Engineers, by their very nature, like to design new things. To us, there is an undeniable appeal to creating something no one has ever seen before. However, there are compelling reasons to leverage your company’s past. It can be far more cost-effective and less risky to re-use past successful designs.
But this can be a daunting challenge. First, you need to have a database that contains the historical record of your company’s designs. Each design could be broken down into non-dimensional parameters, as this is far easier way to build a method by which dissimilar specific designs could be directly compared to analogous designs. Then, you need a smart search routine to sift through the mountain of data to find pertinent references. To top it all off you need intuitive design tools that enable you to visualize the changes to ensure the new design meets your current needs.
Imagine how much time and money could be saved if a successful historical design could be used over again for a current application. It might be rare, but it is possible. If there is nothing in the database that can be re-used directly, then perhaps scaling an existing design might satisfy the new design requirements. Still a huge timesaver. Currently, Concepts NREC’s Agile Engineering Design System® 3D blade geometry module, AxCent®, offers two tools to scale a design. Users can proportionately expand or shrink a compressor’s flow-channel to adjust for higher or lower flow requirements while not changing basic performance characteristics. Typically, this has been used to create “flow trims” to complete a compressor “family” – a group of scaled designs to map a huge range of flows. There is also a more flexible approach called intelligent scaling. Intelligent scaling enables an engineer to choose the specific dimensions which should be scaled or left unchanged. For a situation in which there are mold fill limitations (for cast impellers, for example), and too-thin blade tips are not viable, the engineer can lock this parameter and the scaling will bypass it.
If scaling doesn’t result in a feasible design, there is still more that can be done to leverage legacy designs. Database candidates which are “close, but not close enough” can be looked at in their preliminary design software and adjusted to match the new condition. In this scenario, even though everything is not being re-used, there is still significant time saving achieved. In a worst-case scenario, you might determine that the new condition is in a completely new sector of the design space, and that a clean-sheet design is inevitable.
This idea of a design database is one that Concepts NREC has already implemented with EASy! Control, which can search design databases and extract relevant meanline modeling parameters. Extending this process to recover portions of the 3D geometry will pay huge dividends to customers with a portfolio of designs. Concepts NREC is working with its customers towards extending the functionality of design databases to include many more parameters. Stay tuned!