Using dilatometer experiments, numerical data can be collected during the sintering process and used to develop a constitutive model for shape changes during sintering. The dilatometer chamber is closed and small compared to a normal industry oven used for sintering. Depending on the final application of the sintered product, different temperatures and heat cycles may be used during the process. It is therefore important for a constitutive model to be robust, giving valid results even when changes are made to the sintering procedure. This is done by optimizing the constitutive parameters with respect to different sintering cycles. Variations of initial density after compaction has also been considered in the constitutive model. The deviatoric part of the model is investigated for experimental results in nonhomogeneous sintered specimens to complete the constitutive model. The model presented is for the full sintering process and relevant physical aspects have been accounted for.

By performing a dilatometer experiment, measured shrinkage can be used to determine the adjustable parameters in a constitutive model for sintering. While the dilatometer machine is excellent at collecting data, its conditions differ from those in an industrial sintering oven. Therefore, the constitutive model must be robust and in the present study, different sintering time cycles have been used to optimize the adjustable parameters in the model. Investigations on initial density has also been performed to better understand the sintering process. Before sintering of particles begins, during the debinding process, densification can be detected, which is dependent on the initial density. The constitutive model for sintering was improved to include this phenomenon by adding particle rearrangement based on the theoretical packing of spheres.

During sintering, a green body of powder particles is heated to high temperatures, fusing the particles together. In cemented carbide production, the sintering process generally results in substantial densification of the material. By using a dilatometer, shrinkage during the sintering process can be measured. For a green body of lower density, early particle rearrangement has been observed. This is investigated here using different initial densities using the same powder, leading to a suggested addition to the constitutive model. The environment in the dilatometer and the sintering furnace differs, especially with respect to heating and temperature during holding. This effect can be minimized by creating robustness in the model, making it independent of the heating cycle. Here, this is done by optimizing the constitutive parameters towards four heating cycles for a specific powder.

Finite element (FE) simulations are frequently used nowadays in order to analyse powder compaction and sintering, for example, when determining the shape of a cutting blank insert. Such analyses also make it possible to determine in detail the stress state in a powder compact during loading and unloading. This is certainly important as (residual) tensile stresses can lead to cracking, either after unloading or during the subsequent sintering step. The magnitude of plastic deformation is also an issue here. Concerning the stress state in the powder compact, the frictional behaviour (between walls and powder compact) is of great importance. For this reason, in the present study, two frictional models are implemented into a commercial FE software, and numerical results based on the stress state before and during unloading are derived. The two friction models produce quite different results, and it is obvious that the frictional behaviour at powder pressing has to be carefully accounted for in order to achieve results of high accuracy.

The densification of cemented carbides during sintering was studied using an existing constitutive model based on powder particle size and material composition. In the present analysis, we study how well the constitutive model can capture the experimental results of a dilatometer test. Three experiments were performed, where the only difference was the transition between the debinding and sintering process. From magnetic measurements, it is concluded that the carbon level in the specimen is affected by changes to the experimental setup. It is shown, using parameter adjustments, that the constitutive model is more suited for a certain experimental setup and carbon level, which is a limitation of the model. In order to capture the mechanical behaviour under different experimental conditions, further constitutive modelling relevant to the carbon level is recommended.

During unloading, pressed hard metal powder compacts expand (spring back), leading to unwarranted tensile stresses and, subsequently, crack initiation in the green body. Here, the elastic spring back and the green strength are analyzed for different types and amounts of the pressing agent PEG (polyethylene glycol). The results show that the plastic behavior, but not the elastic behavior, is influenced by a change of the pressing agent PEG. In this context, it should be stated that the risk for the initiation of cracks is influenced by both the elastic and plastic behavior after compaction. The spring back after compaction and the green strength defines to a large extent the risk for cracks. In addition, it is concluded that a standard three-point bending test is sufficient to analyze the risk for the initiation of cracks when comparing different spray-dried powders.

The correlation between granule strength and green strength of hard metal powders is examined. The approach is based on experiments and numerics. In the latter case, a Design of Experiment software is used. The granule strength of the powder (particle) is determined by GFP-measurements (“Granularfestigkeits-Prüfsystem”). During this test, a single particle is pressed from one side until breakage. The corresponding measurements of the green strength are done using three-point bend (3PB) testing. The experimental results show that the pressing agent has a strong influence on the behavior of both quantities. The statistical evaluation shows that the relation between the two strength properties is very close to linear with coefficient of determination R2 taking on the value 0.97. This of course indicates that it is possible to get information about one of the properties for a similar set of materials by experimentally determining the other one. This is of substantial practical importance as for one thing it can limit the amount of testing required. Even though the present investigation is pertinent to hard metal powders, the results could be of value for many other types of powder materials. 

It is known that the shape of a cutting insert blank, after pressing and sintering, can be predicted with finite element (FE) simulations. It is also known that such simulations have the potential to save costs and time when used for press tool compensation. However, in such simulations (and in a real situation), the frictional behaviour has a great influence on the results. Therefore, in this study, two frictional models are discussed and implemented into a commercial FE-program. The results from the different frictional descriptions, when for instance analysing density after compaction, shows a clear difference. It can be concluded that the frictional behaviour at powder pressing has to be modelled in detail, also at low forces.

Material parameter curves in an advanced material model describing compaction of spraydried cemented carbide powder are determined successfully based on a general approach formaterial characterization of powder materials. Pressing forces from a production machineand equivalent finite element (FE) calculations are used in inverse modelling. A pressingmethod that includes multiple unloading steps is used. The material model is of DruckerPrager CAP kind and friction between powder and pressing tool is modelled as a function ofnormal pressure. The results are verified with density gradient measurements using aneutron source. The method is proven to be robust and the results show good agreementbetween experiment and simulation. Effects that have not been captured numericallypreviously are captured due to the high accuracy of material characterization. The presentapproach is detailed for tungsten carbide powders but is valid for other powder materialswhen properly calibrated for constitutive and frictional effects in the same manner asoutlined here.