Over the span of approximately 200 dyno runs I have done over the past 40-something years, regardless of whether an engine is designed for street use or all-out performance, centrifugal timing advance numbers are remarkably similar between engine manufacturers. Vacuum advance can mostly be ignored, as it only comes into play at light throttle/high vacuum situations, and mostly affects fuel economy, not horsepower or torque figures. Vacuum advance disappears almost instantly as soon as the engine gets a little torque load. To do a dyno pull properly, a specific RPM is selected, and the throttle opening is increased to try to maintain that RPM as load is applied to the engine. There will come a point where no more load can be applied at full throttle without dropping RPM. At that point, the torque reading is recorded and HP is calculated using the following formula:
TORQUE X RPM
5252
"Fudge factors" can be applied to the raw HP figures to account for temperature, altitude, barometric pressure, and relative humidity changes.
RPM is then increased (usually in increments of 500) and a new torque pull is made, until a predetermined max RPM is reached, or HP begins to decrease. Any alteration to operating conditions such as carburetor jetting, cam timing, intake/exhaust configuration, or ignition timing requires a new sequence of pulls. ONLY CHANGE ONE THING AT A TIME!
If we're evaluating centrifugal advance at a particular RPM, for instance, find the maximum torque developed at a particular WOT speed, and start advancing timing one degree at a time and recording the torque change. You'll find the "sweet spot" where torque is at a maximum. Advancing the timing further results in a loss of torque, due to the engine trying to run backwards as the flame front reaches the piston crown too soon. Record the torque loss from the over-advance. Go back to max torque, and start retarding the timing, recording the torque loss as the timing is retarded. You'll find that it takes approximately 5 degrees of timing retard from the max power point to equal the torque loss from ONE DEGREE of over-advance. It's always a good idea to err on the side of "not enough" centrifugal advance, as the law of diminishing returns kicks in very quickly!
The above results apply to virtually every engine I've ever dyno'ed, regardless of manufacturer, or the amount of modification it's had.
Jerry
