While eddy current technology was initially developed for testing of bar, tube, and wire, advances in electronics, automation, and coil design have paved the way for a new generation of systems specifically designed to test critical components, such as gears, that go into automotive and industrial applications. These eddy current test systems not only detect cracks in components, but also verify proper heat-treat conditions, proper alloy composition or material structure, and the presence or absence of features such as splines or teeth.
Eddy current systems are fast, clean, repeatable, and easy to integrate into production processes allowing for in-line testing at production line speeds. In addition to enabling 100% of production components to be inspected, eddy current testing helps to monitor upstream processes notifying operators that something is not functioning correctly on the production line. This greatly reduces scrap and warranty costs for gear manufacturers.
Eddy current testing is often used to replace chemical testing processes such as dye penetrant and magnetic particle inspection for crack testing, and cutting and acid etch testing to verify proper heat treat patterns. These chemical tests are typically performed off-line after a production batch has been run and is subject to operator skill and error. In the case of cutting and acid etch testing, the part is destroyed. Eliminating or reducing off-line chemical testing allows companies to reduce the cost of chemical purchases, reduces hazardous wastes, and allows companies to “go green.”
Eddy Current 101
Eddy current testing is an electromagnetic, non-destructive testing method that measures the flow of eddy currents in a conductive material. Changes in the material due to cracks or “structure” changes, such as heat treatment or alloy variation, cause the eddy currents in the material to flow differently. This difference is detected by an eddy current instrument.
Figure 1 shows how eddy currents are generated and flow in a test scenario. An AC signal generator drives an electrical current through an eddy current coil, generating a magnetic field around the coil. As the coil is held in close proximity to the component under test, the magnetic field induces eddy currents to flow in the component. These eddy currents, which change their flow around cracks or flow differently in different alloys or heat treat conditions, create their own magnetic field which is then sensed by the same coil or by a different coil.
Figure 2 shows a block diagram of an eddy current test system. A signal generator in the real-time signal processing block sends a single frequency or multiple frequencies through a digital-to-analog converter (D/A) out to a single coil or multiple coils where magnetic fields are generated. The eddy current signals are measured by the same coils or different set of coils and are digitized by an analog-to-digital converter (A/D). The real-time signal processing electronics compare the returned signals with the sent signals and determine whether an “alarm condition” has occurred. If an alarm condition has occurred, a signal is sent to a programmable logic controller (PLC) within a material handling station to reject the part from the assembly line. Testing data can be sent to the factory MES or quality systems to track trends and store data. This data can be stored on-premise or in the cloud. Modern touch panel displays greatly simplify user interfaces and allow operators flexibility in setting up their production lines.
Crack Testing
Eddy current crack testing is performed by passing a pair of coil windings over a section of the component to be tested. These coil windings can be made small enough to test between gear teeth, and with multi-coil probes can test very complex shapes. Most crack test applications use one test frequency, as they only require detection of surface flaws. Typical test frequencies range between 10 KHz and 4 MHz with higher frequencies used to find smaller surface flaws. Simultaneous testing with multiple frequencies allows for testing of both surface and sub-surface defects when inspecting non-ferromagnetic parts.
An easy-to-see crack test application is shown in Figure 3. In this case, the carrier gear experienced cracking upon assembly. The probe assembly shown in Figure 4 was used to detect the flaws. This probe uses multiple coil assemblies (the white blocks) so that the testing fixture did not have to rotate 360 degrees for testing. This made for a simpler material handling station.
Heat Treat Verification
While not an absolute hardness test like a Rockwell test, eddy current heat treat verification can achieve sorting results on par with Rockwell testing. This has been demonstrated with both forged and powder metal gears. Eddy current heat treat inspection coils come in both standard encircling coil configurations and multi-coil custom configurations. The custom configurations allow for precise location testing verifying that induction heating parameters were correctly applied. Defects to be tested include misplaced case, shallow case, short quench, delayed quench, air cooled, non-heat-treat, and ground out conditions.
