Back to Basics

[ click here for the Back to Basics Archive ]

Using Eddy Current Testing to 
Solve Industrial Problems

by John Hansen* and Ronald B. Peoples+

 

This month we have a "Back to Basics" article on conventional eddy current testing, which relies for its basic principle on the material properties of electrical conductivity, magnetic permeability and dimensions. Systems and certain applications are discussed.  Roderic K. Stanley Associate Technical Editor

 

Proper performance and compliance of manufactured components require verification of certain component attributes. Dimensional tolerances, metallurgical integrity and proper assembly are examples of important component characteristics. For safety critical applications, verification of 100% of manufactured components is desirable.

Eddy current testing is often applied for surface crack detection, verification of heat treat condition or material type, to ensure that the proper materials are in use and to verify certain component or assembly features such as the orientation or position of a subcomponent in an assembly.

Eddy current testing does not require physical contact between the sensor and the test material. Sensor scanning speeds of several meters per second are common. With the eddy current technique, 100% nondestructive testing (NDT) is possible. For this reason, eddy current testing is often used to verify certain characteristics of manufactured components.

 

PRINCIPLE OF OPERATION
The basic principle of the eddy current technique is well known. An inductive sensor is excited with an alternating current. The alternating current flowing through the sensor generates an alternating magnetic field with the same frequency as the excitation current. When the sensor is brought into close proximity of an electrically conductive material, current flow is induced in the material (as per Lenz's law). The alternating current flowing in the material generates a secondary, opposing, alternating magnetic field that interacts with the primary magnetic field. The resultant interaction between the primary and secondary magnetic fields is commonly represented as a polar or linear display of the test coil's complex impedance.

---

Eddy current testing can be applied to a wide range of industrial applications.

---

In practice, the test coil is balanced on a discontinuity-free section of the test material. As the sensor is scanned across a homogeneous section of the test material, the impedance of the test coil remains relatively constant. When the test coil encounters a discontinuity in the test material (such as a crack or a change in electrical conductivity or magnetic permeability), the interaction between the magnetic fields generates an apparent change in test coil impedance. Varying instrument test parameters such as gain, phase, test frequency and filter frequency optimizes the indication resulting from the impedance change caused by the discontinuity. Optimization is aimed at maximizing the signal to noise ratio (the amplitude relationship of the reference discontinuity to the noise).

An eddy current test coil's impedance is influenced by three primary variables: the electrical conductivity of the test material; the magnetic permeability of the test material; and certain geometric and positioning relationships between the test coil and the test material.

In practice, geometric and positioning variables are controlled by the design of the sensor and the material handling equipment. Eddy current systems are designed to maintain constant positioning and spacing between the sensor and the test material. Since variations in test coil positioning and spacing have a major effect on test results, periodic system verification is performed using a reference standard which generates a known indication. Test coil positioning and test parameters can be varied to resolve the reference indication.

Three types of eddy current sensors are commonly used (Table 1). Probe coils allow for localized testing of materials and are generally very sensitive to minor surface imperfections and localized changes in electrical conductivity and magnetic permeability. Probe coils are particularly sensitive to variations in test coil to test piece spacing (liftoff). The liftoff variable must be closely controlled to ensure consistent test results.

Table 1 Eddy current sensor types
Sensor Type	Application	Considerations
Probe coil	surface crack detection localized metallurgical variations sensing applications	position must be carefully controlled small probes lose sensitivity quickly as liftoff increases  must be protected from being damaged by the test material
Encircling coil	surface crack detection inner/outer diameter testing of tubing from outer diameter heat treat verification material type verification	test pieces must maintain constant position inside test coil sensitivity decreases as test coil size increases must suppress signals from the ends of the test pieces
Bobbin coil	inner/outer diameter testing of tubing from  inner diameter  testing of bore holes	sensor must maintain consistent position inside test piece  sensor must fit well inside test piece test piece inner diameter must be clean and free of obstructions

Outside diameter (torroidal) encircling coils allow for testing of the entire outer diameter circumference of the test material. The sensor diameter is chosen based on the outside diameter of the test material and the test material must be centered in the test coil to ensure consistent sensitivity around the circumference. For heat treat and material type verification, the test material need not be centered in the test coil but positioning should be consistent for subsequent test pieces. Inside diameter (bobbin) type coils allow for testing of the inside diameter of manufactured components and the inside diameters of manufactured and inservice tubes. Positioning is again critical to consistent test results.

For nonferrous materials, variations in metallurgical structure or the presence of discontinuities in the test material generate a corresponding change in electrical conductivity. Through the use of a representative reference standard, electrical conductivity changes generated by discontinuities can be characterized on the impedance plane. For ferrous materials a corresponding change in magnetic permeability is also apparent.

