Why controlling the deep rolling process is essential for component reliability

Technical systems, such as machines or vehicles, are becoming increasingly sustainable. Their resource efficiency is rising, and they are becoming lighter and lighter. As a result, designers are increasingly reaching the limits of what is feasible in terms of further increasing power density. 

One possibility, however, is to adjust and optimize surface and subsurface area properties even more precisely. In this material region, the component is subjected to the highest mechanical stress, which causes cracks to form. Compressive residual stresses, which are specifically introduced into the subsurface area, have been shown to inhibit or even stop the formation of cracks [1]. 

As a designer and manufacturer—particularly of safety-critical components—you face the challenge that the introduced Compressive residual stresses cannot be verified retrospectively through simple quality control. Measuring these stresses, especially deep-seated residual stresses beneath the surface, typically requires a destructive testing method. So how can the introduction of residual stresses be reliably ensured and documented? 

This is where ECOROLL’s tools come into play. With dynamic rolling force control, they reduce fluctuations in Residual stresses by up to 50%—and thus the risk of component failure.

Well-known methods for inducing residual stresses include Deep Rolling and shot peening. In shot peening, quality control is performed using so-called Almen strips. A defined metal strip (material, dimensions, etc.) is placed in the shot peening machine in addition to the component. This strip is shot peened in the same manner as the actual component. The deformation of the Almen strip is then measured. The thin sheet metal deforms due to the change in its residual stress state. The Almen intensity can thus be determined based on the degree of deformation. However, this does not guarantee that the necessary residual stresses have actually been introduced into the component. 

Deep rolling, on the other hand, is significantly more reproducible and, above all, more precisely definable. The deep rolling process can be very well documented using the three main parameters: roll geometry, feed rate, and, most importantly, rolling force. With the ECOsense technology introduced a few years ago, the rolling force can now be measured for each individual process and also documented. With the analog dial gauges previously used on the tools, however, there was a certain degree of leeway in the machine operator’s reading of the gauge deflection. 

Studies conducted at the Institute for Manufacturing Technology and Machine Tools at Leibniz University Hannover [2] have shown, however, that—provided the process was carried out in the same manner—the variation in the residual stress state falls within the measurement uncertainty of the residual stress measurement (Figure 1). 

According to the process signature approach developed by Prof. Brinksmeier (Leibniz-IWT Bremen), however, it is not the external load but rather the internal stress that is responsible for the development of the edge zone properties [3]. For Deep Rolling, this means that the strains and stresses in the contact zone provide more insight into the residual stresses than the externally applied rolling forces. 

Hertzian contact pressure is used as an auxiliary parameter for analysis. If the rolling force and rolling body geometry are the same but the component geometries differ, then the contact pressures also differ, leading to an altered residual stress state (Figure 2). 

If the correct introduction of residual stresses is truly critical for a component, then not only the rolling force but, more importantly, the contact pressure between the rolling body and the component must be taken into account. 

As part of a research project in collaboration with the Institute for Manufacturing Technology and Machine Tools (IFW) at Leibniz University Hannover, ECOROLL AG has developed a dynamic rolling force control system that helps maintain a constant contact pressure across the entire component geometry. 

A fixed rolling tool equipped with ECOsense technology is used for this control system. ECOsense measures the tool’s rolling force during operation and transmits the force signal via Bluetooth to a gateway. This gateway has now been equipped with a control algorithm that compares the actual force value with the target force value. If these values do not match, the tool feed can be corrected via the open OPC/UA machine interface. The corrected feed values are also transmitted to the gateway. 

This enables the gateway not only to ensure the target force value during operation but also to determine the part geometry. 

To determine the contact pressure from the rolling force, it is necessary to create a complete force history for the process. This means that for each individual contact point on the geometry, the contact pressure—and, consequently, the required rolling force—must be determined. The force control system is now capable not only of maintaining a constant rolling force throughout the entire process, but also of reliably achieving the rolling force with spatial resolution. 

To quantify the effect and benefits of this control system, a comparison with measured residual stresses was also conducted as part of the aforementioned “KontRoll” project. For this purpose, a component (Figure 3) was used that features various geometric elements (diameters, radii, straight lines). For deep rolling, a tool with a rolling roller (radius R = 2.5 mm) set at a 45° angle was used. 

The rolling process was performed on two identical components, each of which was designed differently. The developed control system was active throughout the entire process. In the first variant, the rolling force was maintained at a constant level. The entire component was always rolled with a rolling force F_w= 700 N. For the second component, the Hertzian pressure was kept constant at each contact point during the design phase. To achieve this, the rolling force was adjusted locally as needed. Figure 3 shows the resulting rolling force profile. 

Subsequently, the surface residual stresses for both components were measured at six different but characteristic points. The measurements were performed using state-of-the-art X-ray diffraction techniques. 

The result is clear. During deep rolling with a constant force, the residual stress values at the individual measurement points fluctuate around a mean value of s = -613 MPa with a standard deviation of 160 MPa. During deep rolling with a constant Hertzian pressure, the standard deviation is reduced by approximately 50%. This means that the machining result is already significantly better, and that dynamic process control of the rolling force is a fundamental component in the manufacture of lightweight components. 

The attentive reader has surely noticed that calculating the solid rolling force using Hertzian pressure represents a significant simplification of the process. Hertz’s theory is actually only valid in the case of an ideal elastic surface. Furthermore, contact pressure is not the sole factor responsible for the development of Residual stresses . Otherwise, the result in the experiment shown would not be a 50% reduction in the standard deviation, but rather nearly 100%. 

In the future, therefore, more research is needed to determine the causal relationships between process control variables and the process signature. Once these are known, the process can be further optimized through control. 

Would you like to learn more about ECOsense technology? Contact our experts or learn more about the technology here: 

Find out more about ECOsense

Acknowledgements

The collaborative project (KK5032727PK3) ‘Contact-tension-controlled mechanical solid-state rolling’ is funded by the Federal Ministry for Economic Affairs and Energy (BMWE) as part of the Central Innovation Programme for SMEs (ZIM) and is managed by the Association of Industrial Research Organisations (AiF). ECOROLL AG and the IFW would like to express their gratitude for the financial support provided for this project.

Sources: