3 PRINCIPLES AND THEORY OF EDDY CURRENT INSPECTION.
3.1 Induction of Eddy Currents. As the electromagnetic field from a coil penetrates a conductor, it generates eddy currents parallel to the surface of the part and at right angles to the direction of the applied field (Figure 1). The frequency of eddy current flow is the same as the electromagnetic field.
Figure 1. Primary and Secondary Magnetic Fields in ET
3.2 Primary Electromagnetic Field. The primary electromagnetic field is the coil’s magnetic field (Figure 1). This field is called electromagnetic because the magnetic field is produced from electricity rather than from a permanent magnet. The rate at which the electromagnetic field varies is called the frequency. The strength of the electromagnetic field at the surface of the conductor depends on the coil size and configuration, the amount of current through the coil, and the distance from the coil to the surface. The amount of eddy currents the primary field is able to generate is dependent upon the properties of the part under test and the strength of the secondary electromagnetic field that opposes the primary field.
3.3 Secondary Electromagnetic Field. Eddy currents also generate an electromagnetic field in the part. This field, called the secondary electromagnetic field, opposes the primary electromagnetic field (Figure 1) and is a consequence of Lenz’s Law. Lenz’s Law, as applied to this case, states induced currents (eddy currents) act to reduce the magnitude of the inducing current. The opposition of the secondary field to the primary field decreases the overall electromagnetic field strength and reduces both the current flowing through the coil and the resultant eddy currents. Changes to the properties of the inspection article produce changes to the eddy currents and thus their secondary magnetic fields. In this manner, changes in the inspection article produce effects that can be detected by monitoring either the source of the primary electromagnetic field or the overall electromagnetic field.
3.4 Depth of Penetration. The intensity of eddy currents decreases exponentially with depth in a material. The intensity at any given depth is affected by the same variables that influence the surface intensity of eddy currents, although not always in the same manner or by the same amount. To put it another way, the depth of penetration of a specific intensity of eddy currents is affected by the variables, as indicated in Table 4-3 in Paragraph 4.8. Generally, any parameter that increases the depth of penetration would increase the detectability of discontinuities deeper in the part.
3.4.1 Standard Depth of Penetration. Three of these variables (conductivity, relative magnetic permeability, and frequency) are used to define the standard depth of penetration. Standard depth of penetration is the depth below the surface of the inspection article at which the magnetic field strength, or the intensity of the induced eddy currents, is reduced to 36.8 percent of the value at the surface. The standard depth of penetration is expressed by the following formula in Paragraph 4.8.7. Since the depth of penetration is related only to a percentage of surface field strength (eddy current intensity) some test variables are not included in the formula. Coil configuration, size, current, and magnetic coupling are not considered in this formula. These variables affect the absolute magnitude of the eddy currents at a specified depth but not the standard depth of penetration. The standard depth of penetration values for select frequencies for various alloys, bare aluminum alloys, and clad aluminum alloys are shown in Table 4-5 and Table 4-6 in Paragraph 4.8.
3.4.2 Effective Depth of Penetration. Effective depth of penetration is the depth in the inspection article at which the magnetic field strength or the intensity of the induced eddy currents is reduced to 5-percent of the value at the surface. This depth is approximately 3 times the standard depth of penetration (According to ASTM E1004, the effective depth of penetration used for the purposes of conductivity testing is 2.6). The effective depth of penetration is used to determine test frequency when working with thin materials, so the overall electromagnetic field does not extend beyond the back surface of the test part so thickness variation effects can be suppressed. The minimum material thickness required for conductivity testing various alloys at 60 kHz and 480 kHz using the ASTM values of 2.6 is shown in Section IV, Table 3, Paragraph 8.
3.4.3 Temperature and Depth of Penetration. For most applications, temperature is not a major factor in determining depth of penetration. However, if necessary the effects of temperature would be included as adjustments to the values for conductivity and relative magnetic permeability used in the formula to calculate the standard depth of penetration.
3.5 Impedance. Impedance is the total opposition to the flow of current represented by the combined effect of resistance, inductance and capacitance of a circuit.
