Wind Tunnel v CFD - Cost and Speed.

Technical considerations aside, the choice between either a CFD or model scale wind tunnel development program will come down to cost and expediency.  There isn't a single right or wrong choice.  The best choice will depend on the complexity and duration of the prospective program and on the resources available to it.  To illustrate the choice, we'll consider a variety of development programs.

Option 1) Tiny program
We'll run a CFD program with 100 full car runs and the equivalent model wind tunnel testing program (wind tunnel @12 hours for one day only with approx 5 useful runs with 20 ride points ) and we will compare the two.
  • From inception to delivery of data would take approximately 3 months for the wind tunnel program. The CFD solutions should be available within 1 month. CFD is much quicker.
  • The cost for the CFD program would be approximately £20,000.  The cost of the wind tunnel program would be approximately £80,000.  Both estimates exclude the salary of an aerodynamicist (because it would vary enormously depending on experience), but do include all other costs (tunnel hire, model technicians, model manufacture, CFD solutions and analysis).  CFD is much cheaper.

Option 2) Small program.
We'll run a CFD program equivalent to only 5 days of model wind tunnel testing (wind tunnel @12 hours per day, approx 30 runs with 20 ride points and CFD @ 10 complete car solutions per day) and we will compare the two.
  • From inception to delivery of data would take approximately 3 months for the wind tunnel program. The CFD solutions should be available initially within 1 month, but would take 10 months to complete.  CFD is initially much quicker, but the wind tunnel program would be quicker in the longer term.
  • The cost for the CFD program would be approximately £600,000.  The cost of the wind tunnel program would be approximately £120,000.  Both estimates exclude the salary of an aerodynamicist (because it would vary enormously depending on experience), but do include all other costs (tunnel hire, model technicians, model manufacture, CFD solutions and analysis).  The wind tunnel program is much cheaper.
  • It should be noted that whilst it is considerably more expensive to generate the same data using the CFD solution, the CFD solution will probably generate better understanding of the overall vehicle flow.
Option 3) Medium program.
We'll run a CFD program equivalent to 5 days of model wind tunnel testing per month for 1 year (12 hours per day, approx 30 runs with 20 ride points) and we will compare the two.
  • From inception to delivery of data would take approximately 3 months for the wind tunnel program. The CFD solutions should be available initially within 1 month, but after that, the CFD would be unable to keep up with the wind tunnel program and would lag further and further behind.  CFD is initially quicker, but the wind tunnel program would be quicker in the longer term.
  • The cost for the CFD program would be approximately £7,200,000.  The cost of the wind tunnel program would be approximately £500,000.  Both estimates exclude the salary of an aerodynamicist (because it would vary enormously depending on experience), but do include all other costs (tunnel hire, model technicians, model manufacture, CFD solutions and analysis).  The wind tunnel program is much cheaper.
  • It should be noted that current CFD capabilities simply cannot cope with a program of this size, so it is perhaps a little unfair to compare the two.
Option 4) Big program (as might be run by a small, back of the grid F1 team).
We'll run a CFD program equivalent to 5 days of model wind tunnel testing per week for 1 year (12 hours per day, approx 30 runs with 20 ride points) and we will compare the two.
  • From inception to delivery of data would take approximately 3 months for the wind tunnel program. The CFD solutions should be available initially within 1 month, but after that, the CFD would be unable to keep up with the wind tunnel program and would lag further and further behind.  CFD is initially quicker, but the wind tunnel program would be quicker in the longer term.
  • The cost for the CFD program would be approximately £36,000,000. The cost of the wind tunnel program would be approximately £1,650,000. Both estimates exclude the salary of an aerodynamicist (because it would vary enormously depending on experience), but do include all other costs (tunnel hire, model technicians, model manufacture, CFD solutions and analysis). The wind tunnel program is much cheaper.
  • It should be noted that current CFD capabilities simply cannot cope with a program of this size, so it is unfair to compare the two.

Click here to return to Aero Development Techniques.
Comments