannot determine how far the signal traveled, only the amount of time from when it leaves the antenna to when it returns. Knowing the time is not enough to calculate the distance (depth).
In the example below, each car drives for the same amount of time, but they travel very different distances. Distance = Time X Speed. Time and speed are both needed to calculate the distance. GPR travels at very different speeds in different materials. GPR depths are only as accurate as of the dielectric.
Dielectric is not used for improving the data. Dielectric represents the speed of the waves, which has everything to do with depth accuracy, not data quality! As shown in the following examples, the dielectric was changed, and the data shown in the O-Scope stays precisely the same, but the depth is dramatically different. Changing the data while viewing the O-Scope will make the signal appear better or worse but is only an illusion. It’s just the same as zooming in and out. Note: initializing the antenna does not calculate the dielectric. It only sets the surface and the gain.
Radar waves travel very fast through the air, the speed of light. Radar waves travel much slower through water, 9 times slower or 1/9the speed of light. All other naturally occurring materials are some where between air and water.
Dielectric values are ratio numbers without units representing a velocity in relation to the speed of light to make it easier. For example, you could either say that the velocity in air is 299,792,458 meters per second, or we can use the dielectric of 1 to represent the speed of light. We could say that the speed is 33,310,273 meters per second in water, or we could call it 81. The square root of 81=9, and 81 represents 1/9th the speed of light.
All-dielectric numbers work the same way (speed of light/square root of dielectric). Or speed=c/√k if c=speed of light and k=dielectric constant. More examples: Dielectric of 4=1/2 speed of light, the dielectric of 9=1/3 speed of light, the dielectric of 16=1/4 speed of light, etc. These examples should show that the GPR signal can travel at very different speeds regardless of understanding the math. Since GPR does not know the type of material you are scanning, it can’t know the speed unless the user inputs the dielectric.
Below is an example dielectric chart to get familiar with some common materials. All these values can vary; these are only examples. The main factor for variance is moisture. Most materials on their own will be between 3and 12. Moisture (water) is 81, so as more moisture is added, the dielectric value will continue to increase. Note: notice that metals are not listed on this chart. There is no speed listed for GPR waves to travel through metal because metal is a complete reflector of GPR. Practically speaking, the dielectric value of metals is infinite(∞).
Dielectric Values – Underground
This table is meant to be a simplified reference for some typical values worth remembering. As with any dielectric chart, the values can change depending on the moisture content and other factors. These are just approximate examples.
Dielectric Values – Concrete
The following table is a simplified reference for some typical values worth remembering. As with any dielectric chart, the values can change depending on the moisture content and other factors. These are just approximate examples.
• Examples of Very Dry (4-6): fully cured concrete for an elevated slab, interior.
• ModDry (6-8): Slab on grade, elevated slabs after 6 months-1 year
• Moist(8-12): Any slab between 1 and 6 months old. After complete cure, slab-on-grade tends to stay higher than an elevated slab, often 7-10, regardless of age. Any slab that has exposure to moisture can remain in this range regardless of age, exterior slabs, for example.
• Wet(12-15): Recently poured slabs <1 month or slabs with consistent moisture exposure such as a water tank/reservoir.
All of these numbers can vary and can be even higher than15. Many factors affect this, so the age of concrete alone is not enough to determine the dielectric. Potential rule of thumb: Starts at 14 and loses one point each month for 6 months.
The following animation shows how the dielectric of different materials affects the speed of the GPR signal through the material.The drastic change in speed demonstrates why the dielectric can have such a significant impact on depth accuracy. An animation will be shown when the slideshow is viewed or in the webinar version. Otherwise, the travel time can be viewed in ns (nanoseconds). The times have shown, and the time-lapse in the animation are all proportionately accurate.
Depth Variance – Underground
The following chart illustrates how changing the dielectric will affect the depth. A default of 8.0 is typically used as the dielectric for underground scanning. 9is used here to represent the default setting since 9 is a square number, and the math is more straightforward. If the default setting was used initially and the dielectric is changed, the new depths can be seen in this chart.
Depth Variance – Concrete
The following chart illustrates how changing the dielectric will affect the depth. A default of 6.25 is typically used as the dielectric for underground scanning. If the default setting was initially used and the dielectric is changed, the new depths can be seen in this chart.
