We will always need real models and test to get an idea of the real thing. We do not have equations to explain everything in any possible situations, so no, even CFDs, FEA, etc contain their weaknesses and inabilities. I doubt a day will come where we would know enough, .i.e. all - to simulate alone and get results.
Tomba wrote:
Americans... common rail direct injection and turbocharging would be better for the environment.
If you considered manufacturing emissions, I bet it wouldn't. I know its hard to think of C02 coming from anything but a Ford F-150's tailpipe, but over half of a car's total carbon output comes from its manufacture.
Well, Tomba, thank you very much. When I made the comment on simulation becoming more precise than reality, I was thinking, tongue-in-cheek, that this is what happens in traffic simulation: when you simulate on computers you have to understand what you're simulating. On the other hand, measurements (empirical methods) only give results but don't tell you why.
For example, the empirical methods for traffic or pavement design give you very accurate dimensions to design, based on experiments (like when you use a wind tunnel). However, when you simulate, you have to understand what's happening.
The circle here is that for the simulations to work they have to be compared with something. Unless you can measure "something", then you have nothing to revise if your simulations are correct.
A question for the CFD specialists: I have seen a paint that is sensitive to pressure. Why don't they use it to test the cars on track, instead of using it on a wind tunnel? This way you could check the simulations "directly". Traffic engineers don't have that problem: we can count or measure contamination, noise or speed with ease and cheaply. Pavement is different: you need to put instruments into it to validate the models.
So, why don't dispose of wind tunnels entirely and do "real" aerodynamic testing using some kind of instrumentation to check the equations? It is just a matter of money? Or is it lack of instrumentation that can be carried on a real car? Maybe the answer to this question is what will tell us if wind tunnels will carry on forever, as a needed counterpart to computer modeling...
Ciro Pabón wrote:
A question for the CFD specialists: I have seen a paint that is sensitive to pressure. Why don't they use it to test the cars on track, instead of using it on a wind tunnel? This way you could check the simulations "directly". Traffic engineers don't have that problem: we can count or measure contamination, noise or speed with ease and cheaply. Pavement is different: you need to put instruments into it to validate the models.
I have heard that the paint is very messy and it needs to be used under a UV light. It also clogs pressure taps that are used on the models in the wind tunnel.
One 'old way' of seeing what the boundary layer is doing is by using tufts of wool string taped in a grid pattern across the vehicle or on an area of interest. I'm going to do this to my car one day, although it does seem a bit of a pain.