When performing heat-treat inspection, multiple test frequencies are used to reliably detect these various heat treat anomalies. A range of eight frequencies is commonly used with the ratio of the lowest to highest frequency of 1:1000 or more. Structure defects or anomalies are usually discoverable within those frequency ranges. Using an eight frequency test also makes setting up a test much easier.
Figure 5 shows a heat treated reaction internal gear that needed to be heat treat tested. On the left hand side is the custom designed eddy current probe used to test the gear. The two eddy current coil windings are clearly seen as bands in the white nylon housing. Two areas on the gear were tested using this set up.
Figure 6 shows a good example of a complex heat treat pattern in a cut away wheel spindle bearing sample. This sample was prepared using an acid etch process and clearly shows the heat treat pattern. The probe used to test the bearing is shown in Figure 7 and has four individual coils to verify proper placement and depth of heat treat at various locations on the part. The visible coil winding seen on the bottom of the probe inspects the bottom flange area for proper run-out. The metal components of the probe are made of stainless steel to increase the longevity of the probe.
Figure 8 shows a detail of the material handling station used to test the wheel spindle. The probe comes down onto the wheel spindle to perform the test. The actual test takes fractions of a second to perform. If the part is acceptable, it passes along the conveyor to the next station. If it fails, the eddy current instrument sends a signal to the material handling PLC to send the part to a reject chute. If multiple parts are rejected in a row, the eddy current instrument signals the operator who can check for process issues.
Figure 9 shows the material handling station in the production line. It is located just downstream of the heat treat processing station.
Figure 10 shows a set of powder metal gears that were tested just downstream of a heat treat furnace for proper hardness. The test was designed to inspect 100% of the components up to 60 parts/minute. The test was designed using eddy current encircling coils and automated sorting chutes which were designed to accommodate different sizes of rings and stars with minimal set up changes. Figure 11
In order to develop a correlation between the eddy current and Rockwell hardness (HrB) test methods, two tests were conducted. 380 samples (38 sets of 10 samples) were tested with a Rockwell hardness tester and the eddy current test system. Improperly hardened specimens were included in the test batch.
The initial test results showed a very good correlation between the eddy current and Rockwell hardness tests (see Figure 3). These results indicated between 1 and 3 HrB points of variation. The eddy current test reading shown in blue, were more consistent than the Rockwell test results. A sample production test of 35,000 parts showed that the eddy current readings were more consistent than the sample Rockwell tests.
Figure 12 shows a flange area of a full-length axle shaft with splines on the other end. The application was to verify that proper heat treating was applied in the area indicated. Figure 13 shows the results of a multi-frequency test with grouping of the results from different conditions. Eight test frequencies ranging from 500 Hz to 5 KHz were used for this test.
Eddy current testing can also verify that proper materials were used in fabrication. Figure 14 shows a drive gear and eddy current encircling probes. This test was put in place to verify that a supplier was delivering gears with the proper alloy.
Assembly Testing
Eddy current testing can also be used to verify proper feature manufacturing including detection of threads, broaches and splines. It can also test for proper assembly of subcomponents. An eddy current test was designed to verify that all the needle bearings used in the gear assembly shown in Figure 15 were present.
Summary
Eddy current testing offers fast, repeatable testing of gears and other critical metal components. Testing data on each component can be stored electronically and re-analyzed off-line at a later date. Eddy current test instruments are designed to integrate with PLC’s in material handling stations to set up real-time rejection capabilities. These are all features required in today’s modern manufacturing environment.
Printed with permission of the copyright holder, the American Gear Manufacturers Association, 1001 N. Fairfax Street, Suite 500, Alexandria, Virginia 22314. Statements presented in this paper are those of the authors and may not represent the position or opinion of the AMERICAN GEAR MANUFACTURERS ASSOCIATION.
