03. Theoretical Study of Near Field Suppression Coils in a Remote Field Eddy Current Probe

THEORETICAL STUDY OF NEAR FIELD SUPPRESSION COILS IN A REMOTE FIELD EDDY CURRENT PROBE

by

Sean Sullivan

Abstract

The basic Remote Field Eddy Current Technique inspection technique for pipes and tubes uses a bobbin transmit (excitation) coil with one or more passive receive (detector) coils axially separated from the transmit coil by a significant distance. In carbon steel tubes and pipes, the separation distance between the coils of a basic remote field eddy current probe is typically more than two tube diameters.

This report documents a theoretical study evaluating the influence “near field suppression coils” placed between the transmit and receive coils. The results of calculations documented in this report indicate that the use of a near field suppression coil enables transmit-receive coil probes to operate similarly as remote field eddy current probes with a significantly decreased distance between the transmit and receive coils.

Copyright of this paper is owned by Sean Sullivan

INTRODUCTION

The capability of performing through-wall eddy current inspection of carbon steel tubing and piping was developed in the 1950s with the development of the Remote Field Eddy Current Inspection Technique [1]. A clear explanation of the physical basis of the technique was published by Schmidt in 1984 [2]. The advantage of using the Remote Field technique is that the technique has nearly equivalent sensitivity to external and internal flaws (loss of tube wall material).

The transmit coil of a remote field eddy current probe is excited by a time harmonic current that generates an electromagnetic field in its vicinity. The internal electromagnetic field generated by the transmit coil diffuses through the tube wall to the exterior region of the tube and spreads out in the exterior region. At axial distances greater than two tube diameters from the transmit coil, the amplitude of the electromagnetic field outside of the tube is typically greater than the amplitude of the electromagnetic field in the interior region of the tube. In this remote field region of the tube interior, the electromagnetic field is generated by the transmit coil which diffuses to the outside of the tube, spreads out along the length of the tube and diffused back through the tube wall to the interior of the tube. As the electromagnetic field diffuses through the tube wall, in each instance, its amplitude is attenuated and phase is shifted by the tube wall in accordance with electromagnetic skin effect interactions with conducting barriers [2]. Figure 1 is a diagram showing the path of the electromagnetic diffusion for a remote field eddy current probe. The double through-transmission energy coupling path of the remote field technique was subsequently supported by Poynting Vector calculations that were performed using electromagnetic field data generated by Finite Element Modeling [3] and analytic solutions [4].

The excessive length of solid body remote field probes limits the capability of scanning curved tube sections like U-bends. Therefore, a method that enables the probe to function with a shorter length is desirable.

The influence of conducting shields (plates or rings) placed between transmit and receive coils has been evaluated experimentally [5] and by using finite element computer simulations [6, 7]. Results of these simulations have indicated that the use of these conducting shields can effectively attenuate the direct coupling component of the electromagnetic field generated by the transmit coil. This moves the boundary between the direct coupling zone closer to the transmit coil and allows for the construction of remote field eddy current probes with a shorter distance between transmit and receive coils.

However, the effectiveness of the conducting shield is dependent on its dimensions and the amount of space that it occupies inside the tube. Penetrations in the shield would be required for the necessary cabling. In addition, issues with tube ovality, denting and possible internal deposits force probe designers to limit the outer diameter of the probe and the conducting shield so that it can be passed through constricted tube sections when needed. Necessary gaps between the conducting shielding and the tube wall are required because of these practical issues. These gaps can reduce the effectiveness of the shielding [7].

This report documents a theoretical investigation of near field suppression coils on the design of remote field eddy current probes. Near field suppression coils may prove to be an alternative method of reducing the direct field coupling between transmit and receive coils allowing construction of remote field probes with a significant reduction in distance between the coils. Modern instrumentation can allow probes to be operating in multiple modes. Use of transmit/receive eddy current probes with near field suppression coils that can be activated and deactivated in different time-slots during operation can allow these probes to simultaneously operate as conventional near field and remote field transmit-receive eddy current probes.

THEORY

The theoretical basis used to evaluate the electromagnetic behavior of remote field eddy current coils in this report, is based on closed form solutions to Maxwell’s electromagnetic equations derived by Dodd, Cheng and Deeds [8] with modifications proposed by Fisher [4]. These solutions can be used to calculate electromagnetic field components and the magnetic vector potential generated by time harmonic current in the remote field eddy current probe’s transmit coil, and current in other field shaping coils, and they can also be used to calculate the signals induced in the bobbin receive coil of the remote field probe. The influences of the carbon steel tube wall thickness, electrical resistivity and magnetic permeability on the signal in the receive coil are accounted for by this model.

