There's a bit of a black art in this, but as far as I understand it:
The bore of the pipes isn't a restriction, it just makes the gas going through it go faster.
Same idea as holding your finger over the end of a running hosepipe, restrict the flow and the output jet is quicker.
So a bigger bore reduces the speed at which the gas travels along the pipe, which has the effect of making the pipe seem longer.
The scavenging wave travelling back up the pipe assists the gas to leave the cylinder efficiently, and so a larger amount of fresh, new charge can go in, meaning that the engine is able to burn more fuel/air and make more power.
Obviously there's a limit as to when the pipe's diameter becomes too large, and the link pipes help fool the system into thinking it's a better length than it actually is.
Remember that there is only one perfect engine speed (revs) that works for each pipe design, and so the art of adding link pipes is to replicate the effect at different rev speeds, too.
With a simple 1:1 pipe for each cylinder, not linked in any way, the cylinders would each make a peak power at a particular engine speed. But this point is a quite narrow band, with power drop-off's either side.
Now by combining the pipes in a Siamese or by using link pipes, you get other peak occurring elsewhere in the rev range, so that peak power is produce over a wider range of engine speeds, not just one point.
You know the phrase "on the cam" or "power band"? Pipe diameter and length have a big part to play in this.
Any two stroke tuners will know that!
Unfortunately there are also other variables,, not just bore diameter and length.
The transition between one diameter pipe and the next is important; curvature and bend radius, whether the pipes are mandrel-bent (still round when bent) or bent using a pipe-bender (slightly flattened in cross-section) is a factor (remember serpent headers?) length of can, etc, etc.
So there's no simple way to say that a pipe must be a particular length and diameter, and anything else will result in less power.
In the same way the intake stacks are a compromise between the perfect length for peak power at a certain engine speed.
And all of the other engine design characteristics also play a part, compression ratio, number of valves, head design, squish, etc, etc, etc.
But it's still the case that when a flow of air or gas is made to turn a tight radius it slows the gas flow and increases resonance, noise, heat, and produces disturbances in the Force, probably.
If you look into the exhaust port you will clearly see one valve directly in line with the exhaust port, but the other valve is hidden to one side.
So the head port design which allows one exhaust valve to flow it's gas straight down the open mouth of the tube, but tortures the other valve's gas through two 90 degree bends in the space of 75mm is surely going to produce gas disturbance at the ports; some effects will surely be felt at the "straight" exhaust valve, because the way the port is machined as standard the gas flow from the "obscured" port is effectively introduced at 90 degrees from the open mouth of the exhaust pipe, and also at 90 degrees to the "straight" valve.
The gas will try to flow both ways, as it can't go straight on.
Once the "straight" exhaust valve closes the gas from the "obscured" valve can only go down the mouth of the exhaust header, but up to that point it can also try to flow back into the open exhaust valve of it's twin.
Now it's possible that Honda have created this very deliberately, and messing with opening up the port into an oval shape will be detrimental; but I don't think this will be the case.
Only one way to find out, I suppose....
And the fair test would be to open the port into an oval shape, make a short header to adapt to a standard set of pipes, and compare dyno results before and after.
I'd want to do that before playing with exhaust pipe diameter and lengths, so as to isolate as far as possible just the effect of the opening up of the port. One thing at a time.
I have a set of heads set aside for this project, and have planned out what metal needs to be removed form the heads, and how to seal up the coolant chamber and fit a replacement stud. I've also been thinking about whether a better method of clamping the pipe onto the head might be possible- springs? maybe... Some other sort of clamp? Perhaps.... Haven't finally decided on that one as yet.
I'll be able to establish the oval section's dimensions once that's completed, and will then be able to fabricate the header adaptor. Then all I need is an oval O-ring.....no ideas on that yet, though.
I might just have to use a flexible exhaust sealant, or heatproof flexible O-ring/gasket, rather than a compressible copper one.
Making my head hurt, so I'm off to bed to think about this some more while I'm asleep.
In hindsight this might be better as a thread in it's own right, more as a joint consultative/planning project taking all ideas into account. I'll get round to it one day, though....
Apologies for thread hijack
But I haven't yet worked out how to
deepest joy!!
