When to switch to 3D CFD
 

Since Oct., I have been doing 1D CFD on an automobile engine and have made a lot of progress.

However, I am finding the 1D CFD is leading me in direction opposite that taken by the std. textbooks on airflow through engines on a couple of engine airflow topics. For example, I get better response (TQ and HP) when making a helmholtz resonator just big enough to completely cover the velocity stack trumpet (about 1 litre in the HR), and that my exhaust header pipes (primary, secondary, and expansion chamber) are longer than what the std texts describe.

With respect to the helmholtz resonator in particular, the length of the HR neck is very short compared to the width of the neck (L on the order of 10mm w; on the order of 60mm). Is this one of the kinds of simulations that needs 3D aspects accurately modeled because of the 'end effects' of resonating masses in short pipes? In particular, the HR resonator is barely bigger (2X in volume) than the trumpet on the velocity stack that reside inside of it. So, while I modeled the overall shape of a HR resonator with nice gentle curves, I'm not sure that the 1D models what might really happen.

So the question is:: At what point does it become profitable to switch from the simple 1D world to the world of 3D CFD?

Where profit is defined as: making forward progress faster

Thanks

Mitch