To start with, just like headers there are good and bad *cough* stock *cough* logs. "Bad" not only because of the undersized cross sectional area, but the lack of steps, reverberation canceling, etc.
Now the header vs log debate. There have been a million studies done on this and results vary. As I've seen the results can easily be swayed in whatever direction the tester wants them to. You can match a great header with a horrible (cast or fabricated) log and vise-versa.
In my experience a well designed, equal length true header/merge collector setup will make more peak power. But it comes at a cost. A log will almost always carry more power under the curve and as a result is much easier to spool a big charger. The logs are great at retaining the thermal energy and produce a lot of velocity for the given size. If you want to make the absolute most power possible in a small rpm range then a tuned header works well. But in my opinion and experience a well designed log will give up a very very small bit of power up top (>5% in most cases) and give back much more everywhere else.
Thank You! Couldn't have said it better myself.
Last edited: