Validating the Technical Cleanliness of Ball Bearings

For components to perform correctly in finished products they must be technically clean. For this reason, being able to assess the cleanliness of components is important in any industry that produces mechanical devices, such as the automotive industry.

A classic example of components that must be clean are ball bearings, ball bearings play a key role in systems that use a rotating shaft. Bearings are manufactured precisely and often must show only minor levels of variation between units. In many applications a size variation of even 5 μm can result in parts failure, and for this reason particle contamination can also be problematic.

In worst case scenarios particle contamination can result in damage or failure, but even when it is not this catastrophic, they can still result in inefficiencies by reducing torque, stopping the machine from operating smoothly, or they can impact machine lifetime. For this reason, it is important that high levels of cleanliness are upheld throughout the manufacturing process to provide a quality product.

Residual particles can either be introduced by the surrounding environment or from the assembly and manufacturing process itself. To determine if particle levels sit within allowed limits, and to identify where particles originate from, the particle shape and composition must be known.

Conventionally optical light microscopy has been used to count and measure particles; however, this is a lengthy process, restricts the volume of particles that can be counted and does not provide any information on the composition of the particles.

Using an SEM (scanning electron microscope) alongside EDS (Energy Dispersive X-ray Spectrometry) allows counting to take place on a finer scale and information on the nature of the particles can be collected. In addition, this process can be fully automated.

AZtecClean is a tool of the AZtecFeature which is designed to carry out technical cleanliness analysis to international standards that are in accordance with VDA19 and ISO 16232. This article explores an example of using AZtecClean to assess the cleanliness of ball bearing manufacturing to these standards.

Example: Technical Cleanliness in Deep Groove Ball Bearing Manufacture

Deep groove ball bearings are the most commonly used type of ball bearing. They are used in applications as varied as high-speed machine tools, automotive and aerospace, and in medical and dental equipment.

Sample Preparation

A sample was prepared by washing a known volume of finished product to remove any associated particles. These particles were collected using a membrane filter which was then mounted to a sample stub and coated with carbon to reduce the chance of the filter charging when under the SEM beam.

Detecting Particles and EDS Analysis

The SEM’s backscattered electron (BSE) detector was used to image the sample. As the density of the phase is used to generate contrast in BSE images it is an ideal method of locating particles. As the majority of the particles present contain dense elements they appear brighter than the material of the background filter (Figure 1).

Figure 1. BSE image showing particles on a filter.

A single grey threshold was used to separate the identified particles from the dark background, and this information was used to determine where EDS measurements should take place. The shape of all of the particles identified was automatically measured and this data was combined with chemical data from the following 20 kV EDS analysis. An example of a field of view where several particles have been detected is shown in Figure 2.

Figure 2. (A) Particles meeting detection criteria are detected and coloured. (B) Typical spectrum acquired from a silicate particle at 20 kV.

Automated Large Area Analysis

A whole filter sample was automatically analyzed over several fields of view, with all of the data combined into one set. This data was collected over a total circular area greater than 506 mm² (d = 12.7 mm) with a resolution of 1.15 µm. A total number of 31,249 particles were identified and analyzed in real time.

The entirety of a filter was analyzed automatically with the results from each field combined into a single data set. The data shown in this article was obtained over a total circle area in excess of 506 mm² (with a diameter of 12.7 mm) at a pixel resolution of 1.15 µm. 31249 particles were detected and analyzed in real time (Figure 3).

Figure 3. Montaged Image of the Entire Filter Showing Detected Particles.

Classification and Report

During collection of EDS data it was quantified and categorized in real time using a specialized technical cleanliness classification scheme (Figure 4).

Figure 4. Classification totals and colour key for entire filter analysis run.

Table 1. Subset of Technical Cleanliness Results for this sample reported to the size bins defined by ISO16232 and VDA 19.

Separation of Features that Cannot be Separated with BSE

In perfect situations when samples of this type are prepared, they would show even distribution over the sample surface with no particles in contact or overlapping one another. In these perfect cases the contrast and brightness can be set to ensure that all particles can be observed and analyzed separately.

In reality, most samples are not ideal and will have some particles touching or overlapping one another. In addition, different phases may have similar grey levels making it difficult to separate them on this basis.

Alternatively, it could be that when contrast and brightness are set for one group of particles, another group becomes saturated. An example of this is shown in Figure. 5a. In this example the BSE makes it appear as if one particle is present and it is impossible to determine if this is truly the case, or if it is an agglomeration of several particles.

Figure 5. A saturated particle – It appears from the BSE image (A) that one particle is present when in actuality, when it is mapped and processed with FeaturePhase (B) there are two.

AZtec FeaturePhase can provide an answer to this problem. The collection of EDS maps only from within highlighted grey level regions saves analysis time (as opposed to running EDS for the entire sample) and a phase identification algorithm is used to identify EDS characteristics.

By using FeaturePhase to analyze the particle in the center of Fig. 5a, it is possible to determine that two phases (steel & Silicate) within the particle, as shown in Fig. 5b. This information would have been lost without FeaturePhase.

Conclusion

AZtecClean is an extremely useful system for the rapid and accurate determination of how technically clean a sample is. The use of a large area EDS detector at a high throughput is possible to give the shape and chemical information for every particle present.

This means that comprehensive data on the sample is provided for the adherence to international technical cleanliness standards.

The use of a dedicated categorization method with pre-optimized settings keeps data in line with the required standards. In addition, the FeaturePhase tool in AZtecClean allows the separation of phases which would be difficult to distinguish using conventional technologies.

This information has been sourced, reviewed and adapted from materials provided by Oxford Instruments NanoAnalysis.

For more information on this source, please visit Oxford Instruments NanoAnalysis.