3.6 Sensitivity. The ability of an eddy current instrument to detect small variations in test coil impedance is a measure of its sensitivity. This quality is interrelated with the properties of the test coil and the operating frequency. Therefore, instrument sensitivity to a particular flaw condition or material property SHALL be established from reference standards representing this condition.
3.7 Resolution. The ability of a test system to separate the signals from two indications that are close together is defined as resolution. This property plus sensitivity must be considered in every flaw evaluation situation. Probe design, test frequency, and instrumentation design are all factors in determining the resolution of an eddy current system.
3.8 Measurement of Resistivity. Electrical resistance is a measure of the resistance to the flow of electric current in a conductor. Resistance depends on the length and area of the current path, and the conductivity of the conductor. Resistance is commonly measured in ohms. If a material allows one volt (electric potential) of driving force to push one ampere of current through a conductor, the electrical resistance of the conductor is defined as one ohm of resistance. Resistivity is a mate rial parameter independent of the size of a material sample and is related to resistance. Resistivity is defined as ohms times cross- sectional area divided by unit of length (Paragraph 8.1.3).
3.9 Measurement of Conductivity. Electrical conductivity is the reciprocal of electrical resistivity. The reciprocal of the ohm is commonly called the mho. Conductivity is commonly expressed in units of mho’s per unit length; such as mho/inch or mho/meter. The relationships between conductivity, resistivity, and resistance are expressed by the equations in Paragraph 8.12.
3.9.1 Conductivity Based on the Percentage of International Annealed Copper Standard. (%IACS). An alternative way of expressing conductivity is a percent of the conductivity of a known material. The International Electrotechnical Commission has designated the conductivity of a specific grade of high purity copper to be the standard for this alternative method with a conductivity of 100-percent. It is called the International Annealed Copper Standard (IACS). The conductivity of all other metals is then expressed as a percentage of this standard.
NOTE
Values of conductivity of some commonly used engineering materials are listed in Table 2 and Table 7 in Paragraph 8. Percent IACS is the usual way of expressing conductivity in aerospace NDI.
3.10 Overview of Signal Detection, Processing, and Display.
3.10.1 Signal Sources. When performing an eddy current technique, material changes can be detected by monitoring the alternating current in the coil (single coil arrangement) or using a separate sensing coil to monitor the resultant electromagnetic field. These signals can be analyzed for information relevant to the inspection being conducted. The important thing to note is the coil that is acting as the receiver is producing an electrical current that either leads or lags the instruments oscillator current. The difference in this leading or lagging is the phase angle.
3.10.2 Signal Detection. A simple but effective signal detection technique is to use a bridge circuit as illustrated in Figure 2. With current flowing through the test coil and the coil positioned on a flaw-free or reference area, the variable impedance Z 1 can be adjusted so zero current flows through the amplifier. This adjustment is termed either balancing or nulling the bridge. When the coil is placed on a flawed or damaged area, the resultant change in current through the coil “unbalances” the bridge and current flows through the amplifier. This current is the inspection signal. The signal has the same frequency as the current through the coil. The phase and amplitude of this signal contains information on the condition that caused the bridge unbalance.
Figure 2. Simplified Bridge Circuit
3.10.3 Signal Analysis. In the simplest type of instrumentation, analysis of the signal consists of measuring the change in magnitude of the current flowing through the bridge. Changes in the magnitude of the alternating current are amplified and converted to a direct current for display or readout. In more sophisticated instrumentation, both amplitude and phase are measured.
3.10.4 Displays. The method by which eddy current signals are presented is dictated by the type of information required and the complexity of the instrumentation. When only signal amplitude is measured, meters, alarm signals, or recorders are commonly used. When both amplitude and phase information are to be displayed, a two-dimensional display device is normally used.
3.10.4.1 Amplitude Display. Meters may be analog (needle moving over a fixed numerical scale) or digital. Audible or visual alarms may be set to trigger when the signal amplitude exceeds a predetermined threshold. A recorder presents a continuous record of the signal amplitude during an inspection for subsequent analysis.
