Simultaneous Absolute and Differential Operation of Eddy Current Bobbin Probes for Heat Exchange Tube Inspection

In this month's "NDT Solution", the authors try to dissolve some commonly held misconceptions about the operation of eddy current probes used for inspecting heat exchanger tubes. With the help of computer model results, the authors demonstrate that, for optimized operation, absolute/differential probes should be operated with coil current flowing in phase with another. This month's feature should be of interest to researchers as well as for NDT field personnel. G.P. Singh Associate Technical Editor

 

Figures 1-3
Figures 4-6

Introduction
Modern instrumentation permits inspection of heat exchanger tubes with eddy current bobbin coil probes operating in absolute and differential modes simultaneously. Many believe that differential operation of these probes requires the current in one coil to be 180 degrees out of phase with the current in the other. This paper shows that differential operation of these probes does not depend on the relative sense of the drive current in the coils but that optimized absolute operation requires that the drive current in one coil flow parallel to the current in the other.

When eddy current probes were first used for inspecting heat exchanger tubes, they were generally composed of two identical bobbin coils mounted closely together, operating in the differential mode. A schematic of such a probe is shown in Figure 1. The recorded signals were the induced voltage in one coil subtracted from the voltage in the other coil. The advantage of this probe is that its signal is resistant to various anomalous effects, such as probe wobble, temperature variations, and gradual variations in the inspected tube's electrical conductivity, diameter, and ovality. This probe is very sensitive to abrupt anomalies, such as pitting corrosion and fretting wear.

---

Differential signals are not dependent on whether the current in the coils is flowing in phase.

---

 Some researchers had developed computer modeling tools to study the electromagnetic operation of differential probes and to optimize probe designs for future inspections. These researchers used finite element modeling (FEM) to generate magnetic field distribution displays (Palanisamy and Lord, 1980; Palanisamy and Lord, 1983) and impedance plane signals from anomalies in such items as support plates (Palanisamy and Lord, 1980; Palanisamy and Lord, 1983; Ida et al., 1983). They modeled probes with current in one coil flowing 180 degrees out of phase with the other coil.

As reported by Cecco, Van Drunen, and Sharp (1983), and by many inspectors in the field, the differential probe was found to be insensitive to gradually varying wall thinning resulting from corrosive wastage or tube to tube fretting. To detect such wall thinning, which was often quite severe, absolute eddy current probes were required. The first absolute probes were operated with a single bobbin coil, as shown in Figure 2. A second reference coil, which was used for electronic balancing, was electromagnetically isolated (shielded) from the inspected tubing. Although this probe was also sensitive to pitting and fretting wear, it was not as sensitive to these anomalies as a differential probe.

With modern instrumentation, bobbin probes are now generally operated simultaneously in absolute and differential modes. Such probes have optimized sensitivity to abrupt anomalies, while retaining sensitivity to gradual wall loss. For the absolute signals, these probes are electronically balanced with identical probes, which are in isolated reference tubes. It is now common practice to operate these probes with the current in both coils flowing in phase. This is contrary to the intuition of most people, including many certified inspectors, who believe that differential operation of these probes requires the current in one coil to be 180 degrees out of phase with the other.

This paper shows that differential signals are not dependent on whether the current in the coils is flowing in phase or if the current in one coil is 180 degrees out of phase with the current in the other. In addition, computer modeling results are presented, showing that such probes should be operated with coil current flowing in phase in both coils for optimized absolute mode operation.

 

Theory
Eddy current signals are displays of recorded voltage induced in the probe coils during inspections. In the case of bobbin probes, such as the one in Figure 3, the voltage across the leads of each of the probe coils is given by Equations 1 and 2:

(1)	V1 = (I  *  Z11) ± (I *  Z12)
(2)	V2 = (I  *  Z22) ± (I *  Z21)

where I is the current flowing in the coils, Z₁₁ and Z₂₂ represent each coil's self impedance in a tube at a given operating frequency, and Z₁₂ and Z₂₁ are the mutual impedances between the coils. The ± sign in each of the equations is dependent on whether the probe's coil current flows in phase or 180 degrees out of phase in the two coils.

The differential signal generated by the probe is equal to the difference in voltages across the coils. The law of reciprocity states that Z₁₂ = Z₂₁. Equation 3 expresses the voltage used for the differential signal:

(3)	Vdifferential =  V  -  V2  =  (I *  Z11) - (I * Z22)

Because of the law of reciprocity, mutual impedance terms and ± signs are eliminated when subtracting Equation 2 from Equation 1 to obtain the differential voltage in Equation 3. The differential signal is, therefore, independent of whether the current in the coils flows in phase or if the current in one coil is 180 degrees out of phase with the current in the other.

