Porting the manifold is just to take a drill or Dremel with appropriate grinder bits and grind flash off the openings on the intake manifold. Using a manifold gasket as a guide, use a sharpie to mark the material to be removed. Use the grinder to grind off the marked material. Some people use a ball hone to smooth the inside of the manifold - I did not bother. It *would* help to do so, due to laminar flow at the surface. Less roughness -> less surface area -> marginally less resistance to flow.
The reason to do it is that the exhaust in a manifold does not flow smoothly, there is a burst on the exhaust stroke of each cylinder. For every cycle of the engine there are 6 peaks and troughs in pressure (think about it - three in each manifold). If there are obstructions each burst gets reflected back, interrupting the flow. If the obstructions are smoothed, the bursts then do a better job pushing exhaust into and thru the manifold.
If you hunt around this site you'll find a thread (I think I was chasing a P0171) where a couple of the other guys mentioned to me I should port the manifold since I had to replace it (my front manifold had cracked, they all do eventually but in my case it took 7000+ hours of use). So I shrugged my shoulders, and did. I bought a junkyard manifold instead of an aftermarket replacement (actually I was hoping to stumble upon a ZZP p-log but I was delusional at the moment I guess) because the GM manifolds are made (much) better. They are SIMO iron castings. SIMO is a nodular cast iron typically containing 4–6 % silicon and 0.5–2 % molybdenum and can be used in temperatures up to 850–860 °C. The high silicon and silicon–molybdenum content improves the dimensional stability.
I was astonished that porting the front exhaust manifold actually made a difference -- at least, my butt-dyno says it definitely did. Then I read a post from azrockcrawler.com, from another enthusiast who's actually done some testing. The point is, this enthusiast builds/runs dune buggies. He likes the 3800 for various reasons. He points out that the rear manifold is fabricated (hint - its already a header of sorts) and performance gains from porting the stock manifold are (a) not that far off what you'd get by buying and installing a header (b) about 1/4 the cost (less if you do it my way ha ha):
- azrockcrawler.com wrote:
- Performance wise I have heard a set of headers in a dual exhaust configuration will easily gain you 30hp but also requires enough room to route the exhaust pipes on both sides of the buggy. Most of the guys I know running dual exhaust and custom headers with this motor have rear engine cars. For a front engine car I think the stock setup makes for very easy exhaust routing since everything dumps to one side. For performance gains with the stock exhaust routing you can get stock headers for roughly $700 or get nearly the same benefit with ported stock exhaust manifolds for almost a fourth the price.
He later points out that the ZZP "powerlog" stainless manifold isn't a bad deal.
There are lots of youtube videos about doing this. Many of them are, well, overblown; they go on and on about what you can do. What you have to do is smooth the openings so that the pressure waves do not reflect back into the heads. The ball hone to smooth out the insides is the only other thing I can think of that would make any diff - but unless you grind the flash so that the openings match up, smoothing the insides won't make any difference. Worse, the pipe itself is so narrow that its inherent restriction is "the bottleneck."
As for gains, the ZZP "powerlog" supposedly provides an increase of over 10 ft lbs of torque and about 15 HP in the 4500 RPM up range, but you have to retune the PCM to get the gains and control for things like knock retard. Some say you can get almost that much gain just porting and polishing the stock manifold.
Technical articles: https://burnsstainless.com/blogs/articles-1/exhaust-header-theory
http://www.epi-eng.com/piston_engine_technology/exhaust_system_technology.htm