3.10.4.2 Impedance Plane Display. Defects or other variations in material characteristics will alter the strength and distribution of an induced eddy current flow. Changes in the eddy current flow will result in changes in the inducing coil or sensor coil currents. These changes can be expressed as an apparent change in the coil’s electrical impedance. This makes it possible to associate changes in material properties with specific changes in the apparent impedance of either the excitation or sensor coils. The two-dimensional display that permits this is the most commonly used and is called an impedance plane display. The impedance plane is discussed further in Paragraph 4.3.10.8.2.
3.10.5 Impedance Changes. The impedance of a coil appears to change when it is placed adjacent to an electrically conductive or ferromagnetic part. The secondary electromagnetic field created by the induced eddy current in the part opposes the primary field. This opposing field also induces a current flow in the coil in opposition to the primary current. If the part is not ferromagnetic, the net magnetic field resulting from the combination of the primary and secondary fields is decreased in magnitude, as is the current flow in the coil. This is equivalent to decreasing the inductance and increasing the resistance of the coil. If the part is ferromagnetic, the net magnetic field is increased because of the magnifying effect of the relative magnetic permeability, but the current flow in the coil is decreased because of the opposing effect of the secondary magnetic f ield from the induced eddy currents. This is equivalent to increasing both the inductance and resistance of the coil. In this manner changes in a part that affect either the strength of the magnetic field at the surface of the part or the strength and distribution of the eddy currents in the part, change the apparent impedance of the test coil(s). These variations in current f low, both phase and amplitude, can be detected, amplified, displayed, and analyzed as eddy current test results. The amplitude and phase changes in the signals can be related to changes in the parts inspected.
3.10.6 Inductance of a coil. The inductance of a coil depends upon the number of turns in the coil, the size of the coil, the permeability of the material within the coil (e.g., the core of the coil), and total magnetic flux through the coil. An alternate method of expressing self-inductance (L) is:
L = n . deltaphi; / I
Where:
L = Inductance (henry)
n = Number of turns in coil
deltaphi; = Magnetic flux (weber)
I = Current through coil (ampere)
3.10.7 Inductive Reactance. The measure of the amount of opposition or resistance (ohm) to alternating current flow due to inductance in a coil is called inductive reactance. Inductive reactance is dependent upon the value of the inductance of the coil and the frequency of the alternating current. The inductive reactance increases as the inductance or frequency increases. This can be stated by the following equation:
XL = 2 . deltapi; . f . L
Where:
XL = Inductive reactance (ohm)
deltapi; = 3.141596
f = frequency (hertz)
L = Inductance (henry)
3.10.7.1 The inductive reactance results from the electromotive force generated across a coil by the alternating current. The instantaneous value of this induced voltage, increases and decreases as the rate of change of the applied alternating current increases and decreases as shown in Figure 3. The voltage is at its maximum value when the rate of current change is at its maximum; this occurs when the current value is at zero. Conversely, the voltage is zero when the rate of current change is zero; this occurs when the current is at its maximum value. Considering 360-degrees to be one complete cycle, the induced voltage leads the current (e.g., is out of phase with the current) by 90-degrees as illustrated in Figure 3. The induced voltage is in opposition to the electromotive force applied to the coil, reducing the amplitude of the resultant current.
Figure 3. Sinusoidal Variation of Alternating Current and Induced Voltage in a Coil
3.10.8 Combining Out of Phase Quantities. A real coil has a resistive component of the impedance in addition to the inductive reactance. They can be combined to describe the net impedance. A coil can be considered to be a resistor in series with an inductor. Applying an alternating current to this series circuit will result in two voltages, one across the resistor and another across the inductor. The net voltage across the combination of the resistor and inductor (e.g., across a real coil), will be the combination of the two voltages. The voltage across the resistor will be in phase with the current while the voltage across the inductor will lead the voltage across the resistor by 90-degrees. The combination of the two voltages, as illustrated in Figure 4, results in a voltage that will be out of phase with the current but not by a full 90-degrees.
