The benefits of analyzing an induction-heated alloy pin using both hardness and advanced imaging techniques are explored.

Enlarged Image


Fig. 1. Image of the whole sample generated by the scanner

Enlarged Image


Fig. 2. Schematic drawing of the heating coils specially shaped for induction heating of a small U-shaped piece


In some specific applications, the information generated by a single line of indents on a heat-treated sample is not enough. For instance, the proportion of heat-affected area to total sample as well as the average depth of the heat-affected zone can give valuable additional information. The hardness-testing instrument used for this analysis was a Clemex CMT.HD packaged with a commercial flatbed scanner and Clemex Vision Lite Image Analysis software.

For this study, the sample is a U-shaped guide-pin made of a specialized alloy and used in the static part of a large motor. Induction heating allows for precise control of the heat-treated zone, which is needed for such a small piece. Beside the original hardening specs, the difference between the treated and non-treated zones must not exceed 30 HRC (Rockwell C). Also, the area percent of soft area must not be greater than 28%, and the thickness must be at least 650 microns and no more than 2,600 microns.

Enlarged Image


Fig. 3. Stage pattern window showing the position of the traverse on the sample

Induction Heating

Induction heating is a fast, consistent and easily controlled process used to bond, harden or soften metals. An important use is after metalworking, heat treating, welding or metal melting to restore ductility. One advantage of induction heating is that no torch or flame is applied directly to the metal parts. Instead, induction heating occurs when an electrically conducting object, usually metallic, is placed in an electromagnetic field. Eddy currents are generated in the metal by the magnetic field, and the material’s own resistance leads to Joule heating. In this way, there is no contact with an exterior heat source since the heat is generated in the material to be heated itself and not in the surrounding area – except by radiation. Therefore, there is no product contamination. Figure 2 illustrates how induction heating can be used on a U-shaped piece.

Another advantage is the ability to control the area that is being heat treated. Firstly, a general characteristic of alternating currents is that they are concentrated on the outside of a conductor in what is called the “skin effect.” This effect is characterized by its penetration depth, which is defined as layer thickness. Penetration depth depends on the characteristics of the material to be heated and is influenced by the frequency (in Hz) used during induction heating.

Knowledge about the material to be heat treated (e.g., thickness, resistance and composition) is essential. Once all parameters are known, penetration-depth formulas are used to determine the power needed to increase or decrease the magnetic field to achieve the desired results. Unusually shaped pieces, such as U-shaped connectors or hollow tubes, can also be heat treated with the help of specially shaped inductors.

Enlarged Image


Fig. 4. HRC results curve


The sample was scanned in order to get a global view of the area to analyze and position the indent traverse. The image was imported to the software stage pattern window and used to precisely position the location of the traverse where the indents would be made.

Figure 3 shows the scanned image of the whole sample in the stage pattern window with a yellow line drawn for the location of the traverse. After the execution and measurement of the series of Vickers indents, the hardness values are converted to HRC scale. A curve is displayed in the result window, which shows X-Y coordinates and the hardness value for each of the nine indents that were made at the 500-gram load.

Enlarged Image


Fig. 5. a. Binarized area percent; b. Results of area percent


The following steps were executed on the scanned image of the sample at a resolution of 9,000 microns per pixel. Figure 5a shows the binarized image for the area percent measurement. The results (Fig. 5b) show that the inner zone (Bitplane 2, red) represents 26.6% of the total area and the outer zone (Bitplane 1, blue) represents 73.4%. Hence, the required ratio is respected.

For the measurement of the variation of thickness of the heat-affected zone, the average length measurement was used. For the thickness of the inner zone, 20 lines were drawn in red (Fig. 6a) and measured automatically for length. The results in Fig. 6b show that the average thickness is 1,361 microns with the thinnest and thickest values of the zone varying between 798 microns and 2,442 microns. Here again, the requirements are met.

Enlarged Image


Fig. 6. a. Thickness measurements of the soft area; b. Results of thickness measurements


The more information a given sample can yield, the more knowledge one gets out of that sample. Usually when testing for hardness, the machines used had only one particular function and that was to test the hardness of a given material or sample. Any other analyses needed to complete one’s understanding of the sample had to be performed on any number of other systems.

Even basic analyses such as thickness measurements could not be performed using a common microhardness tester. The arrival of the Clemex CMT.HD has changed this. Using the tester as a microscope, the CMT.HD in combination with a common flatbed scanner acquires an image of the sample that can then be used for both positioning it for better precision in hardness testing and to easily analyze other aspects of the sample. IH

For more information: Contact Elizabeth Haslinger, sales coordinator, North America, Clemex Technologies Inc., 800 Guimond, Longueuil (Québec) J4G 1T5; tel: 450-651-6573 x55; fax: 450-651-9304; e-mail:; web:

Additional related information may be found by searching for these (and other) key words/terms via BNP Media SEARCH at image analysis, induction heating, electromagnetic field, Joule heating, microhardness, inductor