I thought I would take this opportunity to expose one of the least understood areas of the automotive engine. Why? Because I like a challenge and Catholicism is too easy a target. If you step back from an engine you can view it as a chemical reaction harnessed by a mechanical system to produce work. To produce peak power the mechanical requirement is for peak cylinder pressure to occur at the same crank angle under all operating conditions. Through testing that angle turns out to be between 15 and 20° after top dead centre (TDC). Any earlier than that and there’s either too much power lost to rising cylinder pressure before TDC, or a risk of knock. Any later than that the pressure front chases the piston down the bore rather than forcing it down.

The mechanical system, having a fixed geometry, works the same regardless of operating conditions. The same cannot be said for the chemical reaction it harnesses. Combustion speed of an air/fuel mixture is dependent upon numerous factors, some set during engine design time, a lot that vary during operation and some that vary even under steady state conditions. Indeed, just like snowflakes, no two combustion cycles are the same. Fortunately, the cycle-to-cycle variations can be safely ignored, as can some other factors, but most cannot. It is important to identify the physical controlling factors when changes are made to a stock engine so that an initial advance curve can be determined. The other factors, let’s call them environmental factors, have to be accommodated in operational controls or at least safeguards provided to ensure the engine does not operate outside safe ignition limits.

What this means is that to ensure the peak cylinder pressure occurs when it can do the most work, the flame front has to be initiated at the correct time. Unfortunately, this time varies, and because we are talking about a rotating machine, so does the angle. Because a cylinder charge takes a particular amount of time to burn and the engine speed varies, so then must the point at which combustion is initiated in advance. A typical ignition advance curve is shown in figure 1.

The major factors affecting ignition timing requirement are volumetric efficiency, engine speed and burn rate. Volumetric efficiency, or VE, is an expression of how much of a lung full a cylinder gets for each breath. As it turns out, a full cylinder burns faster than one that is only partially full so any improvements to VE also affect timing. Improvements to cylinder head ports, inlet manifold or camshaft improve VE at certain rev ranges so the ignition timing must also be changed to suit.

Lets look at a bunch of factors that affect timing requirements and discuss a few:
 
Factor Cylinder filling Flame speed Burn time Timing effect
bore/stroke ratio - You wouldn't have thought so, but this has an effect on rate of compression and piston dwell time at TDC - a long stroke/small bore requires as much as 10° more advance than does an equal displacement engine with short stroke/big bore
Camshaft with more duration improved at higher speed, worse at low speed - - less advance at high speed, more at low speed
combustion chamber shape - - compact designs take less time to burn to farthest reaches Minimized for compact designs, maximized for in-piston chambers
fuel atomization - liquid fuel does not burn fuel globules have to be atomized by combustion in other areas before they will burn Minimized when atomization is complete - atomization also improves with speed
improved exhaust efficiency or lowered backpressure - the presence of residual exhaust gas in the cylinder retards the flame front - less advance required when exhaust extraction effect is working, ie, higher up in the rev range
improved induction efficiency The engine has an easier time getting a 'lung' full faster with improved VE - less advance as VE improves
increased bore size may increase valve shrouding and lower VE - more distance from plug to far side of cylinder more advance required
increased compression ratio - a higher CR results in faster burning - less advance required for increased CR
mixture swirl in combustion chamber - - a well mixed cylinder charge will burn uniformly minimized with good swirl. Westlake type heads have heart shaped chambers to improve swirl
piston shape - - pistons which force the charge into a confined space limit the distance the flame front has to travel minimized for squish pistons, maximum for dished pistons
spark plug position in head - - varies with distance from plug to farthest point in the cylinder, ideally centered in cylinder minimum when centered, more advance as it moves to the farthest corner of the cylinder
air/fuel ratio - anything weaker or richer than the ideal 14.7:1 will burn slower - maximized at stoicheometric
coolant temperature - increased engine temperature affects final charge temperature - less advance required as engine temperature increases
fuel octane rating - octane slows burn rate - more advance with increased octane, or, more importantly, less risk of over advancing without changing advance
heat transfer rate - the higher rate at which the cylinder head can get rid of the combustion heat, the lower the final charge temperature will be - less advance needed for poor cooling iron heads, slightly more for efficient turbo heads, even more for aluminum 
induction air temperature Colder charge means denser air and higher VE, warmer charge, lower VE higher temperature charges burn faster - less timing with increased temperature
engine speed VE increases with speed - - as VE increases, advance decreases, but because flame speed is finite, it must be initiated much earlier with increasing speed
Ambient moisture conditions - moisture in the cylinder charge will slow the flame speed - more advance required under humid conditions

As you can see, there are a lot of factors that decide what a particular engine’s advance curve looks like. If you look at advance curves for a dozen different engines you will have a dozen different advance curves, but, they all have the same general pattern. This has a lot to do with the physics of an internal combustion engine. The fine tuning of exactly what the advance curve looks like for a particular engine is not so much designed in as pulled out of a running engine. The advance curve for a similar engine can safely be used as a starting point and in fact, the engine may run just fine. This is good news for the home tuner, however, until the optimum timing point is found there is horsepower going unharnessed.

If you are planning an engine modification, or have an already modified engine the above table will not help pinpoint where to push and pull the advance curve, or by how much. Some combination of factors cancel each other out, some amplify and some are so subtle they can be ignored. But how each of these factors affect an engine can only be determined by testing. When an engine has been changed in many areas, typically cylinder head, camshaft, inlet manifold and compression ratio all at once, any advance curve that will allow the engine to run should be considered nothing more than a starting point.

There are a few suppliers that offer ‘performance’ distributors, notably Aldon, Piper and Accel. Although they have sexy names like "Red", "Rally" and "Dominator" they are in fact no different from your original distributor except for the advance curve. As far as the engine is concerned it couldn’t care less if it were a distributor made by Lada or a gold plated one with a sexy name as long as the spark hits at the right instant. Considering the long list of factors affecting ignition timing, how is it possible for a 3rd party to supply an advance curve that optimizes an engine they’ve never seen? They can’t so to avoid pre-ignition problems they supply a generic conservative advance curve that may be just as conservative as your original. The actual difference between the ‘performance’ distributor and your own is 2 tiny springs. That’s it. And the chances of those springs delivering the optimum advance curve your engine needs is very low indeed. At the price of a new distributor, that’s an awful high price to pay for a shot in the dark.

Save your money. Don’t trust your ignition advance curve to anybody, but take your car to a rolling road dynamometer with an operator that knows what he’s doing. It will be cheaper in the end and you’ll maximize the return on all the mechanical modifications.