For bobbin receive coils, the signal voltage in each wire turn is directly proportional to the magnetic vector potential within the windings using equation (1) below:

In equation (1), V represents the voltage induced in a single turn coil of radius r. ω represents the angular frequency (2π times frequency) and Aφ represents the circumferential component of the magnetic vector potential. The total signal in a typical multi-turn bobbin receive coil is equal to the sum of the voltages in all of the wire turns.

BASIC REMOTE FIELD EDDY CURRENT PROBE FOR CARBON STEEL HEAT EXCHANGER TUBES

The analytic solutions to Maxwell’s electromagnetic equations were applied to simulating basic transmit-receive eddy current probes inside a typical carbon steel heat exchanger tube.

The outside diameter of the tube was 19 mm (0.75 inches) and the inside diameter of the tube was 15 mm (0.59 inches). The electrical resistivity of the tube material was 15 μΩ-cm and the relative magnetic permeability was 100.

The inner diameter of the transmit coil was 13.7 mm (0.54 inches). The outer diameter of the coil was 14.2 mm (0.56 inches). The length of the coil was 1.5 mm (0.06 inches). When calculating the signal in the bobbin receive coil, this coil had the same physical dimensions of the transmit coil. Calculations were performed using a test frequency of 500 Hz.

Figure 2 shows a graph plotting the amplitude of the magnetic vector potential inside the tube at a radial position of 7 mm (0.28 inches), and the magnetic vector potential outside the tube at a radial position of 9.6 mm (0.38 inches) as a function of axial distance from the centre of the transmit coil. Figure 3 is a graph plotting the phase of the magnetic vector potential inside the tube at a radial position of 7 mm (0.28 inches), and outside of the tube at a radial position of 9.6 mm (0.38 inches) as a function of axial distance from the centre of the transmit coil. These graphs are consistent with previously reported observations [2, 4, 5, 6, 7] regarding the remote field eddy current effect in tubes. In the near field region, the amplitude of the magnetic vector potential inside the tube decreases exponentially at a rapid rate as a function of distance from the transmit coil. In the remote field region, far from the transmit coil, the amplitude decreases at a much slower rate. In the remote field region, the rate of decay of the amplitude of the magnetic vector potential (and the magnetic field) inside the tube is equivalent to the rate of decay outside the tube. The transition zone between the near field and the remote field regions, between 25 and 30 mm from the centre of the transmit coil, is identified as the location where the amplitude inside the tube changes from a rapid exponential decay to a much slower decay. This location can also be identified as the region where the phase of the magnetic vector potential inside the tube changes abruptly as a function of distance (between 25 and 30 mm distance from the centre of the transmit coil). Based on these graphs, the design of a remote field eddy current probe would require a centre-to-centre coil separation distance to be at least 35 to 40 mm to ensure that the receive coil is in the remote field coupling zone.

Figure 4 is a graph showing a voltage plane display of tube wall thinning signals from a remote field eddy current probe with the centre-to-centre distance between the transmit and receive coils equal to 40 mm. The display has been oriented in the accepted configuration [9] for analysis of remote field eddy current signals. The display has been oriented so that the line between the base operating point of the probe in the tube and the zero signal point is horizontal. The signals from tube wall thinning from the inner surface of the tube are nearly equivalent, but not identical to the signals from thinning on the exterior of the tube. This demonstrates a shortcoming of the use of the one dimensional skin effect equation that is often used to predict remote field probe signals. The skin effect equation predicts identical sensitivity of the probe to wall loss from the internal or external surfaces of the tube. The more comprehensive model using the closed form analytic solutions to Maxwell’s electromagnetic equations used for this paper demonstrates that there is a slight difference in remote field probe responses to wall loss from the two surfaces of the tube wall.

REMOTE FIELD EDDY CURRENT PROBES FOR CARBON STEEL HEAT EXCHANGER TUBES WITH NEAR FIELD SUPPRESSION COILS

A diagram of a transmit-receive eddy current probe inside a carbon steel heat exchanger tube with a near field suppression coil is shown in Figure 5.

The inner diameter of the near field suppression coil was 13.7 mm (0.54 inches). The outer diameter of the coil was 14.2 mm (0.56 inches). The length of the coil was 1.5 mm (0.06 inches). Calculations were performed using a test frequency of 500 Hz.

