Eddy current testing

from Wikipedia, the free encyclopedia

The eddy current testing (engl. Eddy current method ) is an electrical method for non-destructive material testing . It is used to test electrically conductive materials.

principle

During the test, an alternating magnetic field is generated by a coil , which induces eddy currents in the material to be examined . During the measurement, a sensor , which usually also contains the excitation coil, detects the eddy current density through the magnetic field generated by the eddy current . The measured parameters are the amplitude and the phase shift to the excitation signal. A second coil in the sensor is usually used to measure it. One then speaks of a fluxgate magnetometer , which in German-speaking countries is also known colloquially as a Förster probe . Occasionally, other magnetic field sensors such as GMR sensors or SQUIDs are also used.

The eddy current test exploits the effect that most of the impurities and damage in an electrically conductive material also have a different electrical conductivity or a different permeability than the actual material.

Since the measuring signal is determined by the three parameters conductivity, permeability and distance between the detector and the material surface, eddy current testing has three different areas of application:

Crack testing

In crack testing, the sensor is moved over or through the object to be tested. As long as there is no damage in the material, its electrical resistance is also homogeneous and the eddy currents flow evenly in the material. If the test part has, for example, an inclusion of a foreign material, the specific resistance of which is lower than that of the rest of the material, the electrical current density in the inclusion will be greater than in the surroundings. The opposite is true for an inclusion with a higher specific resistance or a hairline crack around which the current has to run. In any case, the eddy current density changes compared to the undamaged component. In this test, sensors are used whose coils are switched in such a way that small changes in the material properties or the distance between the sensor and the material surface are largely compensated for.

Layer thickness measurement

A distinction is made between the following cases for coating thickness measurement (based on):

  • non-ferromagnetic, electrically conductive plus insulating layer on a ferromagnetic material (typically iron)
typical application for the magnetic inductive method
  • non-electrically conductive, non-ferromagnetic layer on a non-ferromagnetic metal
here the distance between the sensor and the conductive surface, which is determined by the thickness of the coating, is determined by measuring the amplitude.
  • electrically conductive foil / sheet made of non-ferromagnetic metal
With increasing thickness, the cross-section through which the current flows increases, and thus also the amplitude of the signal, the evaluation of which results in the thickness.
  • electrically conductive non-ferromagnetic layer on a ferromagnetic base material, but hidden under paint
Combined magnetic inductive and eddy current method with phase evaluation
  • electrically conductive, non-ferromagnetic layer on any base material
Eddy current method with phase evaluation
Examination of the material properties (structure examination)
  • Changes in conductivity or permeability are used to determine material conditions, hardness , heat treatment , detection of weld seams or to check for mix-ups.

By changing the frequency of the excitation voltage, the penetration depth of the eddy current changes ( skin effect ), which means that it can be adapted to the test conditions.

Practical implementation of a sorting test for structure

The aim here is to explain how parts can be checked for structure or material mix-ups and sorted.

It is the state of the art today to test the parts with several frequencies. The test setup is often such that two pairs of coils (each consisting of a transmitter and receiver winding) are connected to the transmitter windings in the same direction and the receiver windings in opposite directions. In one pair of coils (the compensation coil), a good part is firmly positioned as compensation in order to design the type of output signals so that in the case of a good part the level remains in the range of 0 V; This makes it all the easier to record the deviations in bad parts, be it as a deflection on a display device ( oscilloscope ) or for digital processing ( A / D converter and digital signal processor ), as is usual with current devices. The other pair of coils is loaded with the parts to be tested by hand or automatically.

Before a sorting test can begin, a number of known good parts must first be stored in the test device as a reference. Certain frequencies are selected in tests and through the experience of the user with which the best separation conditions can be achieved for the application. This can also be done automatically, using the widest possible frequency band with z. B. eight test frequencies and z. B. a frequency ratio of 1: 1000 is used, a typical frequency band would be e.g. B. 25 Hz to 25 kHz. The “response” of the parts is saved for each frequency, from which the test device then determines the tolerance fields in which the parts to be tested must lie. If this range is not reached even for one frequency, then the part must be sorted out as a bad part.

This type of testing of the structure of a material is very sensitive to even minor deviations. By using several frequencies, unexpected errors are also well recognized and yet the throughput is very high. Depending on the type of inspection and the size and geometry of the parts and the required accuracy of the sorting, up to approx. Ten parts per second can be checked and sorted.

Manual testing is particularly useful for random samples, trials and small subsets. For a 100% inspection, an automatic system is usually integrated into the production line. Ideally, the previous throughput is not reduced by adding such a system, which means quality assurance without loss of productivity.

A distinction must be made between dynamic and static testing, especially with automated testing. The dynamic test allows higher throughput, which is bought with lower accuracy. The parts are transported through the coil pair in a continuous flow (either individually or adjacent to one another), while the position of the parts is monitored by suitable sensors in order to start and end the test at the right time. With many types of parts and material combinations to be tested, this procedure is generally impossible because the separation is too imprecise. Dynamic testing is often used to check balls (for ball bearings) to identify the correct heat treatment or surface hardening. A pair of through-coil coils is not used here, but a probe probe without a compensation coil facing the test objects from one side. Instead, there is a compensation winding in the sensor housing.

During the static test, the test item is stopped in the sensor coil and tested. Although the actual inspection process does not take longer, stopping the parts can significantly reduce throughput.

See also

Web links

Individual evidence

  1. ↑ http://www.helmut-fischer.com/fileadmin/user_upload/default/Brochures/de-german/BROC_PMP10Duplex_911-025_de.pdf Company publication Helmut Fischer GmbH