Feasibility Report

With conventional capsules, the machine supports the capsule in such a way that the capsule itself does not need to carry any significant load caused by the pressure in the capsule. However, when combining capsules into a single plastic part, the wall separating the different compartments of the capsules is not supported by the machine, and thus needs to be strong enough to withstand the pressure in the machine.

Though it would be possible to just make the wall separating the compartments thicker, that goes against the primary purpose of having combined capsules, which is to reduce the amount of plastic wasted per coffee serving. Thus ensure that the final design uses less material than existing solutions, its shape needs to be optimized.

(Figure 1) shows a cross section of the capsule as viewed from above. The walls separating the compartments have a variable wall thickness. There are two reasons for this. The first reason is that by making the middle of the wall thin enough, we can expect the wall to bulge out. This is important as a curved wall can resist pressure without having to be thick enough to resist the bending. To be able to resist bending moments the wall thickness has to be relatively thick.

The top of the capsules is covered and sealed by a single piece of aluminum foil. Because there could be significant time between the different compartments being used, it is important that using one capsule will not destroy the other ones. Because the separating wall is designed to have significant deformation, the part of the capsule where the seal is connected to the capsule is going to move. Aluminum, however, is rather stiff, and cannot compensate for these deformations.

For this reason the top of the capsule is designed like a beam, being able to withstand bending forces. (Figure 2) shows a cross section from the side. This also results in the tops being wide enough to be able to glue the aluminum seal to them. A small chamfer has been added to prevent high stresses in the corners.

Finite Element Analysis was used to analyze how the capsules behave at different temperatures and in different conditions. (Figure 3, left) shows an assembly containing the capsules and a part that houses the capsules. The bottom of the housing is fixed. This is fine from a simulation point of view, because any horizontal forces are going to cancel out. Furthermore, this setup is primarily designed to analyze the separating wall. The maximum pressure inside the capsule is 9 bars. A safety factor of 1.2 is assumed. Thus, a 1.08 MPa pressure load is added to every surface on the inside of one of the capsules.

(Figure 3, right) shows the mesh quality. The mesh is somewhat rough in the curved parts of the capsule, but since we are primarily analyzing the separating wall, this is not an issue

The capsules are made out of a heat resistant version of PMMA. Since almost boiling water is flowing through the part, a service temperature of 100°C is assumed. This higher temperature lowers the strength and E-modulus of the material. A lower strength of 64 MPa and E-modulus of 2.77 GPa is assumed because of the influence of temperature.

Let’s first look at the top part of the wall. (Figure 4, left) shows the top view, which can be seen to bend slightly outwards. The exact max deformation at the surface is 0.46 mm. This is acceptable since the aluminum seal will be slightly loose. In (Figure 5, left) it can be seen that the top part does indeed function as a beam, since there are high stresses on the sides, while the middle experiences a lower amount of stress, which is characteristic of stress in beams.

Having a look at the deformation chart (Figure 4, center), it can be seen that there is a high amount of deformation in the middle of the wall. This is expected since we designed it to bulge out. This also serves as proof that it does indeed bulge outwards creating a curved surface.

(Figure 5, left) shows the stress in the wall. It can be seen that there are no high stresses in the middle part of the wall. 

The outside, however, does have some high stresses. This was theorized to be the result of the middle part not having to resist any bending moments. The sides, which do need to resist bending moments, carry some high stresses, but this is to be expected. Also notice the low stresses in (Figure 5, right). This is due to the walls experiencing a combination of bending stresses and shear stresses, which cancel each other out on the outside of the wall.