A small summary
There is no worse justification of an engineer for a design solution than: „We have always done so“. For this reason, it is one of the most important tasks of every engineer to continuously question and reinvent himself and his products.
The design freedom enabled by industrial 3D printing should be an incentive to rethink existing solutions creatively. Even if additive manufacturing is economically and technically reasonable only for selected areas, it nevertheless shows what potential exists in existing designs.
We demonstrated the revealing of unused potential in a hydraulic manifold using ANSYS Discovery Live. In this way we could not only significantly improve the flow behavior of the medium but also significantly reduce the weight of the manifold.
Problem and motivation
SMILE-FEM brings together a team of simulation experts who are passionate about simulating designs under real-world conditions that currently exist only as data. For over 10 years we have been supporting our customers and partners in recognizing and using the full potential of their technical application. Our focus is on static simulations of strength and fatigue strength, vibration simulations and dynamic simulations. For this purpose, different simulation tools from ANSYS have been used ever since.
It is part of our daily business to openly examine and test the designs of customers and partners. There are not only critical positions with too high capacity utilization that attract attention, but also areas that offer potential for optimization. New manufacturing processes such as 3D printing allow the user to partially overcome existing production limitations and to exploit unknown potentials. Using this prototype, we demonstrate the potential in a structure that has already been optimized using conventional methods.
We decided together with our partners to examine a hydraulic manifold (see Fig. 1) for unused potential. These components have the advantage to be installed multiply in a hydraulic assembly while still been designed individually for each use case. This means they are ideal for individualisation and 3D printing.
The technical challenge can be divided into two areas. On the one hand, there is the challenge of optimizing the function of the component. The second is saving weight and thus material. Both effects always play together. The focus may be different for each application, but it always takes both considerations to create a really good and successful product.
The whole procedure is based on a single assumption: the limits of the previous production method are being shifted. Currently, the channels are being drilled or milled into a massive block (see Fig. 2). So only straight connections with a constant or tapered cross-section are possible. Additive manufacturing now also allows for round connections, branches and undercuts. All the potentials shown here result from this one assumption.
Optimising the function
The function of a hydraulic manifold is to create connections between connectors and direct the media. In addition, functional units such as pressure reducers, check valves and controllers can be integrated. Our focus was on one hand to reduce the pressure loss in the channels to allow the use of smaller pumps and to achieve a better energy efficiency. In addition, a structurally desired pressure reduction can be realized more accurately. Also turbulences should be reduced. This leads to a reduction of the noise emission and a shortening of the switching times as side effects.
Functional optimization also includes the integration of multiple components. In this case, two connected units could be combined into one component. This not only saves assembly steps but also reduces sealing surfaces and thus potential sources of error.
Optimising the mass
Lightweight design is one of the driving factors in additive manufacturing and component optimization. Lightweight design is a driving factor, especially in aerospace, the automotive industry but also in special fields of the maritime industry. This can reduce fuel consumption and increase payload. But also the assembly is simplified because potentially fewer people are needed or tools can be saved. An important factor is also the saving of energy in the production and the better use of the existing raw materials.
In addition, there was a desire to save space in this application. This allows the limited space in the overall assembly to be used more efficiently and the entire product to be smaller.
The main motivation from a financial point of view is to handle small batches more efficiently if the production is designed for larger batches. This way it comes to a better utilization of the machines. Also, the storage is simplified because existing components can be made on Demand.
An additional financial advantage is given if the following assembly steps can be simplified or reduced. This is the case when several components are combined in one unit.
Methodology and software tools
The basis for the optimization was the CAD geometry of the previous application. Based on the function and the previous arrangement, the channels were removed and the connections repositioned. The aim was to keep the connections as efficient and short as possible. In addition, all dead ends were removed. Then the channels were redrawn in SpaceClaim (see Fig. 3). The direct modeling in SpaceClaim was immensely helpful. Subsequently, the channels were examined directly for their flow behavior in Discovery Live (see Fig. 4). Corners could be directly rounded off and gradients adjusted. For a comparison of the original and the optimized channels see Fig. 5. The direct presentation of the results after minor changes greatly simplified the optimization.
For the flow simulation of individual channels, a small graphics card was enough. To look at the overall complexity of the component and visualize the interaction of each channel, an NVIDIA Quadro RTX 8000 with 56 GB of memory was made available. Only then could the entire complexity of the channels be investigated.
The channels were imported into ANSYS CFX to resolve fluid flow details. Thus, local turbulence and detachment could be identified. For the CFX analysis, a hydraulic oil HLP 46 with a Reynolds number of 391 was used. The boundary layer was dissolved by a prism layer. Within this flow simulation showed the potential of Discovery Live. Although the numerical values are not as precise as those of a CFD simulation, the time it takes to create the model and calculate it is significantly shorter and makes variational calculations more efficient.
After the channels were finished, a topology optimization of the structure was performed. Not only could Discovery Live make various variations quickly and efficiently, changes in the design had a direct impact on the look of the finished structure.
The optimized geometry has been imported back into SpaceClaim. Here the facets were reduced from 1.3 million to 400,000. The structure was smoothed and functional surfaces such as flanges and holes were traced. Then the structure could be prepared for 3D printing. This first happened on our own filament printer before it could be realized in full-sized by affiliate companies. The topologically optimized structure is shown in Fig. 6 and 7.
Through the use of Discovery Live and a structured approach, we managed to pull the channels in a way that the height of the manifold was reduced by 28% see Fig. 8. This allows a space-saving installation as well as an enlargement of the usable space within the overall assembly.
Based on an exemplary selected channel (Fig. 2-4) we were able to reduce the pressure loss of the channels by 55%. This not only allows smaller pumps to be used, but also more accurate adjustment of the desired reduction in hydraulic pressure. By using curves instead of sharp corners, it was possible to locally reduce the pressure loss of only this component by 45%, as a detailed CFD calculation shows (see Fig. 9 and 10). It also shows how a turbulence forms in the depressions of the boreholes. In addition, a large flow separation is created after the sharp corner. This detail illustrates the enormous potential for optimizing the flow pattern. In addition to the reduction of the pressure loss, the shortening of the switching times is an advantage of the optimized geometry. Thus, a stable state within the flow sets faster than with the original model. A reduction of the acoustic emission by the reduction of turbulence is also an advantage.
A geometry was designed by the use of topology optimization and a boundary condition of a minimum necessary wall thickness. This assured the compliance with the permissible strength under working pressure. Overall, the weight of the manifold was reduced by 76.8%.
The application example illustrated here shows that the tools used here and their coupling worked very well. Using Discovery Live, flow simulations could already be performed during the creation of the new channels and then validated in details by CFX. Through this detailed investigation, we also gain significantly deeper insights into the flow behavior.
Once the channels had been created, the Topology Optimized Structure could be determined using Discovery Live. This showed the advantage of the efficient comparison of different variants.
Using SpaceClaim, the model could be processed and a structured strength analysis could be carried out in the ANSYS Workbench.
The immense potential for optimization is shown by 76.6% weight reduction, 28% reduction in overall height, integration of several components and a reduction of 55% in the pressure loss. This potential was found in a component that has undergone a huge process of conventional optimization in recent years and decades. This result could be achieved only through the disruptive technology of additive manufacturing as well as through modern simulation tools and not to mention good engineering performance. The project illustrates how much potential still be slumbers and can be found in conventional products and components.
This should be an incentive to question the own production methods as well as the design of the own structures and to become aware of the slumbering potential. Additive manufacturing is not worthwhile for every application. But to question the previous procedure is always worthwhile.