When testing manufactured components for cracks, eddy current testing is most suited for detecting discontinuities that are open to the surface of the test material. Test frequencies in the range of 500 kHz and higher are often used to provide high sensitivity to shallow surface discontinuities. When deeper surface discontinuities are expected, the test frequency can be reduced to provide better linearity for indications in the range of anticipated discontinuity depths.

For heat treat and material type verification, lower test frequencies provide deeper penetration and a more volumetric view of the test material. Physically larger sensors are often used for these applications. These larger coils are less sensitive to minor surface imperfections that can create noise signals that interfere with the ability to measure bulk material properties.

Eddy current testing, like most nondestructive testing techniques, is a comparative technique. System calibration using a representative reference standard is crucial to obtaining meaningful test results. When using automated testing systems to test manufactured components, calibration techniques commonly result in a go/no go sorting decision that removes anomalous components from the process.

 

SYSTEM CONFIGURATION - TEST TECHNIQUES
Three commonly used test techniques are manual testing, testing with the sensor installed as an integral part of an existing manufacturing process and testing using a specially designed "turnkey" test system that can be installed either as part of an existing process or as a standalone "offline" testing system (Table 2).

Table 2 Eddy current system installation
Installation	Advantages	Disadvantages
Manual	low investment for test equipment   necessary for larger components  appropriate for low volume production	operator must control test variables operator makes go/no go decision slow throughput speed
In-process	minimum investment in material handling equipment      real time test results can lead to actions that minimize  occurrences of process-induced discontinuities  fully automatic operation can provide more consistent  test results	process design might have to be modified to accommodate installation does not verify that test piece quality is  consistent after subsequent processes correct operation must be periodically verified
Offline	can test product from more  than one processing line delays do not affect production rates can be used to verify product quality as the last step prior to shipment	mechanical system design must be more complex  to accept a wide range of test piece geometries test results not real time - limited opportunities for process control  adds an operation to the processing path

Manual testing is sometimes suitable for low volume testing. The disadvantage to manual testing is that it requires that the operator control the key test variables (such as sensor positioning and scanning speed), which can have a negative effect on the test result. Throughput speed is normally lower when compared with automated testing. In many cases, manual testing is more subjective than automated testing since reliability depends primarily on the techniques used and the judgments made by the operator.

In-process testing is often very cost effective due to the relatively small investment required for mechanical automation. The sensor must be installed in a position in the process where the test material's throughput speed and motion is well controlled. For in-process testing, it is often necessary to install a test fixture that is designed to maintain constant sensor positioning. The process should allow for segregation or paint marking of anomalous test pieces. Examples of in-process installations include encircling coil testing of wire in the drawing process, testing of the weld zone of welded tubing in the welding process, heat treat verification of manufactured components after induction hardening and surface testing of manufactured components during machining operations.

Turnkey test systems are often integrated as an operation in an existing process. Test pieces are delivered to the test system by a conveyor or robotic pick-and-place system. The test pieces are automatically oriented and scanned using the eddy current sensors. Anomalous test pieces are removed from the process by appropriately designed reject gates. The advantage to in-process testing is that 100% of the components produced by the process are tested. If test piece geometry varies within the process, considerations must be made to adapt the sensor system to the varying test pieces.

Offline test systems operate as standalone systems apart from the manufacturing process. The test pieces can be introduced either automatically (by material handling) or manually (fed by the operator). Testing is performed automatically and anomalous test pieces are segregated and identified for removal from the test process. Although offline testing normally involves additional material movement within the manufacturing facility, it has the advantage of being able to be used to test components from a number of individual manufacturing processes.

 

APPLICATION EXAMPLES
As previously mentioned, eddy current testing is often applied for surface crack detection, verification of heat treat condition or material type, to ensure that the proper materials are in use and to verify certain component or assembly features such as the orientation or position of a subcomponent in an assembly. Examples of actual applications follow.

Surface Crack Detection
Surface cracks can be created by many factors including discontinuities present in raw materials, abusive grinding or machining, and thermal stresses created during heat treating and quenching. For many applications, particularly when catastrophic failure of the component can occur under loading, the presence of surface cracks is detrimental to the service of the component.

Safety critical components used in the automotive industry are particularly prone to catastrophic failure when surface cracks are present. Premature failure of components used in braking, suspension and wheel systems can have dire consequences and therefore must be verified prior to use. Premature failure of drive train components such as axles, transmission shafts, valves and engine blocks generate warranty claims and sometimes expensive recalls for evaluation and replacement, often at the manufacturer's expense.

Raw materials such as bar, wire and tubing used to manufacture components are often subjected to eddy current surface testing prior to introduction into the manufacturing process. Depending on their geometries, these materials can often be tested prior to shipment by the primary metals producer or as an incoming test by the component fabricator. Eddy current encircling coil, rotating probe and rotating test piece systems are all commonly used to verify the surface quality of these components.

