>I would think the ideal diameter is a function of actual CFM >througput.
>Or perhaps the target is an ideal velocity that is a function >of CFM and diameter.

Here's something I've always wondered about. There's a variable that complicates the calculation, and that's the decrease in exhaust gas temperature. As the exhaust gas cools, it reduces in volume, and as we all know, temperature and volume are proportional. The cooler the gas the less the volume. Using that reasoning, as the temperature drops, so could the cross sectional areas of the pipe. Does that mean the pipe diameter can get smaller downstream and still handle the volume? What about density? As the volume drops, density goes up, doesn't it? So how does that figure into the equation?

Bob