Gas Entrainment and Inclusions in the Die Casting Filling (Part Two)

How to avoid gas entrainment and inclusions？

We know that gas entrainment and oxidation inclusions can be formed in the die casting filling process, so how to avoid these defects through reasonable mold design and process parameter adjustment? As mentioned above, for the obvious gas entrainment caused by back pressure, usually we can rely on experience to avoid it through certain mold structure adjustments. For the oxides formed in the smelting process, we use vacuum smelting or protective gas smelting to keep the molten metal from oxygen. For other forms of defects, especially the gas and oxides involved in the molten metal in high-speed filling, we need to use CAE simulation methods to analyze and avoid them.
 

CAE simulation technology is based on computational fluid dynamics, using discrete grids to solve fluid dynamics control equations to obtain the flow pattern of the metal alloy filling process. By adjusting the three-dimensional geometry of the pouring gate and using numerical simulation technology, we can obtain the filling shape and movement characteristics of the fluid under different gating systems. Then, by analyzing these flow characteristics, we can select the best gating plan to avoid defects like gas entrainment.
 
There is no absolutely effective criterion for using CAE technology to judge the flow and filling quality. At this stage, not only academic research but also commercial software have put forward many kinds of judgment standards. However, no standard has been proven to be completely effective. Most of the standards only work for a certain type or specific range of castings. Generally speaking, we generally believe that the intersection of different liquid flows should be avoided in the filling process, and ensure that the front edge of the interface advances gradually so as to avoid gas entrainment or oxides. As a matter of fact, for most products, the die casting filling process is extremely turbulent, and the liquid front spatters in most cases. In this case, it is difficult for us to adjust the gating system to change this flow state. At the same time, complete turbulence and spattering will not necessarily have a catastrophic impact on the quality of the product. We have found in many actual die casting processes that greatly increasing the high speed and using the U-shaped runner for filling (that is, reducing the filling path and filling time) can even lead to castings with higher quality under certain circumstances. This situation is actually similar to the dispersion strengthening in the material field. Under extremely high speed conditions, although a lot of gas entrainment or inclusions are caused by the turbulence of the front interface. However, because of an increase in liquid momentum, the gas entrainment or inclusions will be broken to varying degrees, so that castings did not produce defects that could seriously affect the performance of the die casting at the end. These defects are dispersed in the microstructure of the die casting with a small size, which has a positive effect on improving the quality of die castings. However, we can also infer that this so-called dispersive broken defect is extremely difficult to quantitatively control in actual production. Therefore, mold design based on this concept is difficult to be absolutely effective to a large extent.

 
Figure 3 CAE simulation for predicting gas entrainment

It is unrealistic to use numerical simulation technology to obtain accurate calculation of gas entrainment, including gas entrainment volumes, gas entrainment positions, inclusion volumes, and inclusion positions. The main reason is that the gas entrainment and oxide film generally occur on the surface of the fluid, and precise positioning must be achieved by using a very fine grid for calculations on the liquid surface. Taking the oxide film as an example, the thickness of the general oxide film is only from nanometers to micrometers. If such a small grid is used in the simulation calculation, the grid volume of the entire fluid will be very large, and this numerical calculations with a large number of grids are inherently unrealistic. Therefore, most of the existing models are not direct simulations, but use indirect methods to evaluate defect locations and defects.  The existing gas entrainment models that have been applied to commercial software include: cumulative gas entrainment surface methods, eddy current judgment methods, cumulative variable methods, gas involvement methods, dimensionless number evaluation methods, bubble entrainment methods, multiphase flow methods, oxide layer entrainment methods (steel) and liquid surface folding methods. Each method has its own advantages and disadvantages, and you can refer to other articles for in-depth understanding if you are interested.
 
At the present stage, the gas entrainment model and oxide entrainment model used are separate. The core of the gas entrainment model algorithm is based on the isolation domain judgment. Each calculation will automatically determine the gas isolation domain of the entire area, and then calculate and calibrate the pressure and volume of each gas isolated area. These isolated areas, that is, air masses, will follow the fluid flow in the cavity, and the air masses themselves will decompose and fuse; determine the actual possibility and volume of gas entrainment based on the pressure and volume of each air mass. The oxide entrainment model is realized by particle tracking. In each calculation, we set the particle in the liquid surface position unit, and track the movement of the particle in the liquid in the subsequent flow process. Finally, determine whether the location will become a potential oxide inclusion according to the characteristics of the particles. This method can simultaneously consider the oxide film on the surface of the liquid in the pressure chamber and the oxide film involved in the liquid in the filling process, so the calculation accuracy is greatly improved.