This is the spark table from a 1989 Mustang EEC:
http://www.mustangworks.com/cgi-bin/Pho ... s/fig6.gif
Instead of using a distributor with centrifugal advance plus a vacuum signal from the carb, spark timing is governed by engine load. Load is determined by how much air is entering the engine devided by the total amount of air the engine can injest. It can be thought of as volumetric efficency.
At low engine loads, you can see it takes a lot of spark to ignite the mixture. Closed and part throttle gives high manifold vacuum and low charge densities that burn slowly, so you have to start combustion early. As throttle setting and load increases, cylinder filling increases - charge density increases and your flame front or combustion speed increases, so you need less spark. Remember, the goal is to generate maximum cylinder pressure when the engine reaches maximum mechanical advantage. This is when the engine rotates a specific amount past TDC to where the connecting rod has reached a specific angle to the crankshaft. In most engines - depending on the rod ratio or length of the connecting rod with respect to the crankshaft stroke, this will be somewhere in the neighborhood of 20 degrees past TDC. So you need to vary the spark timing depending on the engine load to give max cyl pressure when you reach max mech advantage. The lighter the charge density (load), the more spark timing you need. The higher the charge density, the less spark you need.
With EFI, you can create a spark map similar to the one in the diagram. However, each engine will be different and it will take a certain amount of tuning to get your spark map to match your particular engine's requirements. With a distributor using vacuum/mechanical advance, this is not so easy. What you need to do is determine what baseline of initial spark you need and then dial-in how much more you can add with mechanical. At WOT where you have essentially zero intake manifold vacuum, your spark will be your initial + mechanical. How much mechanical advance you'll need and how fast it comes in is tuneable. With this as a baseline, you can now add vacuum advance. This works only under part throttle. Again, this is tuneable and needs to be determined for any particular engine.
Generally speaking, you will get better engine performance in terms of throttle response and part throttle power if you run more initial spark rather than less. Let's take one example: In some years, slant's used close to zero BTDC initial plus as much as 30 degrees of mechanical. When you accelerate at low speeds, you loose much of your vacuum advance and are left with your baseline initial plus mechanical. Because you cannot ramp the mechanical spark in too fast, under these conditions, you will have too little spark. If instead, you run a higher initial - something like 15 degrees - with less mechanical, you will have more spark at part throttle and, as a result, better performance. Most slants can take at least 15 degrees initial plus another 13 to 15 degrees mechanical.
The vacuum advance should be used to run as much spark as possible at light throttle (load) and tuned so it will remove enough spark under load to avoid detonation. The idea is to end up with something that will resemble the spark table where you loose spark as you go to the higher load areas. Again - this is much easier when you can use a lap top to change a spark map rather than changing spring weights and adjusting vacuum advance spring preloads. But the idea in both cases is the same. You want as much initial spark + mechanical with the modifier being the vacuum advance. The vacuum unit should add spark in under low load and take it away as load increases.
In the case of forced induction - either supercharging or turbocharging - you have a real challenge if you're using a distributor. Particularly in the case of turbocharging, you need to run a close to normal spark curve, but you need to pull timing in the midrange - and then add it back in at higher RPM. If you had a mass air meter on the car and you could look at the volumetric efficiency of the engine, you need to pull timing several hundred RPM before, during and after max volumetric efficiency to avoid detonation. Another way to look at this is to pull timing at the torque peak. Max torque output occurs at max volumetric efficiency. At higher RPM where volumetric efficiency drops off - even under forced induction, you can add spark back in. How you are going to do this with a distributor and carb is beyond me.
Mitch
|
Whoa. Thanks Mitch! Great post! 