Figure 4. Combining of Out-of-Phase Voltages
3.10.8.1 X-Y Plot Representation. Another way to illustrate the combination of out-of-phase quantities in a coil is illustrated in Figure 5. Here the two voltages drop; one across the resistor (VR) and the other across the inductor (V L) are plotted at right angles to each other. This represents the two quantities being 90-degrees out of phase. The combination of the two quantities is represented by the diagonal line OA that is at the angle with respect to the voltage drop across the resistor.
Figure 5. Vector Diagram Showing Relationship Between Resistance, Reactance, and Impedance
3.10.8.2 Impedance Plane Representation. Just as the two voltages can be combined to produce the net voltage across a coil (Figure 6); the resistive and inductive impedance components can be combined to produce the net impedance of a coil. In Figure 5, inductive reactance (X L) is plotted on the y-axis and resistance (R) is plotted along the x-axis. These two values define the impedance that is represented by the vector OA. The value of the angle is the same as the anglefor the net impedance illustrated in Figure 6 for the net voltage. This is important because it shows that the impedance of a coil can be displayed as the combination of two out-of-phase voltage drops. The amplitude of the impedance may be determined from the known values of resistance and inductive reactance according to the following formula:
Z = (XL2 + R2 + XC2)1/2
Where:
Z = Impedance magnitude (ohms)
XL = Inductive reactance (ohms)
R = Resistance (ohms)
XC = O Capacitive reactance is negligible.
Figure 6. Diagram Showing Relationship of Voltage Drops Across Coil Resistance and Coil Reactance
3.10.8.3 The phase angle ( ) of the impedance can be calculated from the values of resistance and inductive reactance as follows:
Tan = XL / R
Where:
deltateta; = Phase angle (degrees)
XL = Inductive reactance (ohms)
R = Resistance (ohms)
XC = Capacitive reactance (ohms)
3.11 Impedance Diagrams.
3.11.1 Purpose. The impedance diagram shows how changes in eddy current test variables change the apparent impedance of a coil. Typical variables displayed are electrical conductivity, relative magnetic permeability, fill-factor or lift- off, part thickness, and test frequency. Impedance diagrams are very useful in determining optimum inspection parameters and understanding eddy current results when more than one variable is changing. The vector representation of inductive reactance on the y-axis and resistance on the x-axis of Figure 7 is the basis of the impedance diagram. Let point A represent the impedance of a test coil while on a part. If the probe is moved to a place on the part with a flaw, the impedance will change. This new impedance can be represented by the point B, as shown in Figure 7. Each change in the impedance will create a new point on the diagram.
Figure 7. Vector Representation of Impedance
3.11.2 Development of an Impedance Diagram. To make the impedance diagram into a useful tool for understanding eddy current testing, it is necessary to systematically change a single test parameter such as conductivity, and observe the changes in the impedance. Using an eddy current instrument with a two-dimensional graphical display, a surface probe, a piece of ferrite (a nonconductive, ferromagnetic ceramic) and several nonmagnetic metal specimens representing a range of conductivity’s from low (titanium, Inconel) to high (copper, silver), approximate impedance diagrams can be developed and demonstrated. The specimens must have clean, flat, and bare surfaces. When the eddy current probe is held away from the part (in the air) and the instrument is nulled, an indication (dot) will appear on the display. The null point can be repositioned near the upper left hand corner of the display, as indicated by point A in Figure 8 and Figure 9. The null point in air will be used as a point of reference for the rest of the diagrams. Next, the ferrite specimen is used to establish the direction of in ductive change. Place the probe on the ferrite and adjust the phase control so that the change from air to ferrite is vertical (parallel to the y-axis). When the probe is placed on the copper specimen, the point will move to a new location on the screen, represented by point I in Figure 9. As the probe is lifted from the specimen, the point will move back to the air null point (A), as shown in Figure 8 and Figure 9. The path that the indication follows as the probe is moved onto and off the specimen is called the lift-off trace/line.