During processing, manufactured components can be eddy current tested at various positions in the manufacturing processes. Cold formed components are most susceptible to failure during initial processing due to the high mechanical forces imparted during cold forming. Testing after cold forming but prior to machining is common to verify that the incoming "blanks" are free of critical discontinuities.

In-process systems are used for verification during processing. Integral sensors are normally positioned in a location in the process where the motion of the test piece is both predictable and consistent. Testing of valve lifters during surface honing is an example of an application for an integral sensor. Turnkey test stations are sometimes integrated into a process and provide the consistent motion required to allow for reliable testing. An example of an integrated turnkey system is a test station that provides for simultaneous surface testing of formed wheel hubs using a multichannel eddy current system.

Offline testing stations are used as a final verification of surface quality. These systems can be manually or automatically fed and are often used to verify that the finished product complies with surface quality specifications. A disadvantage associated with offline systems is that the discontinuities are detected at some time interval after processing, making real time adjustments to the process more difficult.

Detection of surface discontinuities as early as possible in the manufacturing process is often most desirable to avoid further investment in the materials and to help to guarantee final yields. Whenever possible, in-process systems should be deployed to maximize cost effectiveness.

Heat Treat/Material Type Verification
While verification of material type is often accomplished by the primary processor using spectrometry or other techniques, it is sometimes necessary to verify the materials at the beginning of the manufacturing process. An absolute type eddy current system can be characterized on a sample of known material and a go/no go verification can be realized in some cases. Reliability of such a system requires that: the normal variation in the eddy current response for the material under test is known; the eddy test parameters are optimized using samples of all potential nonconforming materials; and critical test variables are controlled during testing. Applications must be optimized on a case by case basis and risk areas must be understood and accepted.

Heat treat verification can be accomplished directly after the thermal treating processes, integral to a subsequent process, or offline either before or after subsequent processing. Variables such as test piece temperature, positioning and throughput speed must be well controlled to ensure consistent test results.

For nonferrous materials such as aluminum and copper, eddy current electrical conductivity measurements are often used to verify material characteristics. Raw materials determine the purity of copper products. Electrical conductivity measurements can be conducted after the primary melting and rolling processes to ensure that the raw copper has the correct purity.  The electrical conductivity of aluminum products can be measured after thermal processing to verify that the correct metallurgical structure is present. Thermal processing of aluminum affects the metallurgical structure of the material, which in turn affects the electrical conductivity. Published charts list expected electrical conductivity values for the various heat treat conditions of aluminum alloys.

For ferrous materials, case carburizing, thermal treatment and normalizing determine the microstructure of the materials. Variations in microstructure and surface carbon can influence the electrical conductivity and magnetic permeability of the materials. Using an absolute type eddy current test system, conforming product can be characterized while optimizing test parameters by comparing known samples of nonconforming product. Manual, semiautomated or automated test systems can then be implemented to obtain go/no go sorting decisions based on the optimization using known parts.

Determination of Certain Product Features
For the purposes of this paper, these applications will be referred to as "sensing applications." The objective of sensing applications is often to verify certain test piece features such as correct geometry, correct assembly or the presence of a specific geometric feature of the component or assembly.

A common area of application for eddy current sensing is to verify the presence of threads on threaded components such as bolts and other fasteners. For this application, a suitable sensor is constructed and the test results are optimized by varying the test parameters while comparing correctly and incorrectly threaded test pieces. Often the objective of the calibration is to reliably separate threaded parts from unthreaded and partially threaded parts. Automated testing is normally implemented with each part generating a go or no go decision.

Other sensing applications include verification of correct orientation and location of components (for example, springs, clips, collars and inserts) within an assembly, verification of the presence and correct location of component features, such as bores, grooves and edges, butt weld detection in strip stock for a welded tube manufacturing process, and counting of individual components as they pass by an installed sensor.

In some cases, sensing applications can be combined (using a common sensor) with crack detection or heat treat verification installations. Multichannel test instruments allow for separate discrete outputs from the discontinuity detection and sensing channels.

Conclusion
Eddy current testing can be applied to a wide range of industrial applications. The most common areas of application are crack detection, heat treat verification and verification of certain test piece/assembly features. Testing can be performed manually, but the use of an automated or semiautomated test system can increase reliability and throughput. Sensor design and control is critical to maintaining a reliable test technique. Since eddy current testing is a comparative test method, representative samples of test material and a reference standard are critical to establishing a valid test.

* GE Inspection Technologies, 129 - 135 Camp Road, St. Albans AL1 5HL, England.

 ⁺ GE Inspection Technologies, 50 Industrial Park Road, Lewistown, PA 17004; (717) 447-1365; e-mail <ronald.peoples@ae.ge.com>.

Copyright © 2006 by the American Society for Nondestructive Testing, Inc. All rights reserved.