Detailed information about the density within a part produced by compression of a granular material is perhaps the most important piece of information you need if you are to predict its properties. Nearly all mechanical properties are strongly dependent on the density and the density of a very small region within the part can potentially be disastrous if it has a density level considerably lower then the rest of the part. Filling induced density variations is a problem that long has been overlooked by powder metallurgy scientists. One of the factors that are making investigations difficult is the lack of methods to make experimental data available. In this work four different methods for this purpose have been evaluated.
Has proven to be a reliable method to provide information about the density variations in cut, 2 dimensional cross sections of PM samples. To produce the amount of data necessary to study the filling operation is however considered too labour intensive.
Bulk densitv measurements of compressed samples
A simple and reliable method, perfect for thorough investigations of densityvariations in industrially produced PM components.
Density measurements of filled powder
Although not as accurate as bulk density measurements the increased flexibility of this method can be an advantage when investigating filling behaviour. A novel test rig was constructed that was able to identify and link several different factors to the resulting density variations.
This method has the potential to deliver the same information as metallographic examinations in a much faster and more reliable way. It has however not yet been established if the accuracy ofthe method is sufficient to provide information about the density variation in PM samples caused by the filling operation.
With the bulk density and the filled powder methods deeper studies were also performed to increase the knowledge of powder filling behaviour. It has been established that the filling operation causes density variations within PM parts and that these variations are highly systematic. With the construction of the test rig it was possible to determine several aspects of the feed shoe cycle as responsible for these variations.
Linköping: Linköpings universitet , 2001. , 18 p.