Manufacturers seeking material with high corrosion resistance and an excellent surface finish often turn to ferritic stainless steel (FSS) for its exceptional forming properties in deep-drawing and stretch-forming processes. However, this can sometimes result in surface defects known as “ridging.” These are small parallel elevations and valleys that develop in the rolling direction when the sheet material is elongated (see Figure 1).
For some products, ridging will impact the aesthetic appeal, and polishing or buffing is required to remove it. Historically, the way to assess the severity of ridging was to examine test specimens visually. This method is both subjective and lacks repeatability. Therefore, to provide a more rigorous methodology, Outokumpu has developed a ridging index (RI) based on surface measurements made by profilometers.
Using the RI gives manufacturers a practical way to assess how the specific grade of FSS they plan to use will be affected by forming. An important point to consider is that it is not only the composition of the FSS that influences the likelihood of ridging. Many other factors are involved, especially how the sheet material was processed. In fact, nominally identical FSS samples produced by different manufacturers can show very different ridging behaviors.
Ridging causes
Ridging results from anisotropic (i.e., direction dependent) plastic deformation caused by colonies of different crystal orientations in the rolled sheet. This means that the chances of ridging taking place are largely predetermined during the casting process as this is when the crystal structure forms.
Ridging is at its maximum level when FSS sheet is strained along its rolling direction. It is absent when elongated along its transverse direction. Ridges are usually only a few millimeters apart. The increase in profile height that results from ridging may be up to ±50 microns, depending on the applied strain (see Figure 2).
Traditionally, ridging is assessed by straining 20-mm-wide tensile test specimens of FSS sheet to a defined elongation, typically between 7 and 20 percent. A visual inspection of the specimens is then carried out by trained personnel so that they can be rated. The results are often very subjective and lack repeatability.
Solving ridging
When ridging is an issue, mechanical polishing or buffing will be required. In extreme cases, the added cost could make the product economically unviable, causing it to be rejected.
The level of ridging can be reduced by adopting a special FFS alloy that contains niobium (Nb) such as Core 4622 (EN 1.4622). Another approach is to carry out an additional cold-rolling process on the FFS sheet before the forming operation.
A more rigorous approach, however, is to use a profilometer. Various types are available that use a sensitive stylus in contact with the surface or non-contact optical techniques. Typically, the arithmetic average of the roughness profile (Ra) and the maximum height of the profile are the measure for the intensity of ridging.
The raw profile recorded by the profilometer is often relatively short, covering only a few ridges. This profile is then filtered electronically, but this uses standards that were not developed for ridging and typically do not consider the spacing of the ridges. Therefore, the results may not be representative.
An improved method
Outokumpu has focused on finding a way to combine better sample geometry with a more suitable filtering method. Ultimately, this has been the basis for a calculated RI that provides information on the spacing of the ridges and their height.
Straining a 20-mm specimen by 15 percent reduces its width to about 17 mm, which is sufficient for a standard surface roughness measurement. However, it is not suitable for use with ridging because the distances between ridges are typically between 1 mm to 3 mm, so only four to 12 complete surface features are included. To achieve good statistical reliability, larger sheet specimens are used. They are typically 300 mm long and 100 mm wide.
In the RI development program, two types of profilometers were used: an optical 3-D white light interferometer and a non-skidded benchtop 2-D stylus. Electronic filtering was used to separate the ridging profile from the specimen shape, surface roughness and instrument noise (see Figure 3).
To achieve a smooth polished profile on the finished product, neither peaks or valleys are good. And the greater the difference in height between them, the more detrimental ridging becomes. This was measured by the term Rz that reflects the average maximum height of the profile.
It is important to measure the number of ridges and valleys. This is because as their number increases then so does the level of polishing required to remove them. This was measured by the peak count (Pc). The threshold set to determine Pc is indicated by the horizontal dotted line in Figure 3. The values for the Rz in microns and the Pc in mm-1 are multiplied to produce the RI, the dimensionless ridging index.
Testing the RI
To test the RI, surface profiles of FSS sheet samples were measured after elongations of 0, 2, 5, 10 and 15 percent. The graphs in Figure 4 indicate that the RI and Rz have a linear relationship with elongation, while Pc is nearly constant. Because the highest RI was seen to result from the highest elongation (15 percent), all subsequent tests were carried out at that level.
Profile measurements made on the top and bottom surfaces of a sample confirmed that the choice of the sampled side did not influence the measured RI. This is because the peaks on one side correspond to valleys on the other.
Repeatability is a critical factor in the RI methodology. To check this, an FSS sheet was cut into different specimens (see Figure 5a). Following multiple tests at 10-mm spacing on one sample (C0), an average RI of 4.4 was produced (see Figure 5b).
Figure 5c indicates the calculated RIs of all seven specimens. Four of them are similar to C0, the center sample. Two samples further out from the center of the sheet showed RI values about one unit higher, possibly indicating a locally different texture in the sheet.
Samples were measured with the 2-D stylus and the optical 3-D profilometer to determine if the method used influenced the results. The RIs generated are plotted against each other in Figure 6. The good correlation indicates that the RI determined is sufficiently independent of the technique used.
RI vs. visual assessments
For a final check, the RI was compared against the commonly applied rating based on visual inspection. A set of five samples from different batches was rated by five people, all trained and experienced in the visual inspection of stainless steel surfaces, against an arbitrary scale of 0 to 10 (see Table 1).
In some cases, the variation in ratings by the different surface inspectors was huge. Even so, the average result from the reference group and the RI measurements ranked the specimens in the same way. Furthermore, the average rating numbers correlated well with the RIs.
The RI is a repeatable and easy-to-use method for measuring ridging for both users and manufacturers of FSS. Outokumpu believes that its wider adoption is so important that the company is sharing the methodology freely.
