When engine test cells first came to the scene, air flow prediction was a guessing game. Sometimes designers made good guesses; sometimes they made bad guesses. Then came Computational Fluid Dynamics (CFD), shedding some much needed light on the critical aero systems engineering analysis that always takes place when engineers design propeller and turbine engine test cells.
Background
During the early years of propeller and turbine engine test cell design, there was a great deal of iterative trial and error regarding test cell airflow dynamics. Engine manufacturers and dedicated suppliers in the business developed rule of thumb designs to ensure that flowfield-related things like Cell Depression, Cell Pumping, Entrained Airflow and Section Velocities all met average aero systems engineering guidelines for the aerothermodynamics of mixed gas flow in a controlled volume. Various approximations were used to predict the engine test cell flow performance for a particular engine. The problem with the trial and error method was that it cost both time and money, for the supplier and the client.
Test Cell Aerodynamic Theology
Test cells typically do not fit the standard design practice for clean airflow that you find in a standard wind tunnel system. Because of the need for very controlled laminar flow in a typical wind tunnel test section, the application of standard fluid flow aero systems engineering equations works well for wind tunnels; the designer keeps obstructions to an absolute minimum, ensuring that all transition zones reduce flow separation and turbulence as much as possible.
Test cells, however, are the antithesis of a wind tunnel. They tend to have anything but a clean entrance to the test section, so using standard fluid flow equations from aero systems engineering practices model cell airflow dynamics.
Due to this fact, some designers turned to physical scale models using modified Bunsen burners as engine simulators in an attempt to avoid building a twenty million dollar albatross with airflow problems. Still, these scale model systems were limited in successful airflow prediction even on their best day.
Hence, trial and error test cell designs led to that great engineering practice of rule of thumb. In other words, we found a solution that worked over here, let’s scale it up or down and repeat the process. However, as engine bypass ratios and mass airflows have continued to increase, and cost sensitivities on projects have become paramount, the trial and error rule of thumb method has presented obvious drawbacks.
CFD to the Rescue
Test cell design companies were instantly drawn to the science of Computational Fluid Dynamics for this very reason. CFD allows an engineer to construct a model of a flow field, such as a wind tunnel or an engine test cell and then apply boundary and state conditions to its flow media. Similar to a wind tunnel, and engineer can then visualize pressure, temperature and velocity profiles within the flow field boundaries. However in the early days of CFD, this high level of aero systems engineering analysis was reserved for massive computer systems at universities and research labs. Eventually, CFD made its way to computer systems that could be purchased by large corporations and finally to major players in the industry.
In the same way that computer aided design (CAD) has revolutionized drafting, CFD, coupled with computer aided engineering (CAE) tools with 3D modeling capabilities such as SolidWorks™, has revolutionized test cell aerothermodynamic performance prediction.
With the advent of PC Workstation CFD software bundled with industry-standard CAE tools, the ability of the test cell designer to visualize the potential issues in a test cell airflow management system has reached new plateaus. CFD has become an essential tool in the aero systems engineering toolbox for test cell design teams.
Summary
There is still a definite need for engineers to do their due diligence by the tried and true calculation method for gross airflow performance. However, the complex flow field of an engine test cell demands that the responsible engineer develop a full 3D model of the test cell complete with all obstructions in the flow field at all levels, to ensure the performance of the test cell to the best of their ability.
Of course, even the best of CFD models, should be examined against standard rule of thumb conventions to make sure the model falls in the range of expected performance. There is rarely a need to reinvent the wheel when it comes to aero systems engineering design methods.
There is no substitute for good engineering practice. All engineers must always remember that the computer software is just a tool, not the law, when it comes to designing a complex system such as a propeller or jet engine test cell. The basic tenets of aero systems engineering, however, always apply.