A graph of the amplitude of the magnetic vector potential inside the tube plotted as a function of axial distance from the centre of the transmit coil is plotted in Figure 6 that shows the effects of different near field suppression coils. One near field suppression coil was placed at a distance of 5 mm from the centre of the transmit coil. This coil was energized with a time harmonic current that was 19% of the amplitude in the current in the transmit coil. The phase of the energizing current in this coil was 180 degrees with respect to the energizing current in the transmit coil.

Another near field suppression coil was placed 10 mm from the centre of the transmit coil. This coil was energized with a time harmonic current that was 3.7% of the amplitude of the current in the transmit coil. The phase of the energizing current in this coil was 65 degrees with respect to the energizing current in the transmit coil.

Figure 7 is a graph of phase of the magnetic vector potential inside the tube, plotted as a function of axial distance from the centre of the transmit coil. This graph contains phase curves from a basic remote field probe design, and designs with the two different near field suppression coils. This graph also indicates that the presence of these near field suppression coils has moved the remote field coupling zone much closer to the transmit coil. Based on the graphs in Figures 6 and 7, the use of these types of near field suppression coils would enable a remote field eddy current probe to be constructed with a centre-to-centre separation distance of about 25 mm between the transmit and receive coils. Without the near field suppression coils, the 25 mm separation distance would place the receive coil either in the near field zone, or the transition region between the near field and remote field zones.

Figure 8 is a graph showing a voltage plane display of tube wall thinning signals from a transmit/receive eddy current probe with a near field suppression coil placed 5 mm from the centre of the transmit coil. The centre-to-centre distance between the transmit and receive coils was 25 mm. The display has been oriented in the accepted configuration [9] for analysis of remote field eddy current signals. The display has been oriented so that the line between the base operating point of the probe in the tube and the zero signal point is horizontal. The signals from tube wall thinning are similar, but not identical to the signals from a true remote field probe (shown in Figure 4). The reason for the differences in the thinning signals from the two probes is that the changes of geometry due to the thinning has affected the magnetic shielding interaction of the near field suppression coil which has had an affect on the overall thinning signals. The disadvantage of these signal differences is that they would create the necessity of added signal analysis training of inspectors with experience in using remote field techniques if they desire to use the modified remote field technique with the shorter probes. The advantage of these differences is that they might allow eddy current inspectors to improve their capability to differentiate between signals from wall loss on the interior or exterior surfaces of the tube wall.

SUMMARY/CONCLUSIONS

A theoretical study has been performed to evaluate the influence of two simple near field suppression coil designs placed between transmit and receive coils of eddy current probes

designed for inspecting carbon steel heat exchanger tubes. The results of calculations documented in this report indicate that the use of near field suppression coils enables transmit/receive coil probes to operate similarly as remote field eddy current probes with a significantly decreased distance between the transmit and receive coils. Further studies to optimize the near field suppression coil designs would likely provide improved results.

REFERENCES

1. MacClean, W.R., U.S. Patent 2 573 799 (November 6, 1951).

2. Schmidt T.R., “The Remote Field Eddy Current Inspection Technique”, Materials Evaluation, Vol. 42, No. 2, February 1984, pp 225-230.

3. Atherton D.L. and Czura W., “Finite Element Poynting Vector Calculations for Remote Field Eddy Current Inspection of Tubes with Circumferential Slots”, IEEE Transactions on

Magnetics, Volume 27, Number 5, pp 3920-3922.

4. Fisher J.L., “Remote Field Eddy Current Model Development”, Southwest Research Institute Report, SwRI Project 17-9431, January 1987.

5. Atherton D.L. and Sullivan S., “The Remote-Field Through-Wall Electromagnetic Inspection Technique for Pressure Tubes”, Materials Evaluation, Volume 44, Number 13, December 1986, pp 1544-1550.

6. Atherton D.L., Szpunar B. and Sullivan S., “The Application of Finite Element Calculations to the Remote Field Inspection Technique”, Materials Evaluation, Volume 45, Number 9,

September 1987, pp 1083-1086.

7. Atherton D.L., Czura W. and Schmidt T.R., “Finite Element Calculations for Shields in Remote-Field Eddy Current Tools”, Materials Evaluation, September 1989, pp 1084-1088.

8. Dodd C.V., Cheng C.C. and Deeds W.E., “Induction coils coaxial with an arbitrary number of cylindrical conductor”, Journal of Applied Physics, Volume 45, Number 2, February 1974.

9. Atherton D.L., Mackintosh D., Sullivan S., Dubois J.M.S. and Schmidt T., “Remote-Field Eddy Current Signal Representation”, Materials Evaluation, Volume 51, Number 7, July

1993, pp 782-789.