August 2022 Volume 4

MATERIALS

“Simulation itself doesn’t improve die life, but it can tell you what to expect.” There is much that goes into analyzing the behavior of forging dies. “There is no routine that calculates die life in every situation, but in the last few years there has been an increasing interest in modeling die behavior,” says Nicolas Poulain at Transvalor Americas. The Transvalor platformhas been used for die modeling for decades, but early on the analyses were performed with a simpler surface mesh analysis on the die. This was because in most applications the forging dies did not deform in such a way that affected the part’s final shape. As die analysis has come under increasing scrutiny, a lot more input data is required to predict die behavior. Such simulations have to be calibrated by die material, forging process, forging size, etc. ArchardWear Model Wear is a critical part of die life, and more importantly one that engineering departments must monitor to anticipate maintenance. While cracks are the dramatic failure of the tools, making them unusable, wear might be less noticed but leads to out-of-tolerance parts. This is why it is essential to predict and quantify die wear. One common model used for that purpose is the Archard model. When calibrated properly this model estimates the amount of tool material removed after each forged part. Knowing this value is critical in scheduling maintenance to ensure the production of in tolerance parts. The model implemented in Transvalor’s FORGE platform is defined as follows: σ n .∆v δh= ∫ K w K f H � m (T) dt With: • K f as a coefficient related to lubrication • K w as a coefficient related to the die material • m as a wear coefficient from tribology testing • H v (T) as Vickers hardness function of the temperature The hardness (Vickers or others) must be temperature dependent to make sense in this model. Figure 1 shows an example of data entry.

For the die material, information such as hardness, yield and tensile strengths, and thermal properties must be input to the system. Process information relates primarily to the strain rate and the material of the part to be deformed. For example, a part formed by a hammer has a much higher strain rate than one formed by a mechanical press. Forging size relates to the volume and weight of the billet or preform used to form the part. In simulating a forging die’s behavior, the die(s) can be simulated separately from the workpiece in what is called an uncoupled analysis. In this most common method, the dies are considered rigid (do not deform) during the forging operation and the normal stresses are recorded on the surface of the dies. After the forging operation, a separate simulation can be run in which the dies are volume meshed. Stresses can be analyzed easily this way. It is also mandatory to run a coupled die stress analysis (see main article) in order to have a temperature distribution record within the tools. We recommend the use of a steady state temperature distribution in order to have a more accurate distribution while in full production. A steady state simulation can be run with the simulation considering die heating dies during the forging operation and the cooling phases during lubrication and wait time. We repeat this sequence until the temperature distribution within the dies has reached a steady state. Figure 2 shows the amount of material (in 100th of mm) removed in one forging blow as given by the ArchardWear model.

Figure 2. Quantitative wear results (100th of mm) given by Archard Wear Model. The Archard Wear model is a powerful way of predicting the wear evolution of a tool set. It enables the engineering department to schedule preventive maintenance and avoid production disruption due to out-of-tolerance parts. It increases die life through preventive care. – Courtesy of Transvalor Americas

Figure 1. Vickers Hardness temperature evolution

FIA MAGAZINE | AUGUST 2022 26