SpinOffs

PCs and workstations are getting faster every year, following Moore's Law. With all this power and speed, the “virtual prototyping” of turbomachinery, using CFD simulations and optimization algorithms is getting more and more common. In this context, one may question if meanline design still makes sense. Let's explore the implications of this question. To do this, we have to go back to basics.

In my last blog, Specific Speed Demystified, I covered the mathematical definition of specific speed and how it relates to flow and work coefficient. The concept of specific speed has been a guiding principle for radial turbomachinery design for many years. Use of specific speed has been most heavily emphasized by Balje in his famous textbook and early publications.  In these works, he laid out several graphs that are still widely used today. 

In my blog Flow Coefficient and Work Coefficient, I outlined the basic concept behind the flow and work coefficient. These nondimensional parameters are widely used to characterize axial and radial turbomachinery. Another widely used parameter for radial design is “specific speed”. For something with such a finite name, specific speed is perhaps the most mysterious and non-intuitive parameter in all of turbomachinery. In this blog, I'll lay the ground work for understanding specific speed.

 When designing compressors, engineers often use ruled-surface blades with the goal of making a shape that’s easily manufactured on a 5-axis machine.  Theses blades can be quickly machined in one pass by aligning the side of a cutting tool to the rulings. This process is often referred to as “flank milling.”  The alternative is to make many passes with the tool tip, a process known as “point milling”. For the right application, flank milling is often favored for shorter cutting times and better surface quality, but there are some caveats.