 

Computer Modeling Results
We used the MagNet FEM code (Infolytica, 1995) to generate magnetic field patterns, as shown in Figure 3. The pattern in Figure 3a is from a probe with the current in one coil 180 degrees out of phase with the current in the other and is similar to those produced by Palanisamy and Lord (1980 and 1983). The pattern in Figure 3b is generated when the current in both coils is in phase. Although the patterns are significantly different, the differential probe signal is independent of the relative sense of the currents.

Figure 4 shows absolute signals from a true absolute bobbin probe and from a probe operating simultaneously in absolute and differential modes. Figure 4a shows the signal from the true absolute probe. Figure 4b shows a calculation of the absolute signal from an absolute/differential probe with the current in one coil 180 degrees out of phase with the other coil. This signal resembles a distorted differential signal. Because of such distortion in the signal, practical usage of such a probe would be extremely difficult even for trained analysts of eddy current signals. The absolute signal calculated with in phase coil current is shown in Figure 4c. Although this signal is not completely identical to the true absolute probe signal, it is similar, and it possesses the important unipolar qualities required in an absolute probe. The absolute signal shown in Figure 4c looks much more like a true absolute signal (shown in Figure 4a) than the absolute signal shown in Figure 4b. In Figure 4b, the absolute signal is composed of the self impedance from one coil minus the mutual impedance from the cross talk between the two coils. In Figure 4c the absolute signal is composed of the self impedance of the one coil plus the mutual impedance from the cross talk between the two coils. Because of results such as these (which had been observed previously by probe manufacturers and inspectors), such probes are now manufactured so that they operate with the drive current flowing in phase in both coils.

 

Experimental Verification
Figure 5a shows differential probe signals, measured in a laboratory, from an absolute/differential probe with the current in one coil 180 degrees out of phase with the current in the other coil. Figure 5b shows a differential anomaly signal, measured in a laboratory, from an absolute/differential bobbin probe operating with the drive current flowing in phase in both coils. The lower traces are strip chart displays of the respective imaginary (left) and real (right) components of the eddy current probe signals. The signals in Figures 5a and 5b are virtually identical, which is in agreement with the prediction of Equation 3.

Figure 6a shows a signal from a true absolute probe scanning an anomaly in a heat exchanger tube measured in a laboratory. The lower traces are strip chart displays of the respective imaginary (left) and real (right) components of the eddy current probe signals. Figure 6b shows an absolute signal, measured in a laboratory, from an absolute/differential probe with the current in one coil 180 degrees out of phase with the current in the other coil. Figure 6c shows an absolute signal, measured in a laboratory, from an absolute/differential bobbin probe operating with the drive current flowing in phase in both coils.

 

Summary/Conclusions
Absolute/differential probes should be operated with current in the two bobbin coils flowing in phase with one another to optimize absolute signals. Differential probe signals are independent of the relative sense of current flowing in the coils. That is, differential signals from a probe with the current flowing in the same direction in both coils are identical to signals from a probe with the drive current in one coil 180 degrees out of phase with current in the other.

 

Acknowledgments
The authors are indebted to William Lord, Nathan Ida, and Valentino Cecco for helpful discussions and comments. This work was supported by the CANDU Owners Group (COG) Work Package Identification and Release (WPIR) number 2029.

 

References
Cecco V.S., G. Van Drunen, and F.L. Sharp, "Eddy Current Manual," Vol. 1, AECL Report, AECL-7523, Chalk River Laboratories, 1983.

Ida, N., R. Palanisamy, and W. Lord, "Eddy Current Probe Design Using Finite Element Analysis," Materials Evaluation, Vol. 41, No. 12, November 1983, pp. 1389–1394.

Infolytica, MagNet 5, 2D ToolBox Reference Manual, Infolytica Corp., March 1995.

Palanisamy, R., and W. Lord, "Prediction of Eddy Current Probe Signal Trajectories," IEEE Transactions on Magnetics, Vol. 16, No. 5, September 1980, pp. 1083–1085.

Palanisamy R., and W. Lord, "Prediction of Eddy Current Signals for Nondestructive Testing of Condenser Tubing," IEEE Transactions on Magnetics, Vol. 19, No. 5, September 1983, pp. 2213–2215.

 

* Components & Systems Division, Nondestructive Testing Development Branch, Chalk River Laboratories, Chalk River, Ontario, Canada K0J 1J0; (613) 584-8811; fax (613) 584-4523; e-mail sullivans@aecl.ca.

 

Copyright © 2000 by the American Society for Nondestructive Testing, Inc. All rights reserved.

[ Materials Evaluation ]