Ok, well here goes in a little more detail, although im afraid i cant match Stu on either writing style or depth of knowledge (although by the sounds of that line i can compete with the best of them when it comes to a thorough rim job
)
WHAT IS VOLMENTRIC EFFICIENCY? (VE)
The volumetric efficiency of an engine is basically a % value representing how well it has managed to fill its cylinders in a cycle.
Now, picture an engine sat there on your bench at home.
You slowly turn it over, as the piston comes down and the valve opens it draws in air to fill the cylinder.
First thought is that it will draw in 500cc (in a 2 litre) of air exactly, as thats the amount of space there is available to fill.
However this wont be the case in reality.
The reason is that the valve is never open till EXACTLY the bottom of the cylinder, and if it either stay open after bottom dead centre or shuts before bottom dead centre it means that you have sealed in the amount of air at that point in the cycle instead.
Now, complicate things further by having this happen very quickly (the engine turning over i mean, instead of you doing it slowly by hand), this will drop the VE even further because the cylinder wont have time to fill properly.
So from that, you can see that as RPM increases, the VE will naturally drop in this simple example.
So why does Stu mention it being a curve, surely thats just a pretty much linear result???
Well, in that simple example yes, but in the real world things are far more dynamic than this simple example.
Im actually NOT going to go into a massive amount of detail on this as TBH it would end up longer than this topic is so far if we start getting down to thing like reversion at the valves and cam overlap etc.
But suffice to say that based on the specs of the cams (in terms of lift and duration) and the amount of overlap dialled into the cams, its going to effect the VE differently at different points in the rev range, port size in the head is another of these factors, and the reason why when you port a head you need a remap all the way up, its NOT just that a ported head means it breathes higher up the top, it can actually make it breath LESS well lower down, especially off boost, due to drops in gas speed.
HOW DOES THIS EFFECT THE MAPPING REQUIREMENTS OF THE ENGINE?
If any of you have had a very hot cam in a normally aspirated motor you will have noticeabley felt it "come on cam" at a point in the rev range, thats the point at which the VE suddently increases dramatically, and this will show up as an increase in torque, but in terms of the mapping (what this topic is about) it will require:
more fuel
less timing
So you will find that on many ignition maps you will get a line that starts to increase, then levels off, (or even goes back down in an extreme case!) and then starts to increase again as the RPM continues to rise still further.
The fuelling obviously will basically be a mirror image of this, with more fuel required at the points of higher VE where less timing is required. (well in our oversimplified example anyway, lol)
WHY ARE THE NUMBER OF LOAD SITES SO IMPORTANT?
This again points towards why its SO important to have as many map points as possible and why you need to have them positioned in the correct places to allow the interpolation between them to be as near to correct as possible.
Imagine this sitation, a peak in VE at 5500 rpm, and map points every 1000rpm.
What is going to happen is that you want timing of 12 degrees at 5000, 10 degrees at 5500, then back to 12 degrees at 6000. (for example, not actual figures from any specific engine!)
In order to avoid Det on that engine on 1000rpm increments you are going to have NO CHOICE but to dial in 10 degrees of timing at 5000 and 6000, costing you both economy and horsepower through that hole chunk of the rev range with the excepction of the 5500 point.
Obviously if you could move the load site to where this peak is, then even on the same number of sites, you can still manage to better map it.