Dielectric vs. Conductivity
Dielectric should not be confused with signal attenuation. Conductivity is the primary cause of signal attenuation (the signal dying out or being absorbed). This is where it gets a little more confusing. Higher dielectrics mean that the signal travels slower but does not necessarily mean that the signal cannot penetrate as deep. Electrical conductivity is what primarily limits depth penetration. More water generally raises dielectric and usually increases conductivity, so they tend to go hand-in-hand, but not always. The most common offender is clay. Clay tends to be conductive even when dry due to its molecular structure. Even water itself is somewhat confusing. Pure fresh water is not very conductive compared to water containing minerals. The conductivity from water in soil results from minerals being put into solution instead of just the fact that water was added. Ice and snow have a dielectric of3 or 4, and GPR can penetrate them very well. But in general, water being added to concrete and soil is the enemy ofGPR.
In the following example, the materials all have the same dielectric, and therefore, the signal travels at the same speed. However, the conductivity of the materials is different, so different depths are achieved.
In the following example, the materials have different dielectrics, and therefore, the signal travels at different speeds. The conductivity of the materials is still different, so different depths are achieved.
Wave Speed and Reflections
The following illustration shows how reflections happen when the speed of the signal changes. Objects are only visible when a reflection of the signal is received. Reflections only occur when there is a change in speed(dielectric). Change from higher to lower dielectric is a negative reflection(black). Change from lower to higher dielectric is a positive reflection(white). The speeds are actually the opposite since low dielectric equals high speed. The dielectric value will be used to determine positive or negative reflection. The contrast in speed determines the strength (amplitude) of the reflection.
GPR reflections occur due to changes in wave speed (changes in dielectric). Reflections do not happen because of a material change. Two very different materials with the same dielectric will not cause any reflection and will not be visible to GPR. Reflections do not occur because of a change in density. Two materials with significantly different densities having the same dielectric will not cause any reflection and will not be visible to GPR.Example: sand with a small amount of moisture (dielectric of 8) vs. granite(dielectric of 8) = no speed change and no reflection. Reflections do not occur because of a change in conductivity. Again, conductivity and dielectric often correlate but not necessarily. Example: Sand and clay both have enough moisture content for a dielectric of 10. There would be no reflection from a change between the two, but the clay would be more electrically conductive and allow less depth penetration than the sand.
The following image visualizes two different materials with the same dielectric causing no reflection.
Methods for Changing Dielectric
There are three methods we use for adjusting the dielectric.
1. Educated Guess
The first method for adjusting the dielectric to achieve accurate depths is to use a published reference, a.k.a.taking an educated guess. Be familiar with dielectric values for standard materials and choose an appropriate dielectric. This course contains instructional videos for entering dielectrics on different GSSI GPR systems. Depth accuracy estimated to be between 10%-20%.
2. Hyperbola Match
The second method for adjusting the dielectric to achieve accurate depths is to perform a hyperbola match. This can be done on all GSSI systems except for the SIR 3000. If you use a SIR 3000 and this is the only available method, raw data files can be sent to someone who has Radan, and a hyperbola match can be performed within the software. This course contains instructional videos for entering dielectrics on different GSSI GPR systems. Depth accuracy estimated between 5%-15%
3. Ground Truth
The third and most accurate method for adjusting the dielectric to achieve precise depths is to perform a ground truth. Ground truth can be done when the depth of an object can be confirmed. This can be done on all of the GSSI systems that we use. This course contains instructional videos for entering dielectrics on different GSSI GPR systems.Depth accuracy estimated between 5%-10%
Potential Problems and Things to Keep in Mind When Adjusting Dielectric
Only one dielectric value can be entered into the system. This means that changes in dielectric cannot be accounted for, only the average dielectric for all the data. Performing a hyperbola match or ground truth should be done on the deepest item possible if it’s within the desired medium. Soil can always change vertically and horizontally. We can assume that concrete is consistent.
GPRS specializes in ground penetrating radar, video pipe inspection, and mapping and modeling services. Our Project Manager shave the equipment and expertise to handle all subsurface challenges presented.GPRS does this by utilizing various equipment paired with their industry-leading SIM process.
Visit our website to learn more, or click here to schedule a project with GPRS and allow our highly trained Project Managers to keep your projects on time, on budget, and safe.