Sorry, can't remember what the technical term is here but I remember from the dim & distant past that matching headers to the exhaust port precisely can lead to worse exhaust extraction from the cylinder - having a 'rim' (?) err, mismatch where the header is a larger diameter can reduce the exhaust sucked back into the next cycle. Which is a good thing. For one, more exhaust gases left over from the previous cycle increases the liklihood (how do you spell that???) of knocking.
Just thought I'd bring this up as a popular pastime at the moment seems to be porting headers for an easy (& cheap ) performance increase. And this might help?
Or am I barking up the normal-aspirated tree here? It was some time ago.

 

Anti-reversion headers they're called

Most 'proper' race car headers are stepped up in diameter some 6" from the head

 

Aha, I noticed this on F1 engine.

I assume enlarging the headers to bigger than port has a similar effect too.

I also remember reading in A Graham Bells "Four stroke engine tuning in theory and practice" (library copy, excellent, might buy a new updated copy) that he found a larger overlap on one side of a rectangular port helped. I think it was the bottom edge, closest to the smallest bend rad of header.

Paul

 

Ah-ha. OK. One sided.
So I presume that this is all a bit 'advanced' at makes such a small effect that, well, it's not really worth it? As soon as it has anything to do with F1 the advances seem to be, well, 'technical' as opposed to useful.
Or does it just mean it doesn't really matter if we get a bit carried away with the angle-grinder?

 

I think it means grind away, better to flow into a header larger than port, instead of smaller or the same size.

Grind away

Paul