Figure 8. Vector Representation of an Impedance Change due to Lift-Off
Figure 9. Impedance Diagram Illustrating Effects of Variable Conductivity
3.11.3 Typical Uses of an Impedance Diagram. The impedance diagram (shown in Figure 9) illustrates the conductivity curve can be used to measure the relative conductivity of an unknown material by comparing the position of its indication on the conductivity curve to the positions of indications from known materials. Notice also the lift-off lines are in a different direction than the conductivity line. Changes in conductivity and lift-off are said to have different phase angles. This phase angle difference is further illustrated in Figure 10. The lift-off curve can also be used to measure the thickness of non-conductive coatings on a conductive surface. This is done by comparing the length of lift-off line for an unknown coating thickness to the lengths of lift-off lines for known thickness.
Figure 10. Phase Angle Difference
3.11.4 Conductivity Curve. The gain and phase controls can be adjusted to place point I anywhere on the display. Because copper has high conductivity, it will be convenient to adjust the gain to put point I in the lower part of the screen, (Figure 9). When the probe is placed on the other metal samples, the respective impedance points “B through H” (Figure 9) are recorded. Note for each of the different materials, the point will be located at a different location on the screen (e.g., each different specimen has a different impedance). Each line from the null point A, to the impedance point for a particular specimen represents a liftoff trace. If a smooth curve is drawn from the null point A through each of the impedance points B through I, a conductivity curve will be formed. The point on the curve closest to the air null point represents the material with the lowest conductivity (e.g., titanium). The point on the curve farthest from the air null point represents the material with the highest conductivity (high purity copper). This diagram also shows the relative conductivity of the other specimens.
3.11.5 Thickness Variations. When the part thickness is less than the effective depth of penetration of the test coil at the inspection frequency employed, the impedance curve departs from the conductivity curve as shown in Figure 11. Typically, there is an increase in the resistive component of the impedance with thinner parts, as compared to parts that have thickness equal to or greater than the effective depth of penetration. As the thickness of the parts increase and approach more closely the effective limit of penetration, the curve tends to spiral as it approaches the end point (T=1) on the conductivity curve, where T equals the ratio of the specimen thickness to the effective depth of penetration in that specimen.
Figure 11. Impedance Diagram Showing the Effect of Specimen Thickness
3.11.6 Conductive Layers. The impedance curve for thin conductive layers on a substrate of different conductivity is also represented as a change in the impedance curve for conductivity. The impedance for the layered material departs from the conductivity curve at the value corresponding to the substrate conductivity and forms a loop that rejoins the conductivity curve at the conductivity of the metal in the outer layer. Increasing thickness of the outer layer corresponds to a clockwise direction along the loop. The point at which the loop rejoins the curve represents the effective depth of penetration in the outer layer.
3.11.7 Normalization of Impedance. To illustrate general principles of eddy current inspection or to present data in a universal form independent of specific coil impedance values, impedance diagrams are usually normalized. In normalization, the inductive reactance and the resistance of the coil on the part are divided by the value of the inductive reactance of the coil in air. Therefore, the vertical axis of the impedance diagram equals the relative inductive reactance (XLN) of the test coil; and the horizontal axis of the impedance diagram equals the relative resistance (RN) of the test coil. Normalization is a convenient method to allow comparisons of eddy current data from a large number of tests using different probes and materials. The shapes of the impedance diagrams remain the same. However, the air null point A in Figure 12 becomes 1 on the y-axis of the impedance diagram after normalization. The impedance diagrams in this manual will all be represented by the normalized reactance (XLN), on the y-axis and normalized resistance (RN) on the x-axis.
Figure 12. Impedance Diagram Showing the Effect of Lift-Off
3.12 Impedance Plane Analysis. Most eddy current applications have two major problems to overcome. The first is to ignore changes in parameters not of interest during the test; an example is lift-off variation while inspecting for cracks. The second is to recognize valid indications while other changes are occurring. Another way of stating this is insignificant variations such as those associated with lift-off should not be mistaken for valid defect indications, and valid defect indications should not be hidden by nonrelevant changes such as lift-off. Impedance plane analysis, also called phase analysis, is a tool that is effective in solving these problems.
3.12.1 Phase Detection. Phase angle measurements are a good way to detect a variety of flaw conditions. The information in the vector diagram (Figure 5) illustrates this. Decreases of conductivity (e.g., cracks) and permeability could produce the same signal amplitude, and it would be difficult to differentiate between cracks and normal permeability changes in a part. However, the phase angle of a conductivity change is very different from a permeability change if the correct test frequency is chosen. Using phase detection techniques, it becomes a simple matter to detect the difference between permeability variations and cracks. This also applies to determining the depth of a flaw, which is phase sensitive, or separating lift-off effects from flaw conditions. Phase sensitive detectors use a variety of techniques such as phase splitters, phase shifters, averaging, half-wave and full-wave detection, sampling, and subtractive and additive techniques. The presentation of the impedance plane on waveform display eddy current instrument; uses two-phase sensitive detectors to provide horizontal and vertical phase detection. This information is combined to produce a dot or point on the screen which represents the relative phase and amplitude of an eddy current signal. Some types of meter instruments utilize an adjustable phase control or phase gate to allow only signal detection at a particular phase angle of interest.
3.12.2 Cracks, Lift-Off, and Conductivity. The impedance changes due to surface cracks of different depths. The change for cracks will lie between the lift-off and conductivity. As the crack depth increases, the response moves farther from lift-off and closer to decreasing conductivity.
3.12.3 Crack Detection in Non-Ferromagnetic Materials. The amplitude of the response from a surface crack increases as the crack gets deeper. When the crack reaches three standard depths (Paragraph 3.4.1) it is interrupting essentially all of the eddy current flow and no increase in amplitude is seen as it gets still deeper. Besides an amplitude increase for deeper cracks, the phase angle of the crack indication changes. A shallow crack interrupts little of the eddy current flow, so the amplitude of its signal is small. Also, it is essentially a surface condition, so the direction (phase) of the signal response is very close to that of lift-off (Figure 13). A deeper crack interrupts more of the eddy current flow, so its signal has greater amplitude. It extends well below the surface, the direction (phase) of its signal is farther away from lift-off (Figure 14). The three standard depths crack has the largest amplitude response. It interrupts the eddy currents as far down in the metal as the test can sense, it looks like a change in the bulk property of lower conductivity, and the crack signal direction (phase) is along the conductivity curve (Figure 15).
Figure 13. Shallow Surface Crack
Figure 14. Deeper Surface Crack
Figure 15. Three Standard Depths of Penetration
3.12.3.1 Making the three standard depths crack deeper will not change the signal response because there will be no eddy current flow for it to interrupt. However, there will be a change in the signal response for a subsurface crack. First, eddy currents will flow over the top of the crack (at the surface), the subsurface crack will not block as much of the eddy current
f
low and the amplitude of the signal must decrease. Second, the crack is now farther away from the surface so its phase angle must still be further away from lift-off (Figure 16).
Figure 16. Subsurface Crack
3.12.3.2 Signal response decreases as the depth of the crack below the surface increases. As the subsurface defect gets
further away from the surface, the signal amplitude gets smaller and the phase angle rotates clockwise, away from lift-off (Figure 17).
Figure 17. Deep Subsurface Crack
3.13 Phase Lag at Depth. A phase angle shift can occur and change the eddy current field time and travel distance. Changes at the surface of the part are seen immediately by the coil, while disturbances to the field at some depth in the part require some travel time to return to the surface where they are seen by the coil. Electrically, this is described as phase lag at depth, and the amount of phase lag is 1 radian (57°) per standard depth of penetration Figure 18). This phase lag from the lift-off (surface) signal may be used to measure the depth of defects. The phase angle of a defect signal correlates to defect depth.
Figure 18. Phase Lag and Depth in Part
3.14 Effects of Inspection Conditions on Eddy Currents.
3.14.1 Frequency. The magnitude of the induced eddy currents in the part increases as the frequency of the inducing current increases. In turn, the higher intensity eddy currents generate a stronger opposing magnetic field, reducing the penetration of the primary field. Therefore, all other factors remaining constant, higher frequencies result in shallower depths of penetration as shown in Figure 19.
Figure 19. Relative Effect of Frequency on Depth of Penetration
3.14.2 Conductivity and Frequency. There is a relationship between conductivity and optimal inspection frequency. As an example, an eddy current inspection for cracks in aluminum alloy 7075-T6, with a conductivity of about 30% IACS uses a frequency of 200 kHz. To perform an inspection with comparable depth of penetration on a titanium alloy, TI 6Al-4V with a conductivity of about 1% IACS, a frequency of about 6 MHz would be required.
3.14.3 Electromagnetic Coupling. The interaction between the primary electromagnetic field generated by the coil and the inspection article is referred to as electromagnetic coupling. Because the field decreases in strength with increasing distance from the coil, resultant eddy currents at the surface of the part will also decrease in intensity. An electrical engineering term that could also be used is inductive coupling.
3.14.4 Fill-Factor. When an encircling coil is used to inspect a cylindrically shaped part, the degree of magnetic coupling is dependent upon the difference between the internal diameter of the coil and the external diameter of the part. This effect is termed fill-factor. For internal coils, electromagnetic (inductive) coupling is determined by the air gap between the external diameter of the coil and the internal diameter being inspected. Fill-factor is calculated using the basic formula, but in this case d i is the inside diameter of the part and D 0 is the outside diameter of the coil placed in the part (Paragraph 8.3).
3.14.5 Coil Current. With all other factors constant, an increase in current flowing through the coil results in a higher magnetic field strength.
3.14.6 Temperature. The temperature at which an inspection is performed affects both the electrical conductivity and the ferromagnetic properties of the inspection article. Electrical conductivity generally decreases with increasing temperature, and conversely increases with decreasing temperatures. The reduction at higher temperatures occurs because of the scattering of conduction electrons by atoms moving with increased thermal oscillations. Temperature effects on the ferromagnetic properties of a material are generally negligible with one exception. Above a specific temperature called the Curie temperature (about 1400°F or 760°C), ferromagnetic properties disappear. Because of the thermal effects on conductivity, increasing temperature of the inspection article slightly decreases the intensity of eddy currents at the surface of a part and slightly increases the depth of penetration. Temperature variations also affect the inductance of the coil. Remember, changes in temperature affect ET results. Therefore, during inspections, time SHOULD be allowed for the test system and the test part to stabilize to the ambient temperature. An example test would be to see if part and standards feel the same to the bare hand.
3.14.7 Geometry. Geometric features such as edges, curved surfaces, changes in thickness, and non-conductive coatings (such as paint) on surfaces affect the distribution and strength of eddy currents. As a probe approaches an edge the eddy current response is known as edge effect and appears similar to a response from a crack. Similarly, curved surfaces and nonconductive coatings can vary the distance between the probe coil and the part. These changes are known as lift-off, and the consequent effects on the eddy current signal are called lift-off effects. Lift-off usually cannot be completely prevented; therefore compensating for some lift-off is part of the setup procedure. Part thickness variations can also produce an interfering response in some eddy current units when the thickness is in the range of the depth of penetration of the eddy current field.
3.14.8 Lift-Off. The effects of lift-off can be used to measure coating thickness. Changes in lift-off can be calibrated to allow measurements of nonconductive coating thickness. Fill-factor applies to parts passed through an encircling coil and, in a manner similar to lift-off, can be used to gauge some dimensions. As a test coil is moved away from a part (increasing liftoff) the coupling between test coil and inspection part is decreased. The magnitude of the impedance change for a specific change in an inspection variable is also decreased. For probe coils, the dotted lines connecting points representing the same material properties but with various amounts of lift-off have some curvature as shown in Figure 12. The line A-B-C represents the increase lift-off for material one. Line D-E-F represents the increased lift-off for material two. The line from point A to point D represents the increase in conductivity of material two compared to material one at one lift-off value. Lift- off lines B-E and C-F are increasingly shorter, indicating a smaller change